GaitRecognitionModel.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.7"
    },
    "colab": {
      "name": "DraftProcessDataTrainModel.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "qEW4nsrJcOzC"
      },
      "source": [
        "import pandas as pd\n",
        "import os\n",
        "import numpy as np\n",
        "import math\n",
        "from tqdm import tqdm\n",
        "from datetime import datetime\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.animation as animation\n",
        "from matplotlib import rc\n",
        "\n",
        "import tensorflow as tf\n",
        "import tensorflow.keras as keras\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense, Activation, Input, Bidirectional, Dropout, LSTM, TimeDistributed, Flatten, RepeatVector, BatchNormalization,Conv2D,GlobalAveragePooling2D,GlobalAveragePooling2D, LeakyReLU, AveragePooling2D,MaxPooling2D,SpatialDropout2D, Reshape\n",
        "from keras.regularizers import l2\n",
        "from keras.layers.convolutional_recurrent import ConvLSTM2D\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nRgjgknPcOzI"
      },
      "source": [
        "### Load and Process Data "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dc_oQnwmcOzI",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0a75ccdc-b2b5-427a-8926-129960019361"
      },
      "source": [
        "# Connect to your Google Drive here\n",
        "#\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "\n",
        "\n",
        "# Set the folder to point to where you downloaded the ESC dataset,\n",
        "# and also the folder to point to where you intend to save the processed data\n",
        "#\n",
        "user_folder = '/content/drive/My Drive/ColabNotebooks/ITI110'"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "scrolled": true,
        "id": "JlhTLFDxcOzJ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "ef1e09bb-1ddb-4881-c692-d44d07306765"
      },
      "source": [
        "#df=pd.read_csv('data_normalized.csv', index_col=False)\n",
        "#df=pd.read_csv('data_augmented.csv', index_col=False)\n",
        "\n",
        "df=pd.read_csv(user_folder +'/'+'data_augmented.csv', index_col=False)\n",
        "df.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "      <th>12</th>\n",
              "      <th>13</th>\n",
              "      <th>14</th>\n",
              "      <th>15</th>\n",
              "      <th>16</th>\n",
              "      <th>17</th>\n",
              "      <th>18</th>\n",
              "      <th>19</th>\n",
              "      <th>20</th>\n",
              "      <th>21</th>\n",
              "      <th>22</th>\n",
              "      <th>23</th>\n",
              "      <th>24</th>\n",
              "      <th>25</th>\n",
              "      <th>26</th>\n",
              "      <th>27</th>\n",
              "      <th>28</th>\n",
              "      <th>29</th>\n",
              "      <th>30</th>\n",
              "      <th>31</th>\n",
              "      <th>32</th>\n",
              "      <th>33</th>\n",
              "      <th>34</th>\n",
              "      <th>35</th>\n",
              "      <th>36</th>\n",
              "      <th>37</th>\n",
              "      <th>38</th>\n",
              "      <th>39</th>\n",
              "      <th>...</th>\n",
              "      <th>55</th>\n",
              "      <th>56</th>\n",
              "      <th>57</th>\n",
              "      <th>58</th>\n",
              "      <th>59</th>\n",
              "      <th>60</th>\n",
              "      <th>61</th>\n",
              "      <th>62</th>\n",
              "      <th>63</th>\n",
              "      <th>64</th>\n",
              "      <th>65</th>\n",
              "      <th>66</th>\n",
              "      <th>67</th>\n",
              "      <th>68</th>\n",
              "      <th>69</th>\n",
              "      <th>70</th>\n",
              "      <th>71</th>\n",
              "      <th>72</th>\n",
              "      <th>73</th>\n",
              "      <th>74</th>\n",
              "      <th>75</th>\n",
              "      <th>76</th>\n",
              "      <th>77</th>\n",
              "      <th>78</th>\n",
              "      <th>79</th>\n",
              "      <th>80</th>\n",
              "      <th>81</th>\n",
              "      <th>82</th>\n",
              "      <th>83</th>\n",
              "      <th>84</th>\n",
              "      <th>85</th>\n",
              "      <th>86</th>\n",
              "      <th>87</th>\n",
              "      <th>88</th>\n",
              "      <th>89</th>\n",
              "      <th>90</th>\n",
              "      <th>91</th>\n",
              "      <th>92</th>\n",
              "      <th>93</th>\n",
              "      <th>94</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.0002448784087915845, 0....</td>\n",
              "      <td>[0.17522129060731412, 0.2874729202431497, 0.0,...</td>\n",
              "      <td>[0.275779359207068, 0.26508140219964754, 0.0, ...</td>\n",
              "      <td>[0.2387925360778001, 0.3069714788637996, 0.0, ...</td>\n",
              "      <td>[0.20456744185837544, 0.30737914614479034, 0.0...</td>\n",
              "      <td>[0.24075421008753317, 0.35584099533804675, 0.0...</td>\n",
              "      <td>[0.25804888554524974, 0.31737329156146776, 0.0...</td>\n",
              "      <td>[0.17495590964359647, 0.2840891447979197, 0.0,...</td>\n",
              "      <td>[0.18076530371415564, 0.2767317541279658, 0.0,...</td>\n",
              "      <td>[0.20757447598040882, 0.28529764968514537, 0.0...</td>\n",
              "      <td>[0.2091182091145263, 0.30232406318830424, 0.0,...</td>\n",
              "      <td>[0.20010743464535025, 0.3298441852673377, 0.0,...</td>\n",
              "      <td>[0.21003737353317106, 0.3333837528591532, 0.0,...</td>\n",
              "      <td>[0.18854283561937127, 0.3064399035053147, 0.0,...</td>\n",
              "      <td>[0.1976422390846095, 0.34858295097922326, 0.0,...</td>\n",
              "      <td>[0.19934305760031792, 0.3278063230091262, 0.0,...</td>\n",
              "      <td>[0.19902218989551468, 0.3322032603021786, 0.0,...</td>\n",
              "      <td>[0.20018250715711758, 0.34316353455236803, 0.0...</td>\n",
              "      <td>[0.26038127720284104, 0.4232156687114082, 0.0,...</td>\n",
              "      <td>[0.27374810853647796, 0.3448762064162408, 0.0,...</td>\n",
              "      <td>[0.24733388351782823, 0.35316343893021707, 0.0...</td>\n",
              "      <td>[0.26387366023831255, 0.3669857277028243, 0.0,...</td>\n",
              "      <td>[0.2776930081004963, 0.33324635588744056, 0.0,...</td>\n",
              "      <td>[0.2492370334608279, 0.2941678728030825, 0.0, ...</td>\n",
              "      <td>[0.26911560583818733, 0.31380696604704733, 0.0...</td>\n",
              "      <td>[0.18519499920824312, 0.29302725548902425, 0.0...</td>\n",
              "      <td>[0.2354231687682618, 0.32608384131184087, 0.0,...</td>\n",
              "      <td>[0.2136604840336812, 0.3321463639378755, 0.0, ...</td>\n",
              "      <td>[0.2042403427475535, 0.3015386834501918, 0.0, ...</td>\n",
              "      <td>[0.22190930430560915, 0.2994399926689259, 0.0,...</td>\n",
              "      <td>[0.2464906595879739, 0.29182635343214014, 0.0,...</td>\n",
              "      <td>[0.2885980837410643, 0.2880883032159185, 0.0, ...</td>\n",
              "      <td>[50.926000000000045, 48.98800000000003, 0.0, 0...</td>\n",
              "      <td>[0.2627569603251295, 0.2632941308460845, 0.0, ...</td>\n",
              "      <td>[0.284142992599211, 0.24185011813109605, 0.0, ...</td>\n",
              "      <td>[0.24701302782674023, 0.32132327751772116, 0.0...</td>\n",
              "      <td>[0.24638835827805955, 0.28910689338446144, 0.0...</td>\n",
              "      <td>[0.30834423960710555, 0.28542341527901877, 0.0...</td>\n",
              "      <td>[0.26640699192961853, 0.28869674492562164, 0.0...</td>\n",
              "      <td>[0.26632188859231626, 0.27645178388849134, 0.0...</td>\n",
              "      <td>...</td>\n",
              "      <td>[0.27970034949713257, 0.23506245698554942, 0.0...</td>\n",
              "      <td>[0.28919639688219234, 0.2465330947788163, 0.0,...</td>\n",
              "      <td>[0.28088683887841037, 0.23807794115973815, 0.0...</td>\n",
              "      <td>[0.25846902264532406, 0.25819463939410875, 0.0...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.10381426449510446, -0.0...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.09852515127200821, -0.0...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.11071956599202543, -0.0...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.09070648604143422, -0.0...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.07684766337453774, -0.0...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.04806560399794436, -0.0...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.022558568811792093, -0....</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 3.8899999999998727, -13.6...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...</td>\n",
              "      <td>[15.664500000000004, 54.947, 0.0, 0.0, -35.286...</td>\n",
              "      <td>[0.1892188610206415, 0.3153112205074499, 0.0, ...</td>\n",
              "      <td>[0.17590752580270297, 0.35160739176797423, 0.0...</td>\n",
              "      <td>[0.1516448199485351, 0.25661458076219495, 0.0,...</td>\n",
              "      <td>[0.17482013733554239, 0.3840012489363702, 0.0,...</td>\n",
              "      <td>[0.241048098836105, 0.3673435076695727, 0.0, 0...</td>\n",
              "      <td>[0.1949698113084814, 0.3785348592735734, 0.0, ...</td>\n",
              "      <td>[0.17591500264396362, 0.3643027559189545, 0.0,...</td>\n",
              "      <td>[0.2325178599625046, 0.3884650580707508, 0.0, ...</td>\n",
              "      <td>[0.03974283792967209, 0.3942731202043322, 0.0,...</td>\n",
              "      <td>[0.17605214161459798, 0.42856259175208655, 0.0...</td>\n",
              "      <td>[0.14609593594836476, 0.3480999650970907, 0.0,...</td>\n",
              "      <td>[0.2604272907765966, -1.739629412911269, 0.0, ...</td>\n",
              "      <td>[0.21556494876003499, 0.3949762547868383, 0.0,...</td>\n",
              "      <td>[0.15470361076157546, 0.45098093843614756, 0.0...</td>\n",
              "      <td>[0.18478309619090205, 0.3565388709832277, 0.0,...</td>\n",
              "      <td>[0.1977850954722787, 0.31224246335434996, 0.0,...</td>\n",
              "      <td>[0.19580637951830854, 0.23294839469192605, 0.0...</td>\n",
              "      <td>[0.25280856329708556, 0.35797572175277925, 0.0...</td>\n",
              "      <td>[0.3138778618429066, 0.3814225768245948, 0.0, ...</td>\n",
              "      <td>[0.24348283055703027, 0.32879800062758646, 0.0...</td>\n",
              "      <td>[0.2897882281464847, 0.22959069813343508, 0.0,...</td>\n",
              "      <td>[0.2038011208318052, 0.26533365275204235, 0.0,...</td>\n",
              "      <td>[0.2260578543394793, 0.24648748552729208, 0.0,...</td>\n",
              "      <td>[0.24095772473713328, 0.24124877028047326, 0.0...</td>\n",
              "      <td>[0.37650409485519976, 0.4133358082843246, 0.0,...</td>\n",
              "      <td>[0.2022628298076117, 0.22313845606547023, 0.0,...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.22717958153780618, 0.04...</td>\n",
              "      <td>[0.23381321605193656, 0.7945594091891115, 0.0,...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.1625655376236216, -0.00...</td>\n",
              "      <td>[0.21155501390321949, 0.23423928751784545, 0.0...</td>\n",
              "      <td>[0.08038629463478304, 0.3101183760999813, 0.0,...</td>\n",
              "      <td>[0.12561980379126356, 0.343640498288127, 0.0, ...</td>\n",
              "      <td>[0.441591649235493, 0.3263840169462218, 0.0, 0...</td>\n",
              "      <td>[0.21029114535795915, 0.2314796974404615, 0.0,...</td>\n",
              "      <td>[0.14122261446022039, 0.20190766574139452, 0.0...</td>\n",
              "      <td>[0.24187136254278502, 0.2319001500218764, 0.0,...</td>\n",
              "      <td>[0.10997234095688892, 0.5358096834399555, 0.0,...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.016032401758037804, 0.0...</td>\n",
              "      <td>...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>[0.10257452267647307, 0.4246603163580588, 0.0,...</td>\n",
              "      <td>[0.18649193357184113, 0.4049084879972422, 0.0,...</td>\n",
              "      <td>[0.12301277212487861, 0.3843246239650838, 0.0,...</td>\n",
              "      <td>[0.23457380548929446, 0.37529098332682614, 0.0...</td>\n",
              "      <td>[0.2153002410793222, 0.41629479220004223, 0.0,...</td>\n",
              "      <td>[0.1597617665370772, 0.4953726490046505, 0.0, ...</td>\n",
              "      <td>[0.12137449338893704, 0.43918605708618, 0.0, 0...</td>\n",
              "      <td>[0.1713245910520363, 0.399893518878473, 0.0, 0...</td>\n",
              "      <td>[0.19694061833561738, 0.4239198807069571, 0.0,...</td>\n",
              "      <td>[0.17590876977087863, 0.36691298569346836, 0.0...</td>\n",
              "      <td>[0.17893179643061727, 0.3878390646076832, 0.0,...</td>\n",
              "      <td>[0.1492047026001898, 0.36490033223130336, 0.0,...</td>\n",
              "      <td>[0.16171666154573694, 0.36483957115870214, 0.0...</td>\n",
              "      <td>[0.1989015544724095, 0.4266222635547114, 0.0, ...</td>\n",
              "      <td>[0.13602030503354343, 0.4541743427494462, 0.0,...</td>\n",
              "      <td>[0.17141797660291272, 0.4052725427329992, 0.0,...</td>\n",
              "      <td>[0.1331245525268514, 0.38174563441068826, 0.0,...</td>\n",
              "      <td>[0.18172374644638603, 0.34884069171407106, 0.0...</td>\n",
              "      <td>[0.1396945256709658, 0.3895763559276461, 0.0, ...</td>\n",
              "      <td>[0.23087881820507222, 0.40021089650693176, 0.0...</td>\n",
              "      <td>[0.17959847883657384, 0.372378986458176, 0.0, ...</td>\n",
              "      <td>[0.1488217950327227, 0.32857323743637135, 0.0,...</td>\n",
              "      <td>[0.13683763403624008, 0.3943424477868977, 0.0,...</td>\n",
              "      <td>[0.19846621986007193, 0.3644267907440836, 0.0,...</td>\n",
              "      <td>[0.14723765699735483, 0.3676626575906125, 0.0,...</td>\n",
              "      <td>[0.013580557601254038, 0.32390600432339683, 0....</td>\n",
              "      <td>[0.12651426069351235, 0.3658919731011718, 0.0,...</td>\n",
              "      <td>[0.23927557722150214, 0.35129842237977377, 0.0...</td>\n",
              "      <td>[-0.04553326304445703, 0.37279024837503183, 0....</td>\n",
              "      <td>[0.13477275620191553, 0.3602788959199325, 0.0,...</td>\n",
              "      <td>[-0.01899632840559515, 0.433809804398888, 0.0,...</td>\n",
              "      <td>[-0.057014504628734704, 0.4422086380485906, 0....</td>\n",
              "      <td>[0.00014306021968095933, 0.45087017790377876, ...</td>\n",
              "      <td>[0.24108401648603492, 0.33161676585751404, 0.0...</td>\n",
              "      <td>[-0.08782426543419683, 0.3223911695038059, 0.0...</td>\n",
              "      <td>[0.16885229248463643, 0.35286050495862015, 0.0...</td>\n",
              "      <td>[-5.893000000000029, 39.228999999999985, 0.0, ...</td>\n",
              "      <td>[0.02969059130320155, 0.33243575694323996, 0.0...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>[0.13302203273083382, 0.4700551967384218, 0.0,...</td>\n",
              "      <td>[0.09421312141340309, 0.45309470504857746, 0.0...</td>\n",
              "      <td>[0.12267374247517351, 0.39958600151310014, 0.0...</td>\n",
              "      <td>[0.10978048721253882, 0.39088555163453004, 0.0...</td>\n",
              "      <td>[0.09430960990988571, 0.37528561218620254, 0.0...</td>\n",
              "      <td>[0.14218348575425158, 0.4118172561654683, 0.0,...</td>\n",
              "      <td>[0.18382754309201266, 0.38373236086954554, 0.0...</td>\n",
              "      <td>[0.22785302668489213, 0.37481173403621754, 0.0...</td>\n",
              "      <td>[0.20807959053145783, 0.38680371184141904, 0.0...</td>\n",
              "      <td>[0.1861201892271649, 0.40442920401173105, 0.0,...</td>\n",
              "      <td>[0.15514330354047048, 0.44789855018844915, 0.0...</td>\n",
              "      <td>[0.15369268547453363, 0.39875919558378164, 0.0...</td>\n",
              "      <td>[0.18348716814073732, 0.38272233075574663, 0.0...</td>\n",
              "      <td>[0.17092274280760783, 0.37176941086196325, 0.0...</td>\n",
              "      <td>[0.1712106294463659, 0.3875541468843235, 0.0, ...</td>\n",
              "      <td>[0.14948090921965937, 0.3868895288876866, 0.0,...</td>\n",
              "      <td>[0.1939554807281083, 0.41626356051917457, 0.0,...</td>\n",
              "      <td>[0.1844341743557257, 0.3829324501601202, 0.0, ...</td>\n",
              "      <td>[0.22328998869161787, 0.37216265526573505, 0.0...</td>\n",
              "      <td>[0.1563739312547487, 0.35645084283180994, 0.0,...</td>\n",
              "      <td>[0.18143655087958602, 0.3634908895468019, 0.0,...</td>\n",
              "      <td>[0.17193425563951686, 0.3741929824960656, 0.0,...</td>\n",
              "      <td>[0.15195659510047493, 0.3639180306698005, 0.0,...</td>\n",
              "      <td>[0.21196671759509972, 0.40700792353101034, 0.0...</td>\n",
              "      <td>[0.2088684413225541, 0.36990128345786366, 0.0,...</td>\n",
              "      <td>[0.2360673657286272, 0.43851656277794787, 0.0,...</td>\n",
              "      <td>[0.19157407414619423, 0.38389080999245256, 0.0...</td>\n",
              "      <td>[0.29062177474560613, 0.3721580853230531, 0.0,...</td>\n",
              "      <td>[0.1402741436122982, 0.3744360685612839, 0.0, ...</td>\n",
              "      <td>[0.18688114904403608, 0.4045769195784821, 0.0,...</td>\n",
              "      <td>[0.1487556813106367, 0.37238726068730815, 0.0,...</td>\n",
              "      <td>[0.2707207947562623, 0.376469422721029, 0.0, 0...</td>\n",
              "      <td>[0.19295997002517407, 0.41702582091229373, 0.0...</td>\n",
              "      <td>[0.015412508932703376, 0.45317872582737345, 0....</td>\n",
              "      <td>[0.123072740134851, 0.384841769324455, 0.0, 0....</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, -0.04521154072867402, 0.0...</td>\n",
              "      <td>[0.1446936230498427, 0.4337260829875747, 0.0, ...</td>\n",
              "      <td>[0.18628622229275194, 0.3550245993473127, 0.0,...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.08814276673984352, 0.00...</td>\n",
              "      <td>[0.08707210872078273, 0.34875441711634775, 0.0...</td>\n",
              "      <td>...</td>\n",
              "      <td>[13.659999999999968, 76.423, 0.0, 0.0, -46.983...</td>\n",
              "      <td>[11.720000000000027, 78.346, 0.0, 0.0, -48.976...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.12858079177789766, -0.0...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>[0.2091141550638824, 0.3157836824543189, 0.0, ...</td>\n",
              "      <td>[0.22492385765393264, 0.28776158834513116, 0.0...</td>\n",
              "      <td>[0.09923219851277483, 0.30944928944164135, 0.0...</td>\n",
              "      <td>[0.17314204277430578, 0.3446916476772184, 0.0,...</td>\n",
              "      <td>[0.15465182266592792, 0.3936274254847282, 0.0,...</td>\n",
              "      <td>[0.20239531520282436, 0.3758250494131653, 0.0,...</td>\n",
              "      <td>[0.1808036850321451, 0.37455491477679015, 0.0,...</td>\n",
              "      <td>[0.21391976983223573, 0.37716067989382956, 0.0...</td>\n",
              "      <td>[0.18913569116945017, 0.3403862566677584, 0.0,...</td>\n",
              "      <td>[0.15853547615392832, 0.38089064555420366, 0.0...</td>\n",
              "      <td>[0.18334017060533178, 0.3409657567684808, 0.0,...</td>\n",
              "      <td>[0.1348169938032137, 0.35006668812827757, 0.0,...</td>\n",
              "      <td>[0.1643155031538285, 0.39645473336103215, 0.0,...</td>\n",
              "      <td>[0.2186854195931858, 0.3698372808597583, 0.0, ...</td>\n",
              "      <td>[0.20503562550521384, 0.3331445032149949, 0.0,...</td>\n",
              "      <td>[0.196444053715724, 0.3210034355957933, 0.0, 0...</td>\n",
              "      <td>[0.18689291553662163, 0.3488069536434329, 0.0,...</td>\n",
              "      <td>[0.17954589766063786, 0.3585748811115384, 0.0,...</td>\n",
              "      <td>[0.2725211332887732, 0.367794164830887, 0.0, 0...</td>\n",
              "      <td>[0.19715555486412287, 0.3418794775717, 0.0, 0....</td>\n",
              "      <td>[0.20694738456837125, 0.298480659411993, 0.0, ...</td>\n",
              "      <td>[0.15050911510294432, 0.3544219962259993, 0.0,...</td>\n",
              "      <td>[0.18931282790330226, 0.3908789578820346, 0.0,...</td>\n",
              "      <td>[0.16658808970550715, 0.36067645797263664, 0.0...</td>\n",
              "      <td>[0.19917102048877638, 0.37168279536191023, 0.0...</td>\n",
              "      <td>[0.1446353477926202, 0.3670714443677828, 0.0, ...</td>\n",
              "      <td>[0.1667631566788031, 0.32061646036652625, 0.0,...</td>\n",
              "      <td>[0.1531921259759367, 0.3072791955047583, 0.0, ...</td>\n",
              "      <td>[0.18418706356156594, 0.3543504811057057, 0.0,...</td>\n",
              "      <td>[0.1751179420333937, 0.3501187759666117, 0.0, ...</td>\n",
              "      <td>[0.15304467097638258, 0.3466548875487812, 0.0,...</td>\n",
              "      <td>[0.15919458346725326, 0.3902107737481062, 0.0,...</td>\n",
              "      <td>[0.1708240345147019, 0.38101892896410483, 0.0,...</td>\n",
              "      <td>[0.1941131691877153, 0.35014856047864684, 0.0,...</td>\n",
              "      <td>[0.12055522654468008, 0.3333786361343153, 0.0,...</td>\n",
              "      <td>[0.19767111313765245, 0.3088508694744043, 0.0,...</td>\n",
              "      <td>[0.19251640639012776, 0.3077605388392634, 0.0,...</td>\n",
              "      <td>[0.20550742231589453, 0.3081957019206334, 0.0,...</td>\n",
              "      <td>[0.1921259856669105, 0.24431570607583286, 0.0,...</td>\n",
              "      <td>[0.26606364892339895, 0.34627816728247046, 0.0...</td>\n",
              "      <td>...</td>\n",
              "      <td>[0.2491028163674967, 0.2849674225056647, 0.0, ...</td>\n",
              "      <td>[0.2514638543278003, 0.29997724961949124, 0.0,...</td>\n",
              "      <td>[0.25042958674802634, 0.37532183442766315, 0.0...</td>\n",
              "      <td>[0.19007688632438316, 0.37868781765961707, 0.0...</td>\n",
              "      <td>[0.2151016469234024, 0.2771265585478549, 0.0, ...</td>\n",
              "      <td>[0.214843713545656, 0.2398881901126822, 0.0, 0...</td>\n",
              "      <td>[0.215190111136998, 0.21542260945635408, 0.0, ...</td>\n",
              "      <td>[0.26284373708064646, 0.26319965731736705, 0.0...</td>\n",
              "      <td>[0.2198747687067439, 0.2458263339555451, 0.0, ...</td>\n",
              "      <td>[0.21020910426964973, 0.2630329423377941, 0.0,...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.059533891254796724, -0....</td>\n",
              "      <td>[0.31328036281762495, 0.2615227384730479, 0.0,...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.07781215828748272, -0.0...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.11395100036777517, -0.0...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.1020373019039024, -0.05...</td>\n",
              "      <td>[0.3007275629345132, 0.2873485544792768, 0.0, ...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.10370976709900372, -0.0...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.08667594420039247, -0.0...</td>\n",
              "      <td>[0.303813623362051, 0.2637786218523216, 0.0, 0...</td>\n",
              "      <td>[0.0, 0.0, 0.0, 0.0, 0.016351077477203848, -0....</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "      <td>[0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 95 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                   0  ...                                                 94\n",
              "0  [0.0, 0.0, 0.0, 0.0, 0.0002448784087915845, 0....  ...  [0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...\n",
              "1  [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...  ...  [0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...\n",
              "2  [0.10257452267647307, 0.4246603163580588, 0.0,...  ...  [0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...\n",
              "3  [0.13302203273083382, 0.4700551967384218, 0.0,...  ...  [0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...\n",
              "4  [0.2091141550638824, 0.3157836824543189, 0.0, ...  ...  [0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, 0, 0.0, ...\n",
              "\n",
              "[5 rows x 95 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "scrolled": true,
        "id": "oDzP15N9cOzK",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3bcf95f2-e6ff-4f19-f538-d6e3d22127fb"
      },
      "source": [
        "df.columns"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12',\n",
              "       '13', '14', '15', '16', '17', '18', '19', '20', '21', '22', '23', '24',\n",
              "       '25', '26', '27', '28', '29', '30', '31', '32', '33', '34', '35', '36',\n",
              "       '37', '38', '39', '40', '41', '42', '43', '44', '45', '46', '47', '48',\n",
              "       '49', '50', '51', '52', '53', '54', '55', '56', '57', '58', '59', '60',\n",
              "       '61', '62', '63', '64', '65', '66', '67', '68', '69', '70', '71', '72',\n",
              "       '73', '74', '75', '76', '77', '78', '79', '80', '81', '82', '83', '84',\n",
              "       '85', '86', '87', '88', '89', '90', '91', '92', '93', '94'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2BYg9C27cOzK"
      },
      "source": [
        "max_frames_to_use=60\n",
        "\n",
        "df.drop(df.iloc[:,max_frames_to_use:], inplace=True, axis=1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oMEQmaNOcOzL",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e9eec425-ef27-48e4-e60f-d9f2dde9de9c"
      },
      "source": [
        "import ast\n",
        "\n",
        "#infilepath='data_resized.csv'\n",
        "\n",
        "\n",
        "def load_x(data):\n",
        "   \n",
        "    #print (\"Processing \" + infilepath)\n",
        "    x = []\n",
        "    \n",
        "    for row in tqdm(data.itertuples(index=False)):\n",
        "\n",
        "        # TODO:\n",
        "        # Load each row from the CSV file and call the \n",
        "        # process_joints function.\n",
        "        #...#\n",
        "        x_row=[]\n",
        "        for i in row:       \n",
        "            #print(ast.literal_eval(i))\n",
        "            x_row.append(ast.literal_eval(i))\n",
        "        \n",
        "        x.append(x_row)\n",
        "    x = np.array(x)\n",
        "    print(x.shape)\n",
        "   # x = np.reshape(x,(x.shape(0), 60, 36)) #each video has 32 frames\n",
        "    \n",
        "    print (\"Done.\")\n",
        "    return x\n",
        "\n",
        "x=load_x(df)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "792it [00:04, 168.01it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "(792, 60, 36)\n",
            "Done.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "scrolled": true,
        "id": "EwGrZW5ocOzL",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "77ff0287-86a2-449e-fb32-def474aed1f9"
      },
      "source": [
        "x.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(792, 60, 36)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pPTMcrQ4aL9_"
      },
      "source": [
        "# #to create row and columns dimensions for use in. convLstm2d model later.\n",
        "# import ast\n",
        "\n",
        "# #infilepath='data_resized.csv'\n",
        "\n",
        "\n",
        "# def load_x2(data):\n",
        "#     #d = []\n",
        "#     #print (\"Processing \" + infilepath)\n",
        "#     x = []\n",
        "#     #y = []\n",
        "    \n",
        "#     for row in tqdm(data.itertuples(index=False)):\n",
        "\n",
        "#         # TODO:\n",
        "#         # Load each row from the CSV file and call the \n",
        "#         # process_joints function.\n",
        "#         #...#\n",
        "#         x_row=[]\n",
        "#         #y_row=[]\n",
        "#         for i in row:       \n",
        "#             #print(ast.literal_eval(i))\n",
        "#             #x_row.append(ast.literal_eval(i))\n",
        "#             lst=ast.literal_eval(i)\n",
        "\n",
        "#             x_coordinates=lst[0::2]\n",
        "#             #x_row.append(x_coordinates)\n",
        "\n",
        "#             y_coordinates=lst[1::2]\n",
        "#             #y_row.append(y_coordinates)\n",
        "#             x_row.append([tuple(a) for a in zip(x_coordinates,y_coordinates)])\n",
        "#             #x_row.append(zip(x_coordinates,y_coordinates))\n",
        "        \n",
        "#         x.append(x_row)\n",
        "#         #y.append(y_row)\n",
        "#     #d.append(x)\n",
        "#     #d.append(y)\n",
        "#     x = np.array(x)\n",
        "#     # y = np.array(y)\n",
        "#     #d= np.array(d)\n",
        "\n",
        "#     print(x.shape)\n",
        "#    # x = np.reshape(x,(x.shape(0), 60, 36)) #each video has 32 frames\n",
        "    \n",
        "#     print (\"Done.\")\n",
        "#     return x\n",
        "\n",
        "# x=load_x2(df)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 607
        },
        "id": "uzyMk63Yra4R",
        "outputId": "b0dbc1f7-b644-4ed1-f61d-269376d85dcb"
      },
      "source": [
        "import PIL\n",
        "fig, ax = plt.subplots(figsize=(10,10))\n",
        "ax.imshow(x[0][0],cmap=plt.cm.jet, aspect='auto')\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7ff8b31d88d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 94
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YhbQHYvUcOzL"
      },
      "source": [
        "y=pd.read_csv('PersonLabel.csv',index_col=False)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eR-UsGDYcOzM",
        "outputId": "48f14bb4-9576-469f-c8ab-f01483ce9550"
      },
      "source": [
        "#personlabel='PersonLabel.csv'\n",
        "#personlabel='PersonLabel_augmented.csv'\n",
        "personlabel=user_folder+'/'+'PersonLabel_augmented.csv'\n",
        "import ast\n",
        "def load_y(personlabel):\n",
        "    df = pd.read_csv(personlabel)\n",
        "    \n",
        "    #print (\"Processing \" + personlabel)\n",
        "    \n",
        "    y = []\n",
        "    index = 0\n",
        "    for row in tqdm(df.itertuples(index=False)):\n",
        "        for i in row:\n",
        "          #print(ast.literal_eval(i))\n",
        "          y.append(ast.literal_eval(i))\n",
        "                    \n",
        "    y = np.array(y)  \n",
        "    return y\n",
        "\n",
        "y=load_y(personlabel)\n",
        "y.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "792it [00:00, 14395.24it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(792, 44)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fTg_aVz6cOzM",
        "outputId": "28edde75-e687-451d-f2c4-da4bddab25e3"
      },
      "source": [
        "# y=np.array(y)\n",
        "# y.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(792, 44)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KjcNM1ducOzM"
      },
      "source": [
        "y=pd.read_csv('Label.csv')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i9bbT5C3cOzM"
      },
      "source": [
        "from tensorflow.keras.utils import to_categorical \n",
        "\n",
        "y = to_categorical(y)\n",
        "\n",
        "\n",
        "#y=np.array(y)\n",
        "#y.dtype"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fwIrFD_2cOzN"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.2, random_state=42,stratify =y)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Nq03yUlYcOzN",
        "outputId": "162edd78-7fc1-4252-be26-5a6e32c2f24b"
      },
      "source": [
        "print(X_train.shape, y_train.shape, X_test.shape, y_test.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(633, 60, 36) (633, 44) (159, 60, 36) (159, 44)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "X80WO1CUf3ES",
        "outputId": "ca12806b-05ab-429c-deb7-bb273c5897ba"
      },
      "source": [
        "len(y_train[0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "44"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2S67PuAlKmZS"
      },
      "source": [
        "labels=np.unique(y, axis=0)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ikFRF2eNcOzN"
      },
      "source": [
        "### Create Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "f5xHuIkUcOzO",
        "outputId": "142614ab-a5a1-4099-f2af-5e531f657d74"
      },
      "source": [
        "#from tensorflow.keras.layers import PReLU\n",
        "from tensorflow.keras import optimizers, regularizers\n",
        "from keras.utils.vis_utils import plot_model\n",
        "\n",
        "# Create our LSTM model here\n",
        "#\n",
        "l2_rate=0.1   #0.1\n",
        "dropout_rate=0.3    #0.15\n",
        "dropout_rate1=0.2\n",
        "dropout_rate2=0.25\n",
        "\n",
        "def create_model():\n",
        "\n",
        "\n",
        "    model = Sequential()\n",
        "\n",
        "\n",
        "    model.add(Bidirectional(LSTM(512,return_sequences=True),input_shape=(95,36 ))) \n",
        "    model.add(BatchNormalization(renorm=True))\n",
        "\n",
        "    model.add(TimeDistributed(Dense(512,activation='elu')))\n",
        "\n",
        "\n",
        "    model.add(Bidirectional(LSTM(512,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='mul')) \n",
        "    model.add(BatchNormalization(renorm=True))\n",
        "\n",
        "    model.add(TimeDistributed(Dense(512,activation='elu'))) #TimeDistributedDense applies a same Dense (fully-connected) operation to every timestep of a 3D tensor.\n",
        "\n",
        "    model.add(Bidirectional(LSTM(512,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='mul')) \n",
        "    model.add(BatchNormalization(renorm=True))\n",
        "\n",
        "    model.add(TimeDistributed(Dense(512,activation='elu')))\n",
        "    \n",
        "    model.add(Bidirectional(LSTM(512,return_sequences=False,kernel_regularizer=l2(l2_rate)),merge_mode='mul'))\n",
        "    model.add(BatchNormalization(renorm=True))\n",
        "\n",
        "\n",
        "    model.add(Dropout(0.5))\n",
        "    \n",
        "    model.add(Dense(512, activation='elu',kernel_regularizer=regularizers.l1_l2(l1=0,l2=0.1)))\n",
        "                                          # bias_regularizer=regularizers.l2(0.01),\n",
        "                                          # activity_regularizer=regularizers.l2(0.01)))\n",
        "    #model.add(Dropout(0.5))\n",
        "    model.add(BatchNormalization())\n",
        "    model.add(Dropout(0.5))\n",
        "    #model.add(Dropout(0.3))\n",
        "    model.add(Dense(512, activation='elu',kernel_regularizer=regularizers.l1_l2(l1=0,l2=0.001)))\n",
        "                                          # bias_regularizer=regularizers.l2(0.01),\n",
        "                                          # activity_regularizer=regularizers.l2(0.01)))\n",
        "    #model.add(Dropout(0.5))\n",
        "    model.add(BatchNormalization())\n",
        "\n",
        "    model.add(Dense(44, activation='softmax'))\n",
        " \n",
        "    adm=Adam(lr=0.001, beta_1=0.9, beta_2=0.999, epsilon=None, decay=0.001, amsgrad=False) \n",
        "\n",
        "    model.compile(loss='categorical_crossentropy', optimizer=adm, metrics=['accuracy'])\n",
        "\n",
        "    model.summary(line_length=150)\n",
        "    plot_model(model, to_file='model_plot.png', show_shapes=True, show_layer_names=True)\n",
        "\n",
        "    \n",
        "    \n",
        "    return model\n",
        "\n",
        "    #model.add(TimeDistributed(Dense(256,activation='elu')))\n",
        "    \n",
        "    #model.add(TimeDistributed(Bidirectional(LSTM(500,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='mul')))\n",
        "    #model.add(BatchNormalization(renorm=True))\n",
        "    #model.add(Dropout(dropout_rate))\n",
        " \n",
        "\n",
        "    # model.add(RepeatVector(500))\n",
        "    # model.add(Bidirectional(LSTM(128,return_sequences=False,kernel_regularizer=l2(l2_rate)),merge_mode='mul'))\n",
        "    # model.add(BatchNormalization(renorm=True))\n",
        "    # model.add(RepeatVector(500))\n",
        "    #model.add(Dropout(dropout_rate))\n",
        "    # model.add(Bidirectional(LSTM(512,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='mul'))\n",
        "    # model.add(BatchNormalization(renorm=True))\n",
        "    #model.add(Dropout(dropout_rate))\n",
        "    #model.add(Dropout(dropout_rate2))\n",
        "\n",
        "    #model.add(TimeDistributed(Flatten()))\n",
        "    # model.add(Bidirectional(LSTM(512,return_sequences=False,kernel_regularizer=l2(l2_rate)),merge_mode='mul'))\n",
        "    # model.add(BatchNormalization(renorm=True))\n",
        "    # # #model.add(RepeatVector(5))\n",
        "    # model.add(Bidirectional(LSTM(64,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='mul'))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(Dropout(dropout_rate))\n",
        "    # model.add(Bidirectional(LSTM(64,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='ave'))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(Dropout(dropout_rate1))\n",
        "    # model.add(Bidirectional(LSTM(64,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='mul'))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(Dropout(dropout_rate2))\n",
        "\n",
        "    # model.add(Bidirectional(LSTM(64,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='mul'))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(Dropout(dropout_rate))\n",
        "    # model.add(Bidirectional(LSTM(64,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='ave'))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(Dropout(dropout_rate1))\n",
        "    # model.add(Bidirectional(LSTM(64,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='mul'))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(Dropout(dropout_rate2))\n",
        "\n",
        "\n",
        "    # model.add(Bidirectional(LSTM(64,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='mul'))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(Dropout(dropout_rate))\n",
        "    # model.add(Bidirectional(LSTM(64,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='ave'))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(Dropout(dropout_rate1))\n",
        "    # model.add(Bidirectional(LSTM(64,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='mul'))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(Dropout(dropout_rate2))\n",
        "\n",
        "    #model.add(RepeatVector(5))\n",
        "    # model.add(LSTM(512,kernel_regularizer=l2(l2_rate),return_sequences=True))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(LSTM(512,kernel_regularizer=l2(l2_rate)))\n",
        "    # model.add(BatchNormalization())\n",
        "\n",
        "    # model.add(RepeatVector(5))\n",
        "\n",
        "    #model.add(Dropout(0.1))\n",
        "    #model.add(Flatten())  \n",
        " \n",
        "    \n",
        "    # One or more repetitions of Dense layers\n",
        "    # model.add(Dense(64, activation='elu'))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(Dropout(0.8))\n",
        "    # model.add(Dense(128, activation='elu'))\n",
        "    # model.add(BatchNormalization())\n",
        "    \n",
        "\n",
        "    # model.add(Dropout(0.3))\n",
        "    \n",
        "    # model.add(Dense(512, activation='elu',kernel_regularizer=regularizers.l1_l2(l1=0,l2=0.2)))\n",
        "    #                                       # bias_regularizer=regularizers.l2(0.01),\n",
        "    #                                       # activity_regularizer=regularizers.l2(0.01)))\n",
        "    # #model.add(Dropout(0.5))\n",
        "    # model.add(BatchNormalization())\n",
        "    # model.add(Dropout(0.5))\n",
        "    # #model.add(Dropout(0.3))\n",
        "    # model.add(Dense(512, activation='elu',kernel_regularizer=regularizers.l1_l2(l1=0,l2=0.2)))\n",
        "    #                                       # bias_regularizer=regularizers.l2(0.01),\n",
        "    #                                       # activity_regularizer=regularizers.l2(0.01)))\n",
        "    # #model.add(Dropout(0.5))\n",
        "    # model.add(BatchNormalization())\n",
        "\n",
        "    # model.add(Dense(44, activation='softmax'))\n",
        " \n",
        "    # adm=Adam(lr=0.001, beta_1=0.9, beta_2=0.999, epsilon=None, decay=0.001, amsgrad=False) \n",
        "\n",
        "    # model.compile(loss='categorical_crossentropy', optimizer=adm, metrics=['accuracy'])\n",
        "\n",
        "    # model.summary(line_length=150)\n",
        "    \n",
        "    \n",
        "    \n",
        "    # return model\n",
        "\n",
        "\n",
        "\n",
        "# Create our model\n",
        "#\n",
        "model = create_model()\n",
        "\n",
        "\n",
        "# Create the training folder\n",
        "#\n",
        "# training_session_id = datetime.now().strftime(\"%Y-%m-%d %H-%M-%S\")\n",
        "# training_session_folder = 'train_%s' % (training_session_id)\n",
        "# os.makedirs(training_session_folder, exist_ok=True)\n",
        "\n",
        "# Configure the checkpoint and stop point.\n",
        "# This allows the training to save the best models and also stop the\n",
        "# training early if it detects that there are no improvements after\n",
        "# a long time.\n",
        "#\n",
        "# callbacks_list = [\n",
        "#     keras.callbacks.ModelCheckpoint(\n",
        "#         filepath=training_session_folder + '/model.{epoch:04d}-val_acc-{val_accuracy:4.2f}-loss-{val_loss:4.2f}.h5',\n",
        "#         monitor='val_loss', save_best_only=True),\n",
        "#     keras.callbacks.EarlyStopping(monitor='val_loss', patience=300, restore_best_weights=True)\n",
        "# ]\n",
        "\n",
        "callbacks_list = [\n",
        "    keras.callbacks.EarlyStopping(monitor='val_accuracy', patience=300, restore_best_weights=True)\n",
        "]\n",
        "\n",
        "\n",
        "# Train our model\n",
        "#\n",
        "history = model.fit(x=X_train, y=y_train, batch_size=100, epochs=5000, verbose=1, callbacks=callbacks_list, validation_data=(X_test, y_test))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_1\"\n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "Layer (type)                                                       Output Shape                                                Param #                \n",
            "======================================================================================================================================================\n",
            "bidirectional_4 (Bidirectional)                                    (None, 95, 1024)                                            2248704                \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "batch_normalization_6 (BatchNormalization)                         (None, 95, 1024)                                            7168                   \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "time_distributed_3 (TimeDistributed)                               (None, 95, 512)                                             524800                 \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "bidirectional_5 (Bidirectional)                                    (None, 95, 512)                                             4198400                \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "batch_normalization_7 (BatchNormalization)                         (None, 95, 512)                                             3584                   \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "time_distributed_4 (TimeDistributed)                               (None, 95, 512)                                             262656                 \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "bidirectional_6 (Bidirectional)                                    (None, 95, 512)                                             4198400                \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "batch_normalization_8 (BatchNormalization)                         (None, 95, 512)                                             3584                   \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "time_distributed_5 (TimeDistributed)                               (None, 95, 512)                                             262656                 \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "bidirectional_7 (Bidirectional)                                    (None, 512)                                                 4198400                \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "batch_normalization_9 (BatchNormalization)                         (None, 512)                                                 3584                   \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "dropout_2 (Dropout)                                                (None, 512)                                                 0                      \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "dense_9 (Dense)                                                    (None, 512)                                                 262656                 \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "batch_normalization_10 (BatchNormalization)                        (None, 512)                                                 2048                   \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "dropout_3 (Dropout)                                                (None, 512)                                                 0                      \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "dense_10 (Dense)                                                   (None, 512)                                                 262656                 \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "batch_normalization_11 (BatchNormalization)                        (None, 512)                                                 2048                   \n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "dense_11 (Dense)                                                   (None, 44)                                                  22572                  \n",
            "======================================================================================================================================================\n",
            "Total params: 16,463,916\n",
            "Trainable params: 16,449,068\n",
            "Non-trainable params: 14,848\n",
            "______________________________________________________________________________________________________________________________________________________\n",
            "Epoch 1/5000\n",
            "WARNING:tensorflow:Model was constructed with shape (None, 95, 36) for input KerasTensor(type_spec=TensorSpec(shape=(None, 95, 36), dtype=tf.float32, name='bidirectional_4_input'), name='bidirectional_4_input', description=\"created by layer 'bidirectional_4_input'\"), but it was called on an input with incompatible shape (None, 60, 36).\n",
            "WARNING:tensorflow:Model was constructed with shape (None, 95, 36) for input KerasTensor(type_spec=TensorSpec(shape=(None, 95, 36), dtype=tf.float32, name='bidirectional_4_input'), name='bidirectional_4_input', description=\"created by layer 'bidirectional_4_input'\"), but it was called on an input with incompatible shape (None, 60, 36).\n",
            "7/7 [==============================] - ETA: 0s - loss: 502.9563 - accuracy: 0.0112WARNING:tensorflow:Model was constructed with shape (None, 95, 36) for input KerasTensor(type_spec=TensorSpec(shape=(None, 95, 36), dtype=tf.float32, name='bidirectional_4_input'), name='bidirectional_4_input', description=\"created by layer 'bidirectional_4_input'\"), but it was called on an input with incompatible shape (None, 60, 36).\n",
            "7/7 [==============================] - 16s 482ms/step - loss: 498.5282 - accuracy: 0.0116 - val_loss: 353.0189 - val_accuracy: 0.0189\n",
            "Epoch 2/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 319.5977 - accuracy: 0.0222 - val_loss: 218.5940 - val_accuracy: 0.0252\n",
            "Epoch 3/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 196.3896 - accuracy: 0.0349 - val_loss: 130.4632 - val_accuracy: 0.0629\n",
            "Epoch 4/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 117.0085 - accuracy: 0.0312 - val_loss: 77.1349 - val_accuracy: 0.0252\n",
            "Epoch 5/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 69.7371 - accuracy: 0.0212 - val_loss: 45.8566 - val_accuracy: 0.0440\n",
            "Epoch 6/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 41.9841 - accuracy: 0.0316 - val_loss: 30.1394 - val_accuracy: 0.0189\n",
            "Epoch 7/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 28.4292 - accuracy: 0.0328 - val_loss: 22.8692 - val_accuracy: 0.0314\n",
            "Epoch 8/5000\n",
            "7/7 [==============================] - 1s 146ms/step - loss: 22.0744 - accuracy: 0.0183 - val_loss: 17.8046 - val_accuracy: 0.0252\n",
            "Epoch 9/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 17.4640 - accuracy: 0.0338 - val_loss: 14.4781 - val_accuracy: 0.0377\n",
            "Epoch 10/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 14.3095 - accuracy: 0.0267 - val_loss: 12.1375 - val_accuracy: 0.0189\n",
            "Epoch 11/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 12.1314 - accuracy: 0.0194 - val_loss: 10.5437 - val_accuracy: 0.0314\n",
            "Epoch 12/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 10.6400 - accuracy: 0.0295 - val_loss: 10.6651 - val_accuracy: 0.0314\n",
            "Epoch 13/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 9.6899 - accuracy: 0.0230 - val_loss: 9.4593 - val_accuracy: 0.0189\n",
            "Epoch 14/5000\n",
            "7/7 [==============================] - 1s 116ms/step - loss: 8.5460 - accuracy: 0.0303 - val_loss: 7.7934 - val_accuracy: 0.0252\n",
            "Epoch 15/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 7.7167 - accuracy: 0.0689 - val_loss: 7.2207 - val_accuracy: 0.0440\n",
            "Epoch 16/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 7.1953 - accuracy: 0.0594 - val_loss: 6.7418 - val_accuracy: 0.0377\n",
            "Epoch 17/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 7.0618 - accuracy: 0.0478 - val_loss: 11.6468 - val_accuracy: 0.0189\n",
            "Epoch 18/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 7.8913 - accuracy: 0.0235 - val_loss: 8.5717 - val_accuracy: 0.0252\n",
            "Epoch 19/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 7.3539 - accuracy: 0.0421 - val_loss: 6.5606 - val_accuracy: 0.0314\n",
            "Epoch 20/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 6.5525 - accuracy: 0.0664 - val_loss: 8.3467 - val_accuracy: 0.0189\n",
            "Epoch 21/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 7.1132 - accuracy: 0.0187 - val_loss: 7.9178 - val_accuracy: 0.0189\n",
            "Epoch 22/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 6.2372 - accuracy: 0.0364 - val_loss: 7.2323 - val_accuracy: 0.0126\n",
            "Epoch 23/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 5.9613 - accuracy: 0.0581 - val_loss: 8.8670 - val_accuracy: 0.0314\n",
            "Epoch 24/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 7.4131 - accuracy: 0.0337 - val_loss: 12.5866 - val_accuracy: 0.0440\n",
            "Epoch 25/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 12.9228 - accuracy: 0.0351 - val_loss: 19.6760 - val_accuracy: 0.0189\n",
            "Epoch 26/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 15.6948 - accuracy: 0.0482 - val_loss: 16.9585 - val_accuracy: 0.0126\n",
            "Epoch 27/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 15.2966 - accuracy: 0.0692 - val_loss: 17.7449 - val_accuracy: 0.0189\n",
            "Epoch 28/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 14.8711 - accuracy: 0.0206 - val_loss: 17.8706 - val_accuracy: 0.0252\n",
            "Epoch 29/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 13.3644 - accuracy: 0.0221 - val_loss: 15.9276 - val_accuracy: 0.0252\n",
            "Epoch 30/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 12.2165 - accuracy: 0.0181 - val_loss: 13.5300 - val_accuracy: 0.0252\n",
            "Epoch 31/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 11.1017 - accuracy: 0.0529 - val_loss: 11.1054 - val_accuracy: 0.0189\n",
            "Epoch 32/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 10.0768 - accuracy: 0.0492 - val_loss: 9.9065 - val_accuracy: 0.0377\n",
            "Epoch 33/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 9.5111 - accuracy: 0.0515 - val_loss: 9.0910 - val_accuracy: 0.0503\n",
            "Epoch 34/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 8.9505 - accuracy: 0.0671 - val_loss: 10.6099 - val_accuracy: 0.0189\n",
            "Epoch 35/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 8.5409 - accuracy: 0.0597 - val_loss: 8.4315 - val_accuracy: 0.0692\n",
            "Epoch 36/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 8.0344 - accuracy: 0.0909 - val_loss: 8.0610 - val_accuracy: 0.0943\n",
            "Epoch 37/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 7.6875 - accuracy: 0.0819 - val_loss: 7.4881 - val_accuracy: 0.0943\n",
            "Epoch 38/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 7.4442 - accuracy: 0.0805 - val_loss: 7.5854 - val_accuracy: 0.0692\n",
            "Epoch 39/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 7.1093 - accuracy: 0.1103 - val_loss: 6.9426 - val_accuracy: 0.1006\n",
            "Epoch 40/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 6.9930 - accuracy: 0.0990 - val_loss: 7.5060 - val_accuracy: 0.0692\n",
            "Epoch 41/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 6.7661 - accuracy: 0.1157 - val_loss: 7.0000 - val_accuracy: 0.1132\n",
            "Epoch 42/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 6.5801 - accuracy: 0.0939 - val_loss: 6.5511 - val_accuracy: 0.1132\n",
            "Epoch 43/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 6.3434 - accuracy: 0.1117 - val_loss: 6.3977 - val_accuracy: 0.1132\n",
            "Epoch 44/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 6.0968 - accuracy: 0.1517 - val_loss: 6.2374 - val_accuracy: 0.1572\n",
            "Epoch 45/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 5.9450 - accuracy: 0.1493 - val_loss: 6.0731 - val_accuracy: 0.1069\n",
            "Epoch 46/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 5.8762 - accuracy: 0.1542 - val_loss: 7.0388 - val_accuracy: 0.0440\n",
            "Epoch 47/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 5.7713 - accuracy: 0.1450 - val_loss: 8.8528 - val_accuracy: 0.0377\n",
            "Epoch 48/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 5.9531 - accuracy: 0.1211 - val_loss: 5.9772 - val_accuracy: 0.1635\n",
            "Epoch 49/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 5.5850 - accuracy: 0.1889 - val_loss: 5.8335 - val_accuracy: 0.1572\n",
            "Epoch 50/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 5.5545 - accuracy: 0.1454 - val_loss: 5.6351 - val_accuracy: 0.1635\n",
            "Epoch 51/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 5.6298 - accuracy: 0.1457 - val_loss: 17.9254 - val_accuracy: 0.0377\n",
            "Epoch 52/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 8.9702 - accuracy: 0.0331 - val_loss: 15.5866 - val_accuracy: 0.0189\n",
            "Epoch 53/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 10.9219 - accuracy: 0.0563 - val_loss: 13.5980 - val_accuracy: 0.0189\n",
            "Epoch 54/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 12.2785 - accuracy: 0.0612 - val_loss: 15.8320 - val_accuracy: 0.0252\n",
            "Epoch 55/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 11.8042 - accuracy: 0.0725 - val_loss: 11.9039 - val_accuracy: 0.0692\n",
            "Epoch 56/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 10.4178 - accuracy: 0.0743 - val_loss: 9.8173 - val_accuracy: 0.0566\n",
            "Epoch 57/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 9.1161 - accuracy: 0.0930 - val_loss: 11.2438 - val_accuracy: 0.0252\n",
            "Epoch 58/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 8.2182 - accuracy: 0.0910 - val_loss: 9.5675 - val_accuracy: 0.0818\n",
            "Epoch 59/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 7.4769 - accuracy: 0.0943 - val_loss: 8.0478 - val_accuracy: 0.0566\n",
            "Epoch 60/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 7.1068 - accuracy: 0.0918 - val_loss: 7.3745 - val_accuracy: 0.0503\n",
            "Epoch 61/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 6.7050 - accuracy: 0.0847 - val_loss: 6.6246 - val_accuracy: 0.0755\n",
            "Epoch 62/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 6.3296 - accuracy: 0.1158 - val_loss: 6.2230 - val_accuracy: 0.0755\n",
            "Epoch 63/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 6.0263 - accuracy: 0.0825 - val_loss: 6.0315 - val_accuracy: 0.1006\n",
            "Epoch 64/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 5.7495 - accuracy: 0.1218 - val_loss: 6.1904 - val_accuracy: 0.0566\n",
            "Epoch 65/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 5.6037 - accuracy: 0.1046 - val_loss: 5.5580 - val_accuracy: 0.1069\n",
            "Epoch 66/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 5.3366 - accuracy: 0.1163 - val_loss: 5.8511 - val_accuracy: 0.0943\n",
            "Epoch 67/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 5.1683 - accuracy: 0.1944 - val_loss: 5.3546 - val_accuracy: 0.1132\n",
            "Epoch 68/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 5.0079 - accuracy: 0.1538 - val_loss: 5.3813 - val_accuracy: 0.1572\n",
            "Epoch 69/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 4.9194 - accuracy: 0.1713 - val_loss: 5.2703 - val_accuracy: 0.1195\n",
            "Epoch 70/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 4.8239 - accuracy: 0.2001 - val_loss: 5.6292 - val_accuracy: 0.0566\n",
            "Epoch 71/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 4.7096 - accuracy: 0.2057 - val_loss: 5.1203 - val_accuracy: 0.1132\n",
            "Epoch 72/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 4.5958 - accuracy: 0.2045 - val_loss: 5.7347 - val_accuracy: 0.1069\n",
            "Epoch 73/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 4.5582 - accuracy: 0.2044 - val_loss: 5.3736 - val_accuracy: 0.0943\n",
            "Epoch 74/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 4.4794 - accuracy: 0.1820 - val_loss: 5.5057 - val_accuracy: 0.0252\n",
            "Epoch 75/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 4.5595 - accuracy: 0.1463 - val_loss: 5.4261 - val_accuracy: 0.0881\n",
            "Epoch 76/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 4.5332 - accuracy: 0.1896 - val_loss: 5.2262 - val_accuracy: 0.0629\n",
            "Epoch 77/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 4.4357 - accuracy: 0.2457 - val_loss: 5.1432 - val_accuracy: 0.1321\n",
            "Epoch 78/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 4.3341 - accuracy: 0.2358 - val_loss: 4.8402 - val_accuracy: 0.1824\n",
            "Epoch 79/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 4.1975 - accuracy: 0.2500 - val_loss: 4.4218 - val_accuracy: 0.1887\n",
            "Epoch 80/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 4.2106 - accuracy: 0.1969 - val_loss: 4.7896 - val_accuracy: 0.1824\n",
            "Epoch 81/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 4.1527 - accuracy: 0.2454 - val_loss: 5.0410 - val_accuracy: 0.1321\n",
            "Epoch 82/5000\n",
            "7/7 [==============================] - 1s 149ms/step - loss: 4.3243 - accuracy: 0.2235 - val_loss: 4.5130 - val_accuracy: 0.2704\n",
            "Epoch 83/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 4.2313 - accuracy: 0.2480 - val_loss: 4.9196 - val_accuracy: 0.1258\n",
            "Epoch 84/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 4.2412 - accuracy: 0.2293 - val_loss: 4.8128 - val_accuracy: 0.1572\n",
            "Epoch 85/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 4.0944 - accuracy: 0.2584 - val_loss: 4.7700 - val_accuracy: 0.1509\n",
            "Epoch 86/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 4.1210 - accuracy: 0.2401 - val_loss: 4.9997 - val_accuracy: 0.1321\n",
            "Epoch 87/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 3.9651 - accuracy: 0.2843 - val_loss: 4.0544 - val_accuracy: 0.2013\n",
            "Epoch 88/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 3.7985 - accuracy: 0.2841 - val_loss: 4.1654 - val_accuracy: 0.2579\n",
            "Epoch 89/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.8000 - accuracy: 0.2610 - val_loss: 4.0851 - val_accuracy: 0.1950\n",
            "Epoch 90/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.6860 - accuracy: 0.2978 - val_loss: 5.3038 - val_accuracy: 0.1132\n",
            "Epoch 91/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.9518 - accuracy: 0.2095 - val_loss: 5.6416 - val_accuracy: 0.0440\n",
            "Epoch 92/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.9381 - accuracy: 0.2638 - val_loss: 4.9186 - val_accuracy: 0.1447\n",
            "Epoch 93/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 3.9535 - accuracy: 0.3363 - val_loss: 4.7124 - val_accuracy: 0.1824\n",
            "Epoch 94/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 3.9438 - accuracy: 0.3403 - val_loss: 4.5136 - val_accuracy: 0.2264\n",
            "Epoch 95/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 3.8113 - accuracy: 0.3260 - val_loss: 4.3545 - val_accuracy: 0.2013\n",
            "Epoch 96/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.6307 - accuracy: 0.3703 - val_loss: 4.1170 - val_accuracy: 0.2516\n",
            "Epoch 97/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.5239 - accuracy: 0.3637 - val_loss: 4.2457 - val_accuracy: 0.1635\n",
            "Epoch 98/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 3.4248 - accuracy: 0.3768 - val_loss: 4.4449 - val_accuracy: 0.1509\n",
            "Epoch 99/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.4067 - accuracy: 0.3526 - val_loss: 4.0269 - val_accuracy: 0.2453\n",
            "Epoch 100/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 3.2746 - accuracy: 0.3872 - val_loss: 4.5953 - val_accuracy: 0.1509\n",
            "Epoch 101/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.3818 - accuracy: 0.3612 - val_loss: 4.0341 - val_accuracy: 0.2390\n",
            "Epoch 102/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.3437 - accuracy: 0.4202 - val_loss: 4.1566 - val_accuracy: 0.2264\n",
            "Epoch 103/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 3.3388 - accuracy: 0.4651 - val_loss: 4.4585 - val_accuracy: 0.1447\n",
            "Epoch 104/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.4161 - accuracy: 0.3761 - val_loss: 4.9048 - val_accuracy: 0.1069\n",
            "Epoch 105/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 3.3030 - accuracy: 0.4113 - val_loss: 4.7034 - val_accuracy: 0.2013\n",
            "Epoch 106/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.3560 - accuracy: 0.3568 - val_loss: 5.2828 - val_accuracy: 0.1258\n",
            "Epoch 107/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 3.2838 - accuracy: 0.3916 - val_loss: 3.8885 - val_accuracy: 0.2390\n",
            "Epoch 108/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.1392 - accuracy: 0.4313 - val_loss: 4.3936 - val_accuracy: 0.1195\n",
            "Epoch 109/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 3.0753 - accuracy: 0.3875 - val_loss: 6.3391 - val_accuracy: 0.0881\n",
            "Epoch 110/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.9315 - accuracy: 0.4949 - val_loss: 3.6478 - val_accuracy: 0.2893\n",
            "Epoch 111/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 2.9317 - accuracy: 0.4486 - val_loss: 5.1907 - val_accuracy: 0.1321\n",
            "Epoch 112/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.1509 - accuracy: 0.3848 - val_loss: 3.7712 - val_accuracy: 0.3270\n",
            "Epoch 113/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.9055 - accuracy: 0.5183 - val_loss: 4.5630 - val_accuracy: 0.1195\n",
            "Epoch 114/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.9696 - accuracy: 0.4549 - val_loss: 4.2761 - val_accuracy: 0.2704\n",
            "Epoch 115/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 2.8790 - accuracy: 0.4698 - val_loss: 3.8325 - val_accuracy: 0.2830\n",
            "Epoch 116/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 2.8043 - accuracy: 0.5136 - val_loss: 3.9689 - val_accuracy: 0.1761\n",
            "Epoch 117/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 2.7590 - accuracy: 0.5027 - val_loss: 3.5379 - val_accuracy: 0.3459\n",
            "Epoch 118/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 2.5791 - accuracy: 0.5746 - val_loss: 4.4154 - val_accuracy: 0.2201\n",
            "Epoch 119/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.7499 - accuracy: 0.5401 - val_loss: 3.9465 - val_accuracy: 0.3082\n",
            "Epoch 120/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 2.8810 - accuracy: 0.5039 - val_loss: 4.0951 - val_accuracy: 0.2767\n",
            "Epoch 121/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 3.0954 - accuracy: 0.4848 - val_loss: 4.5157 - val_accuracy: 0.1761\n",
            "Epoch 122/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 3.0651 - accuracy: 0.4745 - val_loss: 3.6312 - val_accuracy: 0.3396\n",
            "Epoch 123/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 2.9253 - accuracy: 0.4912 - val_loss: 4.1405 - val_accuracy: 0.2201\n",
            "Epoch 124/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 2.8046 - accuracy: 0.5047 - val_loss: 3.6851 - val_accuracy: 0.3208\n",
            "Epoch 125/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 2.6501 - accuracy: 0.5715 - val_loss: 3.5091 - val_accuracy: 0.3396\n",
            "Epoch 126/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 2.6097 - accuracy: 0.5543 - val_loss: 3.3253 - val_accuracy: 0.4528\n",
            "Epoch 127/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 2.4865 - accuracy: 0.5819 - val_loss: 3.2190 - val_accuracy: 0.3648\n",
            "Epoch 128/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 2.5348 - accuracy: 0.5542 - val_loss: 3.4436 - val_accuracy: 0.3711\n",
            "Epoch 129/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.5509 - accuracy: 0.5064 - val_loss: 4.5796 - val_accuracy: 0.1572\n",
            "Epoch 130/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 2.6822 - accuracy: 0.4916 - val_loss: 4.1136 - val_accuracy: 0.2013\n",
            "Epoch 131/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 2.6145 - accuracy: 0.5082 - val_loss: 6.8731 - val_accuracy: 0.1195\n",
            "Epoch 132/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 2.4129 - accuracy: 0.5894 - val_loss: 3.4093 - val_accuracy: 0.4025\n",
            "Epoch 133/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 2.3345 - accuracy: 0.6433 - val_loss: 3.7546 - val_accuracy: 0.2830\n",
            "Epoch 134/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.3452 - accuracy: 0.6042 - val_loss: 3.3522 - val_accuracy: 0.3459\n",
            "Epoch 135/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.2577 - accuracy: 0.6214 - val_loss: 3.2020 - val_accuracy: 0.4277\n",
            "Epoch 136/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.2515 - accuracy: 0.6367 - val_loss: 3.1916 - val_accuracy: 0.3962\n",
            "Epoch 137/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 2.2209 - accuracy: 0.6267 - val_loss: 4.4210 - val_accuracy: 0.2830\n",
            "Epoch 138/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.1489 - accuracy: 0.6716 - val_loss: 4.2388 - val_accuracy: 0.2516\n",
            "Epoch 139/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.3250 - accuracy: 0.5748 - val_loss: 4.0059 - val_accuracy: 0.2264\n",
            "Epoch 140/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.2411 - accuracy: 0.5750 - val_loss: 4.2778 - val_accuracy: 0.1509\n",
            "Epoch 141/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.0710 - accuracy: 0.6657 - val_loss: 3.0027 - val_accuracy: 0.4591\n",
            "Epoch 142/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.9912 - accuracy: 0.6977 - val_loss: 3.0663 - val_accuracy: 0.4780\n",
            "Epoch 143/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 2.0388 - accuracy: 0.6454 - val_loss: 3.2849 - val_accuracy: 0.4025\n",
            "Epoch 144/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.8600 - accuracy: 0.7221 - val_loss: 3.2786 - val_accuracy: 0.4214\n",
            "Epoch 145/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.9886 - accuracy: 0.6610 - val_loss: 4.7119 - val_accuracy: 0.2138\n",
            "Epoch 146/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 2.0858 - accuracy: 0.6354 - val_loss: 4.4214 - val_accuracy: 0.2893\n",
            "Epoch 147/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.2605 - accuracy: 0.5699 - val_loss: 4.5926 - val_accuracy: 0.2327\n",
            "Epoch 148/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.3173 - accuracy: 0.6199 - val_loss: 3.3595 - val_accuracy: 0.4654\n",
            "Epoch 149/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.1651 - accuracy: 0.7138 - val_loss: 4.1588 - val_accuracy: 0.2516\n",
            "Epoch 150/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.1247 - accuracy: 0.6901 - val_loss: 3.1505 - val_accuracy: 0.4843\n",
            "Epoch 151/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 2.0452 - accuracy: 0.6870 - val_loss: 3.6582 - val_accuracy: 0.3522\n",
            "Epoch 152/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 2.0235 - accuracy: 0.6699 - val_loss: 3.2050 - val_accuracy: 0.4843\n",
            "Epoch 153/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.8491 - accuracy: 0.7477 - val_loss: 2.8381 - val_accuracy: 0.5157\n",
            "Epoch 154/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.8363 - accuracy: 0.7285 - val_loss: 3.1666 - val_accuracy: 0.4969\n",
            "Epoch 155/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.7466 - accuracy: 0.7483 - val_loss: 3.5784 - val_accuracy: 0.3396\n",
            "Epoch 156/5000\n",
            "7/7 [==============================] - 1s 145ms/step - loss: 2.0124 - accuracy: 0.6154 - val_loss: 3.2956 - val_accuracy: 0.3836\n",
            "Epoch 157/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.8453 - accuracy: 0.6802 - val_loss: 3.6201 - val_accuracy: 0.3082\n",
            "Epoch 158/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.8923 - accuracy: 0.6938 - val_loss: 3.5090 - val_accuracy: 0.3648\n",
            "Epoch 159/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.8647 - accuracy: 0.6792 - val_loss: 2.9193 - val_accuracy: 0.5409\n",
            "Epoch 160/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.6818 - accuracy: 0.7533 - val_loss: 3.0980 - val_accuracy: 0.4151\n",
            "Epoch 161/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.6625 - accuracy: 0.7892 - val_loss: 2.8535 - val_accuracy: 0.5031\n",
            "Epoch 162/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.6065 - accuracy: 0.7908 - val_loss: 2.8793 - val_accuracy: 0.5409\n",
            "Epoch 163/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.5662 - accuracy: 0.7986 - val_loss: 2.7023 - val_accuracy: 0.5723\n",
            "Epoch 164/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.5524 - accuracy: 0.7756 - val_loss: 2.7200 - val_accuracy: 0.5409\n",
            "Epoch 165/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.5067 - accuracy: 0.8037 - val_loss: 3.4742 - val_accuracy: 0.3585\n",
            "Epoch 166/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.4984 - accuracy: 0.7942 - val_loss: 2.6844 - val_accuracy: 0.6038\n",
            "Epoch 167/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.4648 - accuracy: 0.8003 - val_loss: 2.8648 - val_accuracy: 0.4906\n",
            "Epoch 168/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 1.6722 - accuracy: 0.6872 - val_loss: 3.1666 - val_accuracy: 0.4088\n",
            "Epoch 169/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 1.6858 - accuracy: 0.7316 - val_loss: 3.6644 - val_accuracy: 0.4403\n",
            "Epoch 170/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.6494 - accuracy: 0.7788 - val_loss: 4.2553 - val_accuracy: 0.2642\n",
            "Epoch 171/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.4952 - accuracy: 0.8318 - val_loss: 3.1899 - val_accuracy: 0.3962\n",
            "Epoch 172/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.5875 - accuracy: 0.7739 - val_loss: 2.7327 - val_accuracy: 0.5597\n",
            "Epoch 173/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.6039 - accuracy: 0.7427 - val_loss: 3.2502 - val_accuracy: 0.3899\n",
            "Epoch 174/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.5670 - accuracy: 0.7859 - val_loss: 3.0453 - val_accuracy: 0.4969\n",
            "Epoch 175/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 1.4286 - accuracy: 0.8228 - val_loss: 2.7082 - val_accuracy: 0.5597\n",
            "Epoch 176/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.4588 - accuracy: 0.8137 - val_loss: 3.1144 - val_accuracy: 0.5346\n",
            "Epoch 177/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.5423 - accuracy: 0.7602 - val_loss: 5.3242 - val_accuracy: 0.1572\n",
            "Epoch 178/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.5031 - accuracy: 0.7804 - val_loss: 4.1574 - val_accuracy: 0.2579\n",
            "Epoch 179/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 1.5036 - accuracy: 0.7959 - val_loss: 6.4376 - val_accuracy: 0.1069\n",
            "Epoch 180/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.4932 - accuracy: 0.7937 - val_loss: 4.1331 - val_accuracy: 0.2579\n",
            "Epoch 181/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.5079 - accuracy: 0.8009 - val_loss: 3.1325 - val_accuracy: 0.4843\n",
            "Epoch 182/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 3.1108 - accuracy: 0.5133 - val_loss: 33.9964 - val_accuracy: 0.0252\n",
            "Epoch 183/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 8.9747 - accuracy: 0.1047 - val_loss: 24.6084 - val_accuracy: 0.0943\n",
            "Epoch 184/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 12.5542 - accuracy: 0.1030 - val_loss: 19.1786 - val_accuracy: 0.0692\n",
            "Epoch 185/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 13.8047 - accuracy: 0.0952 - val_loss: 20.9223 - val_accuracy: 0.0818\n",
            "Epoch 186/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 12.3088 - accuracy: 0.1161 - val_loss: 14.2072 - val_accuracy: 0.0755\n",
            "Epoch 187/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 11.0349 - accuracy: 0.1034 - val_loss: 22.1629 - val_accuracy: 0.0314\n",
            "Epoch 188/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 9.2228 - accuracy: 0.1344 - val_loss: 10.0779 - val_accuracy: 0.1069\n",
            "Epoch 189/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 7.8597 - accuracy: 0.1866 - val_loss: 8.2349 - val_accuracy: 0.1321\n",
            "Epoch 190/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 6.7409 - accuracy: 0.2230 - val_loss: 7.5858 - val_accuracy: 0.1384\n",
            "Epoch 191/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 5.9561 - accuracy: 0.2286 - val_loss: 5.7927 - val_accuracy: 0.2201\n",
            "Epoch 192/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 5.7285 - accuracy: 0.2663 - val_loss: 9.4686 - val_accuracy: 0.0629\n",
            "Epoch 193/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 6.7382 - accuracy: 0.0933 - val_loss: 10.6732 - val_accuracy: 0.0692\n",
            "Epoch 194/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 5.8584 - accuracy: 0.1163 - val_loss: 11.7956 - val_accuracy: 0.0252\n",
            "Epoch 195/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 5.9106 - accuracy: 0.1098 - val_loss: 9.4301 - val_accuracy: 0.1006\n",
            "Epoch 196/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 5.3206 - accuracy: 0.2222 - val_loss: 5.8037 - val_accuracy: 0.1572\n",
            "Epoch 197/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 5.0760 - accuracy: 0.2618 - val_loss: 5.7729 - val_accuracy: 0.1384\n",
            "Epoch 198/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 4.6218 - accuracy: 0.3282 - val_loss: 5.0497 - val_accuracy: 0.2453\n",
            "Epoch 199/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 4.1701 - accuracy: 0.3655 - val_loss: 4.6582 - val_accuracy: 0.2579\n",
            "Epoch 200/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.8824 - accuracy: 0.3712 - val_loss: 4.6451 - val_accuracy: 0.1635\n",
            "Epoch 201/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 3.6757 - accuracy: 0.3656 - val_loss: 4.1791 - val_accuracy: 0.2893\n",
            "Epoch 202/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 3.3156 - accuracy: 0.4800 - val_loss: 4.1803 - val_accuracy: 0.2830\n",
            "Epoch 203/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 3.1136 - accuracy: 0.5063 - val_loss: 4.2022 - val_accuracy: 0.1572\n",
            "Epoch 204/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 3.0514 - accuracy: 0.4751 - val_loss: 4.5478 - val_accuracy: 0.1635\n",
            "Epoch 205/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 3.0567 - accuracy: 0.4147 - val_loss: 4.3122 - val_accuracy: 0.2264\n",
            "Epoch 206/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.7865 - accuracy: 0.5072 - val_loss: 3.6145 - val_accuracy: 0.3270\n",
            "Epoch 207/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.8561 - accuracy: 0.4706 - val_loss: 3.5440 - val_accuracy: 0.3396\n",
            "Epoch 208/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 2.7384 - accuracy: 0.4828 - val_loss: 3.4469 - val_accuracy: 0.3333\n",
            "Epoch 209/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 2.6743 - accuracy: 0.5372 - val_loss: 3.6490 - val_accuracy: 0.2767\n",
            "Epoch 210/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 2.6258 - accuracy: 0.5505 - val_loss: 3.8972 - val_accuracy: 0.2704\n",
            "Epoch 211/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 2.5069 - accuracy: 0.5678 - val_loss: 3.1164 - val_accuracy: 0.4906\n",
            "Epoch 212/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.3192 - accuracy: 0.6177 - val_loss: 3.3816 - val_accuracy: 0.3145\n",
            "Epoch 213/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 2.1570 - accuracy: 0.6643 - val_loss: 3.1552 - val_accuracy: 0.4214\n",
            "Epoch 214/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 2.0702 - accuracy: 0.6564 - val_loss: 3.4107 - val_accuracy: 0.2956\n",
            "Epoch 215/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.0837 - accuracy: 0.6526 - val_loss: 3.7592 - val_accuracy: 0.2138\n",
            "Epoch 216/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 2.0781 - accuracy: 0.6284 - val_loss: 3.1808 - val_accuracy: 0.3585\n",
            "Epoch 217/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.9703 - accuracy: 0.6438 - val_loss: 3.7290 - val_accuracy: 0.2642\n",
            "Epoch 218/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.9031 - accuracy: 0.6697 - val_loss: 4.3290 - val_accuracy: 0.1635\n",
            "Epoch 219/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 1.9032 - accuracy: 0.6456 - val_loss: 3.3945 - val_accuracy: 0.3270\n",
            "Epoch 220/5000\n",
            "7/7 [==============================] - 1s 116ms/step - loss: 1.8925 - accuracy: 0.6346 - val_loss: 3.6428 - val_accuracy: 0.2704\n",
            "Epoch 221/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.7852 - accuracy: 0.6722 - val_loss: 3.1334 - val_accuracy: 0.3962\n",
            "Epoch 222/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.7427 - accuracy: 0.7089 - val_loss: 2.8434 - val_accuracy: 0.4591\n",
            "Epoch 223/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.7049 - accuracy: 0.7298 - val_loss: 2.8609 - val_accuracy: 0.5094\n",
            "Epoch 224/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.6546 - accuracy: 0.7382 - val_loss: 2.8699 - val_accuracy: 0.4465\n",
            "Epoch 225/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.8618 - accuracy: 0.6432 - val_loss: 4.1746 - val_accuracy: 0.1761\n",
            "Epoch 226/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.7757 - accuracy: 0.7031 - val_loss: 3.1586 - val_accuracy: 0.3145\n",
            "Epoch 227/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.6641 - accuracy: 0.7295 - val_loss: 2.8912 - val_accuracy: 0.4591\n",
            "Epoch 228/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 1.6090 - accuracy: 0.7382 - val_loss: 2.6610 - val_accuracy: 0.5283\n",
            "Epoch 229/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.4876 - accuracy: 0.7816 - val_loss: 3.0180 - val_accuracy: 0.4340\n",
            "Epoch 230/5000\n",
            "7/7 [==============================] - 1s 146ms/step - loss: 1.4279 - accuracy: 0.8187 - val_loss: 2.7430 - val_accuracy: 0.5283\n",
            "Epoch 231/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.4255 - accuracy: 0.7757 - val_loss: 2.5487 - val_accuracy: 0.5283\n",
            "Epoch 232/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.3528 - accuracy: 0.8045 - val_loss: 2.9807 - val_accuracy: 0.4340\n",
            "Epoch 233/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 1.4013 - accuracy: 0.7857 - val_loss: 2.6140 - val_accuracy: 0.5220\n",
            "Epoch 234/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 1.3738 - accuracy: 0.8240 - val_loss: 2.6881 - val_accuracy: 0.5346\n",
            "Epoch 235/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.2859 - accuracy: 0.8233 - val_loss: 2.8945 - val_accuracy: 0.4465\n",
            "Epoch 236/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 1.2727 - accuracy: 0.8246 - val_loss: 2.7782 - val_accuracy: 0.5220\n",
            "Epoch 237/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.2943 - accuracy: 0.8044 - val_loss: 3.0856 - val_accuracy: 0.3899\n",
            "Epoch 238/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 1.2026 - accuracy: 0.8390 - val_loss: 2.8864 - val_accuracy: 0.5094\n",
            "Epoch 239/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.2398 - accuracy: 0.8295 - val_loss: 2.4327 - val_accuracy: 0.6352\n",
            "Epoch 240/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.2815 - accuracy: 0.8032 - val_loss: 2.8905 - val_accuracy: 0.4591\n",
            "Epoch 241/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.2230 - accuracy: 0.8178 - val_loss: 2.6792 - val_accuracy: 0.5535\n",
            "Epoch 242/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 1.1874 - accuracy: 0.8206 - val_loss: 2.7972 - val_accuracy: 0.4591\n",
            "Epoch 243/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.1770 - accuracy: 0.8294 - val_loss: 2.4961 - val_accuracy: 0.6226\n",
            "Epoch 244/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.1622 - accuracy: 0.8600 - val_loss: 2.7986 - val_accuracy: 0.5472\n",
            "Epoch 245/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.1137 - accuracy: 0.8736 - val_loss: 2.5433 - val_accuracy: 0.6226\n",
            "Epoch 246/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.1163 - accuracy: 0.8797 - val_loss: 2.6995 - val_accuracy: 0.5849\n",
            "Epoch 247/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.0759 - accuracy: 0.8969 - val_loss: 2.4654 - val_accuracy: 0.6101\n",
            "Epoch 248/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 1.0514 - accuracy: 0.8746 - val_loss: 2.4616 - val_accuracy: 0.6352\n",
            "Epoch 249/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.0063 - accuracy: 0.9107 - val_loss: 2.5170 - val_accuracy: 0.6352\n",
            "Epoch 250/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.9960 - accuracy: 0.8902 - val_loss: 2.4542 - val_accuracy: 0.5409\n",
            "Epoch 251/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.0631 - accuracy: 0.8485 - val_loss: 3.0175 - val_accuracy: 0.5723\n",
            "Epoch 252/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.0375 - accuracy: 0.8685 - val_loss: 2.5926 - val_accuracy: 0.6038\n",
            "Epoch 253/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 1.0211 - accuracy: 0.8814 - val_loss: 2.3459 - val_accuracy: 0.6667\n",
            "Epoch 254/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.9142 - accuracy: 0.9256 - val_loss: 2.3720 - val_accuracy: 0.6667\n",
            "Epoch 255/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.9263 - accuracy: 0.9066 - val_loss: 2.4089 - val_accuracy: 0.6038\n",
            "Epoch 256/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.8939 - accuracy: 0.9147 - val_loss: 2.7648 - val_accuracy: 0.5094\n",
            "Epoch 257/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.9109 - accuracy: 0.8926 - val_loss: 2.4266 - val_accuracy: 0.6792\n",
            "Epoch 258/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.9204 - accuracy: 0.8903 - val_loss: 2.3903 - val_accuracy: 0.6289\n",
            "Epoch 259/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.9317 - accuracy: 0.8976 - val_loss: 2.7105 - val_accuracy: 0.5409\n",
            "Epoch 260/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 1.0782 - accuracy: 0.8431 - val_loss: 3.8839 - val_accuracy: 0.3774\n",
            "Epoch 261/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.1806 - accuracy: 0.8036 - val_loss: 3.6744 - val_accuracy: 0.4277\n",
            "Epoch 262/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.2071 - accuracy: 0.7917 - val_loss: 2.7556 - val_accuracy: 0.5723\n",
            "Epoch 263/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.2058 - accuracy: 0.7769 - val_loss: 2.8590 - val_accuracy: 0.6226\n",
            "Epoch 264/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.0223 - accuracy: 0.8819 - val_loss: 3.3290 - val_accuracy: 0.4277\n",
            "Epoch 265/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.0047 - accuracy: 0.9017 - val_loss: 2.6116 - val_accuracy: 0.6478\n",
            "Epoch 266/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.9164 - accuracy: 0.9158 - val_loss: 2.5749 - val_accuracy: 0.6667\n",
            "Epoch 267/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.9201 - accuracy: 0.9030 - val_loss: 2.9758 - val_accuracy: 0.5409\n",
            "Epoch 268/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.0148 - accuracy: 0.8651 - val_loss: 3.4987 - val_accuracy: 0.4340\n",
            "Epoch 269/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 1.0607 - accuracy: 0.8320 - val_loss: 2.8802 - val_accuracy: 0.5660\n",
            "Epoch 270/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.9723 - accuracy: 0.8917 - val_loss: 2.6078 - val_accuracy: 0.6730\n",
            "Epoch 271/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.8567 - accuracy: 0.9388 - val_loss: 2.5125 - val_accuracy: 0.6478\n",
            "Epoch 272/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.8935 - accuracy: 0.9282 - val_loss: 3.7899 - val_accuracy: 0.3899\n",
            "Epoch 273/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.9318 - accuracy: 0.8686 - val_loss: 2.4439 - val_accuracy: 0.6289\n",
            "Epoch 274/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.8360 - accuracy: 0.9233 - val_loss: 2.6264 - val_accuracy: 0.6164\n",
            "Epoch 275/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.7979 - accuracy: 0.9483 - val_loss: 2.5096 - val_accuracy: 0.6289\n",
            "Epoch 276/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.7618 - accuracy: 0.9535 - val_loss: 2.3234 - val_accuracy: 0.6730\n",
            "Epoch 277/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.7198 - accuracy: 0.9633 - val_loss: 2.4802 - val_accuracy: 0.6541\n",
            "Epoch 278/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.7468 - accuracy: 0.9408 - val_loss: 2.6192 - val_accuracy: 0.5786\n",
            "Epoch 279/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.7170 - accuracy: 0.9512 - val_loss: 2.4722 - val_accuracy: 0.6541\n",
            "Epoch 280/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.7144 - accuracy: 0.9393 - val_loss: 2.3005 - val_accuracy: 0.6415\n",
            "Epoch 281/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.7020 - accuracy: 0.9559 - val_loss: 2.3963 - val_accuracy: 0.6415\n",
            "Epoch 282/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.6965 - accuracy: 0.9538 - val_loss: 2.4238 - val_accuracy: 0.6352\n",
            "Epoch 283/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.6998 - accuracy: 0.9409 - val_loss: 2.3861 - val_accuracy: 0.6855\n",
            "Epoch 284/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.6613 - accuracy: 0.9722 - val_loss: 2.4313 - val_accuracy: 0.6101\n",
            "Epoch 285/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.7126 - accuracy: 0.9235 - val_loss: 2.3308 - val_accuracy: 0.6792\n",
            "Epoch 286/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.6948 - accuracy: 0.9540 - val_loss: 2.9294 - val_accuracy: 0.5472\n",
            "Epoch 287/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.7506 - accuracy: 0.9185 - val_loss: 3.0771 - val_accuracy: 0.4465\n",
            "Epoch 288/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.9227 - accuracy: 0.8434 - val_loss: 3.0121 - val_accuracy: 0.5472\n",
            "Epoch 289/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.8574 - accuracy: 0.8805 - val_loss: 2.6829 - val_accuracy: 0.6289\n",
            "Epoch 290/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.7859 - accuracy: 0.9314 - val_loss: 2.9269 - val_accuracy: 0.6352\n",
            "Epoch 291/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.8308 - accuracy: 0.9036 - val_loss: 3.1611 - val_accuracy: 0.5472\n",
            "Epoch 292/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.8162 - accuracy: 0.9156 - val_loss: 2.6561 - val_accuracy: 0.6604\n",
            "Epoch 293/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.8079 - accuracy: 0.9192 - val_loss: 2.5672 - val_accuracy: 0.6164\n",
            "Epoch 294/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.8337 - accuracy: 0.9004 - val_loss: 3.0538 - val_accuracy: 0.6352\n",
            "Epoch 295/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.7759 - accuracy: 0.9258 - val_loss: 3.4217 - val_accuracy: 0.4906\n",
            "Epoch 296/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.7491 - accuracy: 0.9243 - val_loss: 3.0182 - val_accuracy: 0.5660\n",
            "Epoch 297/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.7386 - accuracy: 0.9368 - val_loss: 2.8708 - val_accuracy: 0.6101\n",
            "Epoch 298/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.7013 - accuracy: 0.9447 - val_loss: 3.9446 - val_accuracy: 0.4088\n",
            "Epoch 299/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.7113 - accuracy: 0.9422 - val_loss: 2.7290 - val_accuracy: 0.6038\n",
            "Epoch 300/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.7373 - accuracy: 0.9200 - val_loss: 2.9165 - val_accuracy: 0.6226\n",
            "Epoch 301/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.6994 - accuracy: 0.9438 - val_loss: 2.6239 - val_accuracy: 0.6352\n",
            "Epoch 302/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.6680 - accuracy: 0.9539 - val_loss: 2.5779 - val_accuracy: 0.6667\n",
            "Epoch 303/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.7157 - accuracy: 0.9339 - val_loss: 2.3333 - val_accuracy: 0.7044\n",
            "Epoch 304/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.6493 - accuracy: 0.9471 - val_loss: 2.4366 - val_accuracy: 0.6855\n",
            "Epoch 305/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.5999 - accuracy: 0.9649 - val_loss: 2.5502 - val_accuracy: 0.6604\n",
            "Epoch 306/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.5657 - accuracy: 0.9801 - val_loss: 2.4413 - val_accuracy: 0.6667\n",
            "Epoch 307/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.5700 - accuracy: 0.9636 - val_loss: 2.5950 - val_accuracy: 0.6541\n",
            "Epoch 308/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.5674 - accuracy: 0.9640 - val_loss: 2.4470 - val_accuracy: 0.7044\n",
            "Epoch 309/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.5494 - accuracy: 0.9805 - val_loss: 2.6350 - val_accuracy: 0.6415\n",
            "Epoch 310/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.5473 - accuracy: 0.9776 - val_loss: 2.5469 - val_accuracy: 0.6730\n",
            "Epoch 311/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.5368 - accuracy: 0.9804 - val_loss: 2.5322 - val_accuracy: 0.6352\n",
            "Epoch 312/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.5473 - accuracy: 0.9669 - val_loss: 2.6346 - val_accuracy: 0.6478\n",
            "Epoch 313/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.5414 - accuracy: 0.9694 - val_loss: 2.5021 - val_accuracy: 0.6541\n",
            "Epoch 314/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.6368 - accuracy: 0.9231 - val_loss: 2.5629 - val_accuracy: 0.6101\n",
            "Epoch 315/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.6318 - accuracy: 0.9336 - val_loss: 3.0172 - val_accuracy: 0.5912\n",
            "Epoch 316/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.6719 - accuracy: 0.9119 - val_loss: 2.7792 - val_accuracy: 0.5597\n",
            "Epoch 317/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.6889 - accuracy: 0.9110 - val_loss: 2.7151 - val_accuracy: 0.6352\n",
            "Epoch 318/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.6636 - accuracy: 0.9512 - val_loss: 2.6737 - val_accuracy: 0.6352\n",
            "Epoch 319/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.6669 - accuracy: 0.9461 - val_loss: 3.1377 - val_accuracy: 0.5597\n",
            "Epoch 320/5000\n",
            "7/7 [==============================] - 1s 150ms/step - loss: 0.6404 - accuracy: 0.9659 - val_loss: 2.5782 - val_accuracy: 0.6730\n",
            "Epoch 321/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.6460 - accuracy: 0.9560 - val_loss: 2.7110 - val_accuracy: 0.6855\n",
            "Epoch 322/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.6382 - accuracy: 0.9559 - val_loss: 2.7238 - val_accuracy: 0.6604\n",
            "Epoch 323/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.5959 - accuracy: 0.9617 - val_loss: 2.7108 - val_accuracy: 0.6667\n",
            "Epoch 324/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.5926 - accuracy: 0.9499 - val_loss: 2.7806 - val_accuracy: 0.6604\n",
            "Epoch 325/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.5776 - accuracy: 0.9631 - val_loss: 2.4762 - val_accuracy: 0.7170\n",
            "Epoch 326/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.5531 - accuracy: 0.9698 - val_loss: 2.6596 - val_accuracy: 0.7044\n",
            "Epoch 327/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.5270 - accuracy: 0.9717 - val_loss: 2.7512 - val_accuracy: 0.6415\n",
            "Epoch 328/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.6250 - accuracy: 0.9421 - val_loss: 2.6543 - val_accuracy: 0.6352\n",
            "Epoch 329/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.6552 - accuracy: 0.9302 - val_loss: 2.5802 - val_accuracy: 0.6541\n",
            "Epoch 330/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.7111 - accuracy: 0.9214 - val_loss: 3.2512 - val_accuracy: 0.5660\n",
            "Epoch 331/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.6784 - accuracy: 0.9297 - val_loss: 2.9477 - val_accuracy: 0.5283\n",
            "Epoch 332/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.6664 - accuracy: 0.9437 - val_loss: 2.9647 - val_accuracy: 0.6478\n",
            "Epoch 333/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.6095 - accuracy: 0.9680 - val_loss: 2.8894 - val_accuracy: 0.6604\n",
            "Epoch 334/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.5991 - accuracy: 0.9551 - val_loss: 2.8892 - val_accuracy: 0.6792\n",
            "Epoch 335/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.5598 - accuracy: 0.9753 - val_loss: 2.5227 - val_accuracy: 0.6730\n",
            "Epoch 336/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.5273 - accuracy: 0.9838 - val_loss: 2.5465 - val_accuracy: 0.6792\n",
            "Epoch 337/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.5135 - accuracy: 0.9795 - val_loss: 2.6161 - val_accuracy: 0.7044\n",
            "Epoch 338/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.5196 - accuracy: 0.9772 - val_loss: 2.4665 - val_accuracy: 0.5786\n",
            "Epoch 339/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.5235 - accuracy: 0.9671 - val_loss: 2.6402 - val_accuracy: 0.6855\n",
            "Epoch 340/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.5200 - accuracy: 0.9717 - val_loss: 2.6088 - val_accuracy: 0.6604\n",
            "Epoch 341/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.5437 - accuracy: 0.9554 - val_loss: 2.6703 - val_accuracy: 0.6101\n",
            "Epoch 342/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.5248 - accuracy: 0.9634 - val_loss: 3.8280 - val_accuracy: 0.4214\n",
            "Epoch 343/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.5329 - accuracy: 0.9592 - val_loss: 2.6238 - val_accuracy: 0.6918\n",
            "Epoch 344/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4986 - accuracy: 0.9683 - val_loss: 2.6174 - val_accuracy: 0.6667\n",
            "Epoch 345/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4644 - accuracy: 0.9808 - val_loss: 2.4975 - val_accuracy: 0.6981\n",
            "Epoch 346/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.4503 - accuracy: 0.9871 - val_loss: 2.4182 - val_accuracy: 0.6981\n",
            "Epoch 347/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4872 - accuracy: 0.9647 - val_loss: 2.5200 - val_accuracy: 0.6730\n",
            "Epoch 348/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.4587 - accuracy: 0.9734 - val_loss: 2.3857 - val_accuracy: 0.6604\n",
            "Epoch 349/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.4375 - accuracy: 0.9786 - val_loss: 2.4229 - val_accuracy: 0.6792\n",
            "Epoch 350/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.4459 - accuracy: 0.9834 - val_loss: 2.5743 - val_accuracy: 0.6855\n",
            "Epoch 351/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.4280 - accuracy: 0.9854 - val_loss: 2.4840 - val_accuracy: 0.6730\n",
            "Epoch 352/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.4389 - accuracy: 0.9883 - val_loss: 2.4784 - val_accuracy: 0.6730\n",
            "Epoch 353/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.4330 - accuracy: 0.9798 - val_loss: 2.4533 - val_accuracy: 0.6541\n",
            "Epoch 354/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4769 - accuracy: 0.9643 - val_loss: 2.4060 - val_accuracy: 0.6667\n",
            "Epoch 355/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.4351 - accuracy: 0.9748 - val_loss: 2.6429 - val_accuracy: 0.6730\n",
            "Epoch 356/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4711 - accuracy: 0.9746 - val_loss: 2.6276 - val_accuracy: 0.6792\n",
            "Epoch 357/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.4289 - accuracy: 0.9872 - val_loss: 2.6128 - val_accuracy: 0.6855\n",
            "Epoch 358/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4640 - accuracy: 0.9693 - val_loss: 2.7631 - val_accuracy: 0.6038\n",
            "Epoch 359/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.4347 - accuracy: 0.9804 - val_loss: 2.7672 - val_accuracy: 0.6855\n",
            "Epoch 360/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4272 - accuracy: 0.9815 - val_loss: 2.7319 - val_accuracy: 0.6352\n",
            "Epoch 361/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4146 - accuracy: 0.9852 - val_loss: 2.8831 - val_accuracy: 0.6855\n",
            "Epoch 362/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3982 - accuracy: 0.9867 - val_loss: 2.3049 - val_accuracy: 0.6855\n",
            "Epoch 363/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3991 - accuracy: 0.9920 - val_loss: 2.6581 - val_accuracy: 0.7044\n",
            "Epoch 364/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.3798 - accuracy: 0.9945 - val_loss: 2.5051 - val_accuracy: 0.6792\n",
            "Epoch 365/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3656 - accuracy: 0.9938 - val_loss: 2.3991 - val_accuracy: 0.6855\n",
            "Epoch 366/5000\n",
            "7/7 [==============================] - 1s 116ms/step - loss: 0.3416 - accuracy: 0.9968 - val_loss: 2.6923 - val_accuracy: 0.6667\n",
            "Epoch 367/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3542 - accuracy: 0.9894 - val_loss: 2.3498 - val_accuracy: 0.6667\n",
            "Epoch 368/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.3443 - accuracy: 0.9950 - val_loss: 2.3200 - val_accuracy: 0.6918\n",
            "Epoch 369/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3225 - accuracy: 0.9988 - val_loss: 2.2702 - val_accuracy: 0.6730\n",
            "Epoch 370/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3181 - accuracy: 1.0000 - val_loss: 2.2171 - val_accuracy: 0.6918\n",
            "Epoch 371/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2977 - accuracy: 1.0000 - val_loss: 2.2020 - val_accuracy: 0.6918\n",
            "Epoch 372/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2943 - accuracy: 0.9994 - val_loss: 2.3077 - val_accuracy: 0.6541\n",
            "Epoch 373/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3108 - accuracy: 0.9929 - val_loss: 2.1734 - val_accuracy: 0.6855\n",
            "Epoch 374/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3119 - accuracy: 0.9924 - val_loss: 2.4131 - val_accuracy: 0.6541\n",
            "Epoch 375/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3442 - accuracy: 0.9843 - val_loss: 2.4736 - val_accuracy: 0.6478\n",
            "Epoch 376/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3186 - accuracy: 0.9974 - val_loss: 2.4838 - val_accuracy: 0.6478\n",
            "Epoch 377/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3394 - accuracy: 0.9893 - val_loss: 2.3539 - val_accuracy: 0.6667\n",
            "Epoch 378/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3166 - accuracy: 0.9976 - val_loss: 2.3923 - val_accuracy: 0.6667\n",
            "Epoch 379/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3383 - accuracy: 0.9901 - val_loss: 2.6464 - val_accuracy: 0.6541\n",
            "Epoch 380/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.3290 - accuracy: 0.9848 - val_loss: 2.6461 - val_accuracy: 0.6604\n",
            "Epoch 381/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3452 - accuracy: 0.9784 - val_loss: 3.2142 - val_accuracy: 0.4780\n",
            "Epoch 382/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4406 - accuracy: 0.9525 - val_loss: 3.5891 - val_accuracy: 0.5157\n",
            "Epoch 383/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.6288 - accuracy: 0.8906 - val_loss: 3.0524 - val_accuracy: 0.6478\n",
            "Epoch 384/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4835 - accuracy: 0.9605 - val_loss: 4.4523 - val_accuracy: 0.4717\n",
            "Epoch 385/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.6579 - accuracy: 0.8937 - val_loss: 3.5403 - val_accuracy: 0.6289\n",
            "Epoch 386/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.5772 - accuracy: 0.9412 - val_loss: 3.8114 - val_accuracy: 0.6478\n",
            "Epoch 387/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.5934 - accuracy: 0.9450 - val_loss: 3.2386 - val_accuracy: 0.6541\n",
            "Epoch 388/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.5611 - accuracy: 0.9665 - val_loss: 3.8129 - val_accuracy: 0.6226\n",
            "Epoch 389/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.5499 - accuracy: 0.9653 - val_loss: 3.2865 - val_accuracy: 0.6541\n",
            "Epoch 390/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.5851 - accuracy: 0.9413 - val_loss: 3.6036 - val_accuracy: 0.6604\n",
            "Epoch 391/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.5397 - accuracy: 0.9714 - val_loss: 3.1596 - val_accuracy: 0.6667\n",
            "Epoch 392/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4906 - accuracy: 0.9845 - val_loss: 2.8603 - val_accuracy: 0.6918\n",
            "Epoch 393/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.4682 - accuracy: 0.9752 - val_loss: 3.2387 - val_accuracy: 0.6541\n",
            "Epoch 394/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.5190 - accuracy: 0.9471 - val_loss: 2.8705 - val_accuracy: 0.6792\n",
            "Epoch 395/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.5560 - accuracy: 0.9458 - val_loss: 3.4509 - val_accuracy: 0.5723\n",
            "Epoch 396/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.5993 - accuracy: 0.9392 - val_loss: 2.9847 - val_accuracy: 0.6792\n",
            "Epoch 397/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.5050 - accuracy: 0.9702 - val_loss: 3.0566 - val_accuracy: 0.6478\n",
            "Epoch 398/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.5355 - accuracy: 0.9466 - val_loss: 2.9902 - val_accuracy: 0.6541\n",
            "Epoch 399/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.4817 - accuracy: 0.9732 - val_loss: 2.8145 - val_accuracy: 0.6792\n",
            "Epoch 400/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.4526 - accuracy: 0.9845 - val_loss: 2.8862 - val_accuracy: 0.6918\n",
            "Epoch 401/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4300 - accuracy: 0.9846 - val_loss: 2.4434 - val_accuracy: 0.6918\n",
            "Epoch 402/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3952 - accuracy: 0.9939 - val_loss: 3.0329 - val_accuracy: 0.6101\n",
            "Epoch 403/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3813 - accuracy: 0.9933 - val_loss: 2.4976 - val_accuracy: 0.7044\n",
            "Epoch 404/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3656 - accuracy: 0.9945 - val_loss: 2.6867 - val_accuracy: 0.6855\n",
            "Epoch 405/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3596 - accuracy: 0.9919 - val_loss: 2.5434 - val_accuracy: 0.6981\n",
            "Epoch 406/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.3465 - accuracy: 0.9884 - val_loss: 2.6823 - val_accuracy: 0.6730\n",
            "Epoch 407/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3348 - accuracy: 0.9996 - val_loss: 2.3216 - val_accuracy: 0.6981\n",
            "Epoch 408/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.3393 - accuracy: 0.9893 - val_loss: 2.3811 - val_accuracy: 0.7107\n",
            "Epoch 409/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3134 - accuracy: 0.9962 - val_loss: 2.1653 - val_accuracy: 0.6667\n",
            "Epoch 410/5000\n",
            "7/7 [==============================] - 1s 146ms/step - loss: 0.3273 - accuracy: 0.9922 - val_loss: 2.2107 - val_accuracy: 0.7044\n",
            "Epoch 411/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3041 - accuracy: 0.9934 - val_loss: 2.3709 - val_accuracy: 0.6855\n",
            "Epoch 412/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2995 - accuracy: 0.9918 - val_loss: 2.2049 - val_accuracy: 0.7170\n",
            "Epoch 413/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2910 - accuracy: 0.9962 - val_loss: 2.2114 - val_accuracy: 0.6478\n",
            "Epoch 414/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2967 - accuracy: 0.9965 - val_loss: 2.1681 - val_accuracy: 0.6981\n",
            "Epoch 415/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2732 - accuracy: 0.9991 - val_loss: 2.1766 - val_accuracy: 0.6981\n",
            "Epoch 416/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2722 - accuracy: 0.9984 - val_loss: 2.0741 - val_accuracy: 0.7044\n",
            "Epoch 417/5000\n",
            "7/7 [==============================] - 1s 124ms/step - loss: 0.2632 - accuracy: 0.9957 - val_loss: 2.1603 - val_accuracy: 0.6855\n",
            "Epoch 418/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2657 - accuracy: 1.0000 - val_loss: 2.2906 - val_accuracy: 0.6981\n",
            "Epoch 419/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.2546 - accuracy: 0.9984 - val_loss: 2.3108 - val_accuracy: 0.6667\n",
            "Epoch 420/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2461 - accuracy: 0.9996 - val_loss: 2.0292 - val_accuracy: 0.6667\n",
            "Epoch 421/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2544 - accuracy: 0.9965 - val_loss: 2.2756 - val_accuracy: 0.6667\n",
            "Epoch 422/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3206 - accuracy: 0.9762 - val_loss: 2.4060 - val_accuracy: 0.6289\n",
            "Epoch 423/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4853 - accuracy: 0.9205 - val_loss: 2.9168 - val_accuracy: 0.6038\n",
            "Epoch 424/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.4777 - accuracy: 0.9492 - val_loss: 3.3824 - val_accuracy: 0.6226\n",
            "Epoch 425/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4748 - accuracy: 0.9500 - val_loss: 3.3062 - val_accuracy: 0.6730\n",
            "Epoch 426/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.4418 - accuracy: 0.9753 - val_loss: 3.8900 - val_accuracy: 0.5597\n",
            "Epoch 427/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.4547 - accuracy: 0.9707 - val_loss: 3.1547 - val_accuracy: 0.6604\n",
            "Epoch 428/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.5883 - accuracy: 0.9164 - val_loss: 4.9969 - val_accuracy: 0.4151\n",
            "Epoch 429/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.5591 - accuracy: 0.9291 - val_loss: 4.0496 - val_accuracy: 0.5220\n",
            "Epoch 430/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.5042 - accuracy: 0.9653 - val_loss: 3.4921 - val_accuracy: 0.5975\n",
            "Epoch 431/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.4623 - accuracy: 0.9751 - val_loss: 3.2564 - val_accuracy: 0.6667\n",
            "Epoch 432/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4490 - accuracy: 0.9767 - val_loss: 3.1490 - val_accuracy: 0.6415\n",
            "Epoch 433/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.4334 - accuracy: 0.9779 - val_loss: 2.7968 - val_accuracy: 0.6855\n",
            "Epoch 434/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.4108 - accuracy: 0.9809 - val_loss: 3.0226 - val_accuracy: 0.6918\n",
            "Epoch 435/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3707 - accuracy: 0.9934 - val_loss: 3.0768 - val_accuracy: 0.6667\n",
            "Epoch 436/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.3527 - accuracy: 0.9942 - val_loss: 2.9862 - val_accuracy: 0.6352\n",
            "Epoch 437/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3420 - accuracy: 0.9897 - val_loss: 2.4518 - val_accuracy: 0.7044\n",
            "Epoch 438/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3244 - accuracy: 0.9942 - val_loss: 2.5326 - val_accuracy: 0.7233\n",
            "Epoch 439/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.3156 - accuracy: 0.9939 - val_loss: 2.4863 - val_accuracy: 0.6855\n",
            "Epoch 440/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3065 - accuracy: 0.9953 - val_loss: 2.2414 - val_accuracy: 0.6981\n",
            "Epoch 441/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3138 - accuracy: 0.9902 - val_loss: 2.3383 - val_accuracy: 0.7044\n",
            "Epoch 442/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2826 - accuracy: 1.0000 - val_loss: 2.4148 - val_accuracy: 0.7170\n",
            "Epoch 443/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2784 - accuracy: 1.0000 - val_loss: 2.1677 - val_accuracy: 0.7044\n",
            "Epoch 444/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2712 - accuracy: 0.9961 - val_loss: 2.1363 - val_accuracy: 0.6981\n",
            "Epoch 445/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2624 - accuracy: 0.9958 - val_loss: 2.1469 - val_accuracy: 0.6981\n",
            "Epoch 446/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2485 - accuracy: 1.0000 - val_loss: 2.2757 - val_accuracy: 0.7044\n",
            "Epoch 447/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2483 - accuracy: 0.9954 - val_loss: 2.2672 - val_accuracy: 0.6667\n",
            "Epoch 448/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2378 - accuracy: 0.9983 - val_loss: 2.2222 - val_accuracy: 0.6604\n",
            "Epoch 449/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2340 - accuracy: 0.9980 - val_loss: 2.1739 - val_accuracy: 0.6667\n",
            "Epoch 450/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2454 - accuracy: 0.9919 - val_loss: 2.2386 - val_accuracy: 0.7107\n",
            "Epoch 451/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.2492 - accuracy: 0.9952 - val_loss: 2.7369 - val_accuracy: 0.5597\n",
            "Epoch 452/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2690 - accuracy: 0.9854 - val_loss: 3.3904 - val_accuracy: 0.5094\n",
            "Epoch 453/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3329 - accuracy: 0.9671 - val_loss: 2.8740 - val_accuracy: 0.6415\n",
            "Epoch 454/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3781 - accuracy: 0.9663 - val_loss: 3.1855 - val_accuracy: 0.5912\n",
            "Epoch 455/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3682 - accuracy: 0.9766 - val_loss: 3.0177 - val_accuracy: 0.6352\n",
            "Epoch 456/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3374 - accuracy: 0.9770 - val_loss: 3.0566 - val_accuracy: 0.6855\n",
            "Epoch 457/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3595 - accuracy: 0.9731 - val_loss: 3.0862 - val_accuracy: 0.6038\n",
            "Epoch 458/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.3985 - accuracy: 0.9660 - val_loss: 2.9137 - val_accuracy: 0.6981\n",
            "Epoch 459/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3603 - accuracy: 0.9733 - val_loss: 2.7352 - val_accuracy: 0.6730\n",
            "Epoch 460/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3493 - accuracy: 0.9790 - val_loss: 3.1380 - val_accuracy: 0.6289\n",
            "Epoch 461/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3806 - accuracy: 0.9762 - val_loss: 3.0950 - val_accuracy: 0.6730\n",
            "Epoch 462/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3741 - accuracy: 0.9761 - val_loss: 3.1069 - val_accuracy: 0.6667\n",
            "Epoch 463/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3478 - accuracy: 0.9872 - val_loss: 2.7700 - val_accuracy: 0.6855\n",
            "Epoch 464/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3166 - accuracy: 0.9973 - val_loss: 2.9438 - val_accuracy: 0.6667\n",
            "Epoch 465/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.3289 - accuracy: 0.9858 - val_loss: 2.8305 - val_accuracy: 0.6855\n",
            "Epoch 466/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3971 - accuracy: 0.9618 - val_loss: 2.7417 - val_accuracy: 0.6667\n",
            "Epoch 467/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3920 - accuracy: 0.9667 - val_loss: 3.1972 - val_accuracy: 0.6730\n",
            "Epoch 468/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3966 - accuracy: 0.9637 - val_loss: 2.7375 - val_accuracy: 0.6981\n",
            "Epoch 469/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.3290 - accuracy: 0.9931 - val_loss: 2.8677 - val_accuracy: 0.6352\n",
            "Epoch 470/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3281 - accuracy: 0.9944 - val_loss: 2.8107 - val_accuracy: 0.6918\n",
            "Epoch 471/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3140 - accuracy: 0.9946 - val_loss: 2.6722 - val_accuracy: 0.6855\n",
            "Epoch 472/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2979 - accuracy: 0.9917 - val_loss: 2.6863 - val_accuracy: 0.6792\n",
            "Epoch 473/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2854 - accuracy: 0.9979 - val_loss: 2.7334 - val_accuracy: 0.6478\n",
            "Epoch 474/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2726 - accuracy: 1.0000 - val_loss: 2.7188 - val_accuracy: 0.6855\n",
            "Epoch 475/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.2564 - accuracy: 1.0000 - val_loss: 2.4435 - val_accuracy: 0.6981\n",
            "Epoch 476/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.2450 - accuracy: 0.9984 - val_loss: 2.3382 - val_accuracy: 0.6918\n",
            "Epoch 477/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.2345 - accuracy: 0.9994 - val_loss: 2.2754 - val_accuracy: 0.6918\n",
            "Epoch 478/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2256 - accuracy: 0.9984 - val_loss: 2.2482 - val_accuracy: 0.6981\n",
            "Epoch 479/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2157 - accuracy: 1.0000 - val_loss: 2.3406 - val_accuracy: 0.6855\n",
            "Epoch 480/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2119 - accuracy: 1.0000 - val_loss: 2.2957 - val_accuracy: 0.6918\n",
            "Epoch 481/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2026 - accuracy: 1.0000 - val_loss: 2.2106 - val_accuracy: 0.7044\n",
            "Epoch 482/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1991 - accuracy: 1.0000 - val_loss: 2.2128 - val_accuracy: 0.6792\n",
            "Epoch 483/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2078 - accuracy: 0.9948 - val_loss: 2.4991 - val_accuracy: 0.6038\n",
            "Epoch 484/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2035 - accuracy: 0.9964 - val_loss: 2.2658 - val_accuracy: 0.6855\n",
            "Epoch 485/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3313 - accuracy: 0.9614 - val_loss: 8.0237 - val_accuracy: 0.1132\n",
            "Epoch 486/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 1.3579 - accuracy: 0.6711 - val_loss: 4.9984 - val_accuracy: 0.4465\n",
            "Epoch 487/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.9116 - accuracy: 0.8650 - val_loss: 5.2923 - val_accuracy: 0.5409\n",
            "Epoch 488/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 1.0652 - accuracy: 0.8703 - val_loss: 9.9775 - val_accuracy: 0.2642\n",
            "Epoch 489/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 1.1118 - accuracy: 0.8618 - val_loss: 6.9686 - val_accuracy: 0.3836\n",
            "Epoch 490/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.9451 - accuracy: 0.9174 - val_loss: 4.1553 - val_accuracy: 0.6226\n",
            "Epoch 491/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.8101 - accuracy: 0.9624 - val_loss: 5.0820 - val_accuracy: 0.5849\n",
            "Epoch 492/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.6867 - accuracy: 0.9888 - val_loss: 3.9145 - val_accuracy: 0.6918\n",
            "Epoch 493/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.6087 - accuracy: 0.9786 - val_loss: 3.5747 - val_accuracy: 0.6855\n",
            "Epoch 494/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.5329 - accuracy: 0.9900 - val_loss: 3.3801 - val_accuracy: 0.6918\n",
            "Epoch 495/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.4843 - accuracy: 0.9928 - val_loss: 3.0860 - val_accuracy: 0.6541\n",
            "Epoch 496/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.4492 - accuracy: 0.9938 - val_loss: 2.9241 - val_accuracy: 0.6918\n",
            "Epoch 497/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.4096 - accuracy: 0.9942 - val_loss: 3.1288 - val_accuracy: 0.6415\n",
            "Epoch 498/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3933 - accuracy: 0.9952 - val_loss: 2.7641 - val_accuracy: 0.6792\n",
            "Epoch 499/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3885 - accuracy: 0.9935 - val_loss: 2.7255 - val_accuracy: 0.6730\n",
            "Epoch 500/5000\n",
            "7/7 [==============================] - 1s 148ms/step - loss: 0.3782 - accuracy: 0.9916 - val_loss: 2.8663 - val_accuracy: 0.7044\n",
            "Epoch 501/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3579 - accuracy: 0.9856 - val_loss: 2.5824 - val_accuracy: 0.6667\n",
            "Epoch 502/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.3426 - accuracy: 0.9942 - val_loss: 2.5053 - val_accuracy: 0.7044\n",
            "Epoch 503/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3154 - accuracy: 0.9994 - val_loss: 2.5367 - val_accuracy: 0.6855\n",
            "Epoch 504/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.3112 - accuracy: 0.9965 - val_loss: 2.4091 - val_accuracy: 0.6981\n",
            "Epoch 505/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2842 - accuracy: 0.9988 - val_loss: 2.3635 - val_accuracy: 0.7044\n",
            "Epoch 506/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2701 - accuracy: 1.0000 - val_loss: 2.2266 - val_accuracy: 0.6981\n",
            "Epoch 507/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.2672 - accuracy: 0.9965 - val_loss: 2.2172 - val_accuracy: 0.7044\n",
            "Epoch 508/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2707 - accuracy: 0.9911 - val_loss: 2.5047 - val_accuracy: 0.6855\n",
            "Epoch 509/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2636 - accuracy: 0.9957 - val_loss: 2.3508 - val_accuracy: 0.6730\n",
            "Epoch 510/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2745 - accuracy: 0.9858 - val_loss: 2.6029 - val_accuracy: 0.6918\n",
            "Epoch 511/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2789 - accuracy: 0.9853 - val_loss: 2.6387 - val_accuracy: 0.6855\n",
            "Epoch 512/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2534 - accuracy: 0.9956 - val_loss: 2.5988 - val_accuracy: 0.6415\n",
            "Epoch 513/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.2441 - accuracy: 0.9959 - val_loss: 2.3998 - val_accuracy: 0.6792\n",
            "Epoch 514/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2307 - accuracy: 0.9983 - val_loss: 2.2871 - val_accuracy: 0.6855\n",
            "Epoch 515/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2337 - accuracy: 0.9951 - val_loss: 2.1767 - val_accuracy: 0.6730\n",
            "Epoch 516/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2249 - accuracy: 0.9954 - val_loss: 2.3390 - val_accuracy: 0.6792\n",
            "Epoch 517/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2170 - accuracy: 0.9965 - val_loss: 2.3277 - val_accuracy: 0.6855\n",
            "Epoch 518/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2047 - accuracy: 1.0000 - val_loss: 2.2076 - val_accuracy: 0.7107\n",
            "Epoch 519/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2000 - accuracy: 0.9994 - val_loss: 2.0714 - val_accuracy: 0.6918\n",
            "Epoch 520/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1993 - accuracy: 0.9984 - val_loss: 2.0929 - val_accuracy: 0.6792\n",
            "Epoch 521/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1910 - accuracy: 0.9991 - val_loss: 2.1068 - val_accuracy: 0.6981\n",
            "Epoch 522/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.1855 - accuracy: 0.9988 - val_loss: 2.0393 - val_accuracy: 0.7044\n",
            "Epoch 523/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.1807 - accuracy: 1.0000 - val_loss: 2.0001 - val_accuracy: 0.7044\n",
            "Epoch 524/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1831 - accuracy: 0.9931 - val_loss: 2.0907 - val_accuracy: 0.6855\n",
            "Epoch 525/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1847 - accuracy: 0.9966 - val_loss: 2.0951 - val_accuracy: 0.7170\n",
            "Epoch 526/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1856 - accuracy: 0.9935 - val_loss: 2.0607 - val_accuracy: 0.6541\n",
            "Epoch 527/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1939 - accuracy: 0.9877 - val_loss: 2.4217 - val_accuracy: 0.6604\n",
            "Epoch 528/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1927 - accuracy: 0.9939 - val_loss: 2.3876 - val_accuracy: 0.6415\n",
            "Epoch 529/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2028 - accuracy: 0.9877 - val_loss: 2.4405 - val_accuracy: 0.6541\n",
            "Epoch 530/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1838 - accuracy: 0.9965 - val_loss: 2.5164 - val_accuracy: 0.6855\n",
            "Epoch 531/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.1852 - accuracy: 0.9965 - val_loss: 2.1656 - val_accuracy: 0.6855\n",
            "Epoch 532/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1784 - accuracy: 0.9978 - val_loss: 2.4682 - val_accuracy: 0.7044\n",
            "Epoch 533/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1718 - accuracy: 1.0000 - val_loss: 2.4108 - val_accuracy: 0.6478\n",
            "Epoch 534/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.1733 - accuracy: 1.0000 - val_loss: 2.3287 - val_accuracy: 0.7107\n",
            "Epoch 535/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1633 - accuracy: 0.9988 - val_loss: 2.2984 - val_accuracy: 0.6981\n",
            "Epoch 536/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1663 - accuracy: 0.9961 - val_loss: 2.1515 - val_accuracy: 0.6792\n",
            "Epoch 537/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1628 - accuracy: 0.9992 - val_loss: 2.4195 - val_accuracy: 0.6855\n",
            "Epoch 538/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2094 - accuracy: 0.9808 - val_loss: 4.7969 - val_accuracy: 0.3082\n",
            "Epoch 539/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2236 - accuracy: 0.9784 - val_loss: 5.0406 - val_accuracy: 0.3396\n",
            "Epoch 540/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2908 - accuracy: 0.9603 - val_loss: 3.8514 - val_accuracy: 0.5912\n",
            "Epoch 541/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2891 - accuracy: 0.9704 - val_loss: 3.4479 - val_accuracy: 0.5849\n",
            "Epoch 542/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3491 - accuracy: 0.9563 - val_loss: 3.3147 - val_accuracy: 0.7170\n",
            "Epoch 543/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2890 - accuracy: 0.9860 - val_loss: 3.6930 - val_accuracy: 0.6415\n",
            "Epoch 544/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3062 - accuracy: 0.9737 - val_loss: 4.7863 - val_accuracy: 0.5220\n",
            "Epoch 545/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3269 - accuracy: 0.9750 - val_loss: 3.4196 - val_accuracy: 0.6855\n",
            "Epoch 546/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3223 - accuracy: 0.9816 - val_loss: 2.8960 - val_accuracy: 0.7107\n",
            "Epoch 547/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2954 - accuracy: 0.9842 - val_loss: 2.9604 - val_accuracy: 0.6289\n",
            "Epoch 548/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2916 - accuracy: 0.9818 - val_loss: 3.1296 - val_accuracy: 0.6541\n",
            "Epoch 549/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.2712 - accuracy: 0.9928 - val_loss: 3.0128 - val_accuracy: 0.6792\n",
            "Epoch 550/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2448 - accuracy: 0.9973 - val_loss: 2.5596 - val_accuracy: 0.6918\n",
            "Epoch 551/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.2282 - accuracy: 0.9978 - val_loss: 2.5515 - val_accuracy: 0.6792\n",
            "Epoch 552/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.2153 - accuracy: 1.0000 - val_loss: 2.4620 - val_accuracy: 0.6792\n",
            "Epoch 553/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2059 - accuracy: 1.0000 - val_loss: 2.5662 - val_accuracy: 0.7170\n",
            "Epoch 554/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1966 - accuracy: 0.9980 - val_loss: 2.5256 - val_accuracy: 0.7044\n",
            "Epoch 555/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1841 - accuracy: 1.0000 - val_loss: 2.3241 - val_accuracy: 0.7170\n",
            "Epoch 556/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1795 - accuracy: 1.0000 - val_loss: 2.3293 - val_accuracy: 0.6918\n",
            "Epoch 557/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1707 - accuracy: 1.0000 - val_loss: 2.4538 - val_accuracy: 0.6792\n",
            "Epoch 558/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.1700 - accuracy: 0.9992 - val_loss: 2.3956 - val_accuracy: 0.6918\n",
            "Epoch 559/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1887 - accuracy: 0.9945 - val_loss: 2.2339 - val_accuracy: 0.7044\n",
            "Epoch 560/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1962 - accuracy: 0.9919 - val_loss: 2.4661 - val_accuracy: 0.6604\n",
            "Epoch 561/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2118 - accuracy: 0.9878 - val_loss: 2.4783 - val_accuracy: 0.7044\n",
            "Epoch 562/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1954 - accuracy: 0.9964 - val_loss: 2.4079 - val_accuracy: 0.6918\n",
            "Epoch 563/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2145 - accuracy: 0.9873 - val_loss: 2.7682 - val_accuracy: 0.6478\n",
            "Epoch 564/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2847 - accuracy: 0.9632 - val_loss: 3.1640 - val_accuracy: 0.6667\n",
            "Epoch 565/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2861 - accuracy: 0.9683 - val_loss: 3.0600 - val_accuracy: 0.6038\n",
            "Epoch 566/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3203 - accuracy: 0.9624 - val_loss: 3.1805 - val_accuracy: 0.6352\n",
            "Epoch 567/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3407 - accuracy: 0.9618 - val_loss: 3.5596 - val_accuracy: 0.6855\n",
            "Epoch 568/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2985 - accuracy: 0.9884 - val_loss: 3.3110 - val_accuracy: 0.6792\n",
            "Epoch 569/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2854 - accuracy: 0.9893 - val_loss: 3.2079 - val_accuracy: 0.6981\n",
            "Epoch 570/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2508 - accuracy: 0.9966 - val_loss: 3.2390 - val_accuracy: 0.6730\n",
            "Epoch 571/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2564 - accuracy: 0.9936 - val_loss: 3.1100 - val_accuracy: 0.6541\n",
            "Epoch 572/5000\n",
            "7/7 [==============================] - 1s 147ms/step - loss: 0.2350 - accuracy: 0.9957 - val_loss: 2.9023 - val_accuracy: 0.6981\n",
            "Epoch 573/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2456 - accuracy: 0.9915 - val_loss: 3.1306 - val_accuracy: 0.6792\n",
            "Epoch 574/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2319 - accuracy: 0.9890 - val_loss: 2.6843 - val_accuracy: 0.6792\n",
            "Epoch 575/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2253 - accuracy: 0.9956 - val_loss: 2.6193 - val_accuracy: 0.6918\n",
            "Epoch 576/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2259 - accuracy: 0.9940 - val_loss: 2.7457 - val_accuracy: 0.6730\n",
            "Epoch 577/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2323 - accuracy: 0.9918 - val_loss: 2.5762 - val_accuracy: 0.6667\n",
            "Epoch 578/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2245 - accuracy: 0.9912 - val_loss: 2.7364 - val_accuracy: 0.6918\n",
            "Epoch 579/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2374 - accuracy: 0.9857 - val_loss: 2.9120 - val_accuracy: 0.6730\n",
            "Epoch 580/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2403 - accuracy: 0.9819 - val_loss: 2.8356 - val_accuracy: 0.6604\n",
            "Epoch 581/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.2626 - accuracy: 0.9774 - val_loss: 2.6034 - val_accuracy: 0.6918\n",
            "Epoch 582/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2877 - accuracy: 0.9691 - val_loss: 3.0951 - val_accuracy: 0.6730\n",
            "Epoch 583/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2506 - accuracy: 0.9869 - val_loss: 2.7481 - val_accuracy: 0.6918\n",
            "Epoch 584/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2578 - accuracy: 0.9816 - val_loss: 3.1048 - val_accuracy: 0.6667\n",
            "Epoch 585/5000\n",
            "7/7 [==============================] - 1s 124ms/step - loss: 0.2568 - accuracy: 0.9909 - val_loss: 3.0581 - val_accuracy: 0.6730\n",
            "Epoch 586/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2529 - accuracy: 0.9855 - val_loss: 2.7585 - val_accuracy: 0.7233\n",
            "Epoch 587/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2542 - accuracy: 0.9905 - val_loss: 2.6946 - val_accuracy: 0.6792\n",
            "Epoch 588/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2264 - accuracy: 0.9941 - val_loss: 2.7375 - val_accuracy: 0.6855\n",
            "Epoch 589/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2141 - accuracy: 0.9930 - val_loss: 2.8137 - val_accuracy: 0.6855\n",
            "Epoch 590/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2072 - accuracy: 0.9962 - val_loss: 2.4723 - val_accuracy: 0.6730\n",
            "Epoch 591/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1976 - accuracy: 0.9977 - val_loss: 2.3607 - val_accuracy: 0.6981\n",
            "Epoch 592/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1924 - accuracy: 0.9978 - val_loss: 2.5116 - val_accuracy: 0.7044\n",
            "Epoch 593/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1810 - accuracy: 1.0000 - val_loss: 2.5748 - val_accuracy: 0.6981\n",
            "Epoch 594/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.1744 - accuracy: 1.0000 - val_loss: 2.5155 - val_accuracy: 0.6981\n",
            "Epoch 595/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1768 - accuracy: 0.9965 - val_loss: 2.3892 - val_accuracy: 0.7107\n",
            "Epoch 596/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1638 - accuracy: 1.0000 - val_loss: 2.4310 - val_accuracy: 0.6604\n",
            "Epoch 597/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1689 - accuracy: 1.0000 - val_loss: 2.2168 - val_accuracy: 0.6855\n",
            "Epoch 598/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1558 - accuracy: 0.9991 - val_loss: 2.2746 - val_accuracy: 0.6981\n",
            "Epoch 599/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1514 - accuracy: 1.0000 - val_loss: 2.4341 - val_accuracy: 0.6855\n",
            "Epoch 600/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1457 - accuracy: 0.9984 - val_loss: 2.2346 - val_accuracy: 0.6667\n",
            "Epoch 601/5000\n",
            "7/7 [==============================] - 1s 124ms/step - loss: 0.1402 - accuracy: 1.0000 - val_loss: 2.1831 - val_accuracy: 0.6667\n",
            "Epoch 602/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1451 - accuracy: 0.9991 - val_loss: 2.2577 - val_accuracy: 0.6792\n",
            "Epoch 603/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1348 - accuracy: 1.0000 - val_loss: 2.1395 - val_accuracy: 0.6918\n",
            "Epoch 604/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1308 - accuracy: 1.0000 - val_loss: 2.1099 - val_accuracy: 0.6981\n",
            "Epoch 605/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1297 - accuracy: 1.0000 - val_loss: 2.0750 - val_accuracy: 0.6918\n",
            "Epoch 606/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1230 - accuracy: 1.0000 - val_loss: 2.1139 - val_accuracy: 0.6792\n",
            "Epoch 607/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.1220 - accuracy: 1.0000 - val_loss: 2.0961 - val_accuracy: 0.6541\n",
            "Epoch 608/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1193 - accuracy: 0.9978 - val_loss: 2.2483 - val_accuracy: 0.6352\n",
            "Epoch 609/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1225 - accuracy: 0.9994 - val_loss: 2.0794 - val_accuracy: 0.6730\n",
            "Epoch 610/5000\n",
            "7/7 [==============================] - 1s 124ms/step - loss: 0.1249 - accuracy: 0.9966 - val_loss: 2.0692 - val_accuracy: 0.6918\n",
            "Epoch 611/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1280 - accuracy: 0.9940 - val_loss: 2.2179 - val_accuracy: 0.6981\n",
            "Epoch 612/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1317 - accuracy: 0.9966 - val_loss: 3.0506 - val_accuracy: 0.5157\n",
            "Epoch 613/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1443 - accuracy: 0.9932 - val_loss: 3.1834 - val_accuracy: 0.5472\n",
            "Epoch 614/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1451 - accuracy: 0.9933 - val_loss: 3.2524 - val_accuracy: 0.5786\n",
            "Epoch 615/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1498 - accuracy: 0.9934 - val_loss: 2.7699 - val_accuracy: 0.6730\n",
            "Epoch 616/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1604 - accuracy: 0.9920 - val_loss: 2.9769 - val_accuracy: 0.6478\n",
            "Epoch 617/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1771 - accuracy: 0.9933 - val_loss: 3.2873 - val_accuracy: 0.6604\n",
            "Epoch 618/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1768 - accuracy: 0.9983 - val_loss: 2.8685 - val_accuracy: 0.6541\n",
            "Epoch 619/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1784 - accuracy: 0.9936 - val_loss: 2.6880 - val_accuracy: 0.6667\n",
            "Epoch 620/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1825 - accuracy: 0.9931 - val_loss: 2.8085 - val_accuracy: 0.7107\n",
            "Epoch 621/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1875 - accuracy: 0.9923 - val_loss: 3.6514 - val_accuracy: 0.6164\n",
            "Epoch 622/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.1778 - accuracy: 0.9978 - val_loss: 3.0672 - val_accuracy: 0.6855\n",
            "Epoch 623/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1621 - accuracy: 0.9974 - val_loss: 2.4839 - val_accuracy: 0.7044\n",
            "Epoch 624/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1582 - accuracy: 0.9984 - val_loss: 2.8698 - val_accuracy: 0.6792\n",
            "Epoch 625/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.1570 - accuracy: 0.9964 - val_loss: 2.9440 - val_accuracy: 0.7044\n",
            "Epoch 626/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1553 - accuracy: 0.9984 - val_loss: 3.1092 - val_accuracy: 0.7044\n",
            "Epoch 627/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.1515 - accuracy: 0.9984 - val_loss: 2.4503 - val_accuracy: 0.6604\n",
            "Epoch 628/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1461 - accuracy: 1.0000 - val_loss: 2.5313 - val_accuracy: 0.7107\n",
            "Epoch 629/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.1445 - accuracy: 0.9980 - val_loss: 2.6834 - val_accuracy: 0.6855\n",
            "Epoch 630/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1399 - accuracy: 0.9958 - val_loss: 2.4913 - val_accuracy: 0.6855\n",
            "Epoch 631/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1346 - accuracy: 0.9991 - val_loss: 2.6384 - val_accuracy: 0.6855\n",
            "Epoch 632/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1263 - accuracy: 1.0000 - val_loss: 2.5198 - val_accuracy: 0.6855\n",
            "Epoch 633/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1210 - accuracy: 1.0000 - val_loss: 2.3469 - val_accuracy: 0.7170\n",
            "Epoch 634/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1199 - accuracy: 0.9984 - val_loss: 2.3133 - val_accuracy: 0.6855\n",
            "Epoch 635/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1129 - accuracy: 1.0000 - val_loss: 2.2692 - val_accuracy: 0.6855\n",
            "Epoch 636/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1131 - accuracy: 1.0000 - val_loss: 2.2949 - val_accuracy: 0.6918\n",
            "Epoch 637/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1076 - accuracy: 1.0000 - val_loss: 2.2944 - val_accuracy: 0.6792\n",
            "Epoch 638/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1090 - accuracy: 1.0000 - val_loss: 2.1006 - val_accuracy: 0.6918\n",
            "Epoch 639/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1037 - accuracy: 1.0000 - val_loss: 2.0517 - val_accuracy: 0.7044\n",
            "Epoch 640/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1006 - accuracy: 0.9994 - val_loss: 2.1795 - val_accuracy: 0.7044\n",
            "Epoch 641/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1053 - accuracy: 0.9978 - val_loss: 2.1695 - val_accuracy: 0.6918\n",
            "Epoch 642/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1147 - accuracy: 0.9974 - val_loss: 2.0148 - val_accuracy: 0.6667\n",
            "Epoch 643/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1215 - accuracy: 0.9944 - val_loss: 2.7473 - val_accuracy: 0.6478\n",
            "Epoch 644/5000\n",
            "7/7 [==============================] - 1s 150ms/step - loss: 0.1599 - accuracy: 0.9879 - val_loss: 2.8739 - val_accuracy: 0.6226\n",
            "Epoch 645/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2015 - accuracy: 0.9788 - val_loss: 3.7902 - val_accuracy: 0.5535\n",
            "Epoch 646/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2024 - accuracy: 0.9760 - val_loss: 3.2825 - val_accuracy: 0.6541\n",
            "Epoch 647/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2016 - accuracy: 0.9891 - val_loss: 3.8426 - val_accuracy: 0.6352\n",
            "Epoch 648/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2129 - accuracy: 0.9831 - val_loss: 3.3833 - val_accuracy: 0.6667\n",
            "Epoch 649/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2405 - accuracy: 0.9707 - val_loss: 5.1288 - val_accuracy: 0.3899\n",
            "Epoch 650/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3383 - accuracy: 0.9339 - val_loss: 11.9148 - val_accuracy: 0.1887\n",
            "Epoch 651/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.4183 - accuracy: 0.9224 - val_loss: 3.7323 - val_accuracy: 0.6604\n",
            "Epoch 652/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.3716 - accuracy: 0.9484 - val_loss: 3.9261 - val_accuracy: 0.6478\n",
            "Epoch 653/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3354 - accuracy: 0.9744 - val_loss: 4.2308 - val_accuracy: 0.6352\n",
            "Epoch 654/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.3104 - accuracy: 0.9815 - val_loss: 4.7171 - val_accuracy: 0.5975\n",
            "Epoch 655/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3261 - accuracy: 0.9730 - val_loss: 3.9333 - val_accuracy: 0.6604\n",
            "Epoch 656/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.3073 - accuracy: 0.9751 - val_loss: 4.0742 - val_accuracy: 0.6478\n",
            "Epoch 657/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2877 - accuracy: 0.9770 - val_loss: 3.8351 - val_accuracy: 0.6855\n",
            "Epoch 658/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.2898 - accuracy: 0.9814 - val_loss: 3.9380 - val_accuracy: 0.6667\n",
            "Epoch 659/5000\n",
            "7/7 [==============================] - 1s 124ms/step - loss: 0.2312 - accuracy: 0.9964 - val_loss: 3.5125 - val_accuracy: 0.6730\n",
            "Epoch 660/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.2217 - accuracy: 0.9961 - val_loss: 3.2014 - val_accuracy: 0.6918\n",
            "Epoch 661/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.2295 - accuracy: 0.9901 - val_loss: 3.2894 - val_accuracy: 0.6730\n",
            "Epoch 662/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2396 - accuracy: 0.9848 - val_loss: 3.1448 - val_accuracy: 0.6730\n",
            "Epoch 663/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2484 - accuracy: 0.9801 - val_loss: 3.0933 - val_accuracy: 0.6415\n",
            "Epoch 664/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2556 - accuracy: 0.9826 - val_loss: 2.9586 - val_accuracy: 0.6981\n",
            "Epoch 665/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2235 - accuracy: 0.9910 - val_loss: 3.2270 - val_accuracy: 0.5975\n",
            "Epoch 666/5000\n",
            "7/7 [==============================] - 1s 116ms/step - loss: 0.2274 - accuracy: 0.9838 - val_loss: 2.7982 - val_accuracy: 0.6792\n",
            "Epoch 667/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2017 - accuracy: 0.9972 - val_loss: 2.5700 - val_accuracy: 0.6918\n",
            "Epoch 668/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.1928 - accuracy: 0.9951 - val_loss: 2.7191 - val_accuracy: 0.6792\n",
            "Epoch 669/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1847 - accuracy: 0.9982 - val_loss: 2.4982 - val_accuracy: 0.6918\n",
            "Epoch 670/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1718 - accuracy: 0.9961 - val_loss: 2.4381 - val_accuracy: 0.6981\n",
            "Epoch 671/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1630 - accuracy: 0.9980 - val_loss: 2.6050 - val_accuracy: 0.6981\n",
            "Epoch 672/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1669 - accuracy: 0.9962 - val_loss: 2.4798 - val_accuracy: 0.6855\n",
            "Epoch 673/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1548 - accuracy: 0.9976 - val_loss: 2.4432 - val_accuracy: 0.6981\n",
            "Epoch 674/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1549 - accuracy: 0.9971 - val_loss: 2.4713 - val_accuracy: 0.6981\n",
            "Epoch 675/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1467 - accuracy: 0.9991 - val_loss: 2.3885 - val_accuracy: 0.6981\n",
            "Epoch 676/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1365 - accuracy: 1.0000 - val_loss: 2.3235 - val_accuracy: 0.6855\n",
            "Epoch 677/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1305 - accuracy: 1.0000 - val_loss: 2.2402 - val_accuracy: 0.6981\n",
            "Epoch 678/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1272 - accuracy: 1.0000 - val_loss: 2.1902 - val_accuracy: 0.6981\n",
            "Epoch 679/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.1237 - accuracy: 1.0000 - val_loss: 2.1460 - val_accuracy: 0.6918\n",
            "Epoch 680/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.1214 - accuracy: 0.9991 - val_loss: 2.2887 - val_accuracy: 0.6918\n",
            "Epoch 681/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1165 - accuracy: 0.9996 - val_loss: 2.4424 - val_accuracy: 0.6792\n",
            "Epoch 682/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.1382 - accuracy: 0.9963 - val_loss: 2.5204 - val_accuracy: 0.6667\n",
            "Epoch 683/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1493 - accuracy: 0.9916 - val_loss: 2.7127 - val_accuracy: 0.6415\n",
            "Epoch 684/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.1751 - accuracy: 0.9827 - val_loss: 2.7492 - val_accuracy: 0.6226\n",
            "Epoch 685/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.1858 - accuracy: 0.9770 - val_loss: 2.8538 - val_accuracy: 0.6226\n",
            "Epoch 686/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1732 - accuracy: 0.9845 - val_loss: 2.7802 - val_accuracy: 0.6855\n",
            "Epoch 687/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.1959 - accuracy: 0.9849 - val_loss: 3.0196 - val_accuracy: 0.6352\n",
            "Epoch 688/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1884 - accuracy: 0.9832 - val_loss: 2.9764 - val_accuracy: 0.6855\n",
            "Epoch 689/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1696 - accuracy: 0.9935 - val_loss: 2.7836 - val_accuracy: 0.6541\n",
            "Epoch 690/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.2008 - accuracy: 0.9845 - val_loss: 3.0622 - val_accuracy: 0.6604\n",
            "Epoch 691/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.2236 - accuracy: 0.9719 - val_loss: 3.1922 - val_accuracy: 0.6289\n",
            "Epoch 692/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2599 - accuracy: 0.9681 - val_loss: 3.9962 - val_accuracy: 0.5472\n",
            "Epoch 693/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3440 - accuracy: 0.9367 - val_loss: 5.1523 - val_accuracy: 0.5597\n",
            "Epoch 694/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2897 - accuracy: 0.9770 - val_loss: 4.1622 - val_accuracy: 0.6478\n",
            "Epoch 695/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.3093 - accuracy: 0.9719 - val_loss: 3.6218 - val_accuracy: 0.6541\n",
            "Epoch 696/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.3224 - accuracy: 0.9680 - val_loss: 3.4765 - val_accuracy: 0.6792\n",
            "Epoch 697/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.2906 - accuracy: 0.9851 - val_loss: 3.9337 - val_accuracy: 0.6604\n",
            "Epoch 698/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2993 - accuracy: 0.9816 - val_loss: 3.6506 - val_accuracy: 0.6226\n",
            "Epoch 699/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.4028 - accuracy: 0.9493 - val_loss: 3.5827 - val_accuracy: 0.6604\n",
            "Epoch 700/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.3761 - accuracy: 0.9564 - val_loss: 4.0662 - val_accuracy: 0.5660\n",
            "Epoch 701/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.3720 - accuracy: 0.9651 - val_loss: 3.2135 - val_accuracy: 0.7044\n",
            "Epoch 702/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.3404 - accuracy: 0.9713 - val_loss: 3.2998 - val_accuracy: 0.6792\n",
            "Epoch 703/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.3132 - accuracy: 0.9851 - val_loss: 3.1139 - val_accuracy: 0.6855\n",
            "Epoch 704/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.2796 - accuracy: 0.9846 - val_loss: 3.0175 - val_accuracy: 0.6667\n",
            "Epoch 705/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.2538 - accuracy: 0.9920 - val_loss: 2.8135 - val_accuracy: 0.6792\n",
            "Epoch 706/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2253 - accuracy: 0.9966 - val_loss: 3.0137 - val_accuracy: 0.7107\n",
            "Epoch 707/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.2278 - accuracy: 0.9951 - val_loss: 2.6979 - val_accuracy: 0.7170\n",
            "Epoch 708/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.1992 - accuracy: 0.9969 - val_loss: 2.7067 - val_accuracy: 0.6855\n",
            "Epoch 709/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1845 - accuracy: 1.0000 - val_loss: 2.6545 - val_accuracy: 0.7170\n",
            "Epoch 710/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1756 - accuracy: 1.0000 - val_loss: 2.5560 - val_accuracy: 0.7044\n",
            "Epoch 711/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1690 - accuracy: 0.9994 - val_loss: 2.6935 - val_accuracy: 0.6730\n",
            "Epoch 712/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1687 - accuracy: 0.9991 - val_loss: 2.5490 - val_accuracy: 0.6918\n",
            "Epoch 713/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1549 - accuracy: 1.0000 - val_loss: 2.3842 - val_accuracy: 0.7044\n",
            "Epoch 714/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1487 - accuracy: 1.0000 - val_loss: 2.1565 - val_accuracy: 0.7233\n",
            "Epoch 715/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1455 - accuracy: 0.9988 - val_loss: 2.0679 - val_accuracy: 0.7107\n",
            "Epoch 716/5000\n",
            "7/7 [==============================] - 1s 147ms/step - loss: 0.1389 - accuracy: 0.9991 - val_loss: 2.1181 - val_accuracy: 0.7107\n",
            "Epoch 717/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1320 - accuracy: 1.0000 - val_loss: 2.1886 - val_accuracy: 0.6981\n",
            "Epoch 718/5000\n",
            "7/7 [==============================] - 1s 123ms/step - loss: 0.1315 - accuracy: 1.0000 - val_loss: 2.1084 - val_accuracy: 0.7107\n",
            "Epoch 719/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1245 - accuracy: 1.0000 - val_loss: 2.0511 - val_accuracy: 0.7107\n",
            "Epoch 720/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1182 - accuracy: 1.0000 - val_loss: 2.0114 - val_accuracy: 0.6730\n",
            "Epoch 721/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1206 - accuracy: 0.9962 - val_loss: 2.2048 - val_accuracy: 0.6604\n",
            "Epoch 722/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1148 - accuracy: 1.0000 - val_loss: 2.1257 - val_accuracy: 0.6981\n",
            "Epoch 723/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.1159 - accuracy: 1.0000 - val_loss: 1.9899 - val_accuracy: 0.6981\n",
            "Epoch 724/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1085 - accuracy: 1.0000 - val_loss: 2.0489 - val_accuracy: 0.7044\n",
            "Epoch 725/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.1106 - accuracy: 0.9965 - val_loss: 2.0599 - val_accuracy: 0.6730\n",
            "Epoch 726/5000\n",
            "7/7 [==============================] - 1s 117ms/step - loss: 0.1111 - accuracy: 1.0000 - val_loss: 2.1620 - val_accuracy: 0.6855\n",
            "Epoch 727/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.1080 - accuracy: 1.0000 - val_loss: 2.3621 - val_accuracy: 0.6918\n",
            "Epoch 728/5000\n",
            "7/7 [==============================] - 1s 122ms/step - loss: 0.1075 - accuracy: 0.9996 - val_loss: 2.2981 - val_accuracy: 0.6792\n",
            "Epoch 729/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1116 - accuracy: 1.0000 - val_loss: 2.2041 - val_accuracy: 0.6855\n",
            "Epoch 730/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1078 - accuracy: 1.0000 - val_loss: 2.1909 - val_accuracy: 0.6918\n",
            "Epoch 731/5000\n",
            "7/7 [==============================] - 1s 121ms/step - loss: 0.1029 - accuracy: 1.0000 - val_loss: 2.1694 - val_accuracy: 0.6792\n",
            "Epoch 732/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.0974 - accuracy: 1.0000 - val_loss: 2.1076 - val_accuracy: 0.6667\n",
            "Epoch 733/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.0941 - accuracy: 1.0000 - val_loss: 2.0714 - val_accuracy: 0.6918\n",
            "Epoch 734/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.0915 - accuracy: 1.0000 - val_loss: 2.0694 - val_accuracy: 0.6981\n",
            "Epoch 735/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.0884 - accuracy: 1.0000 - val_loss: 1.9638 - val_accuracy: 0.7044\n",
            "Epoch 736/5000\n",
            "7/7 [==============================] - 1s 118ms/step - loss: 0.0862 - accuracy: 1.0000 - val_loss: 1.9251 - val_accuracy: 0.6981\n",
            "Epoch 737/5000\n",
            "7/7 [==============================] - 1s 120ms/step - loss: 0.0903 - accuracy: 1.0000 - val_loss: 2.0975 - val_accuracy: 0.7044\n",
            "Epoch 738/5000\n",
            "7/7 [==============================] - 1s 119ms/step - loss: 0.0845 - accuracy: 1.0000 - val_loss: 2.1214 - val_accuracy: 0.6855\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 839
        },
        "id": "o7-S9IbJKFPM",
        "outputId": "9bd6c3f1-f065-4e29-f200-41fcb72be875"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# This function is used to display the history the train/test accuracy/loss\n",
        "# of the Keras training.\n",
        "#\n",
        "#   history - Pass in the history returned from the model.fit(...) method.\n",
        "#\n",
        "def display_training_loss_and_accuracy(history):\n",
        "    \n",
        "    plt.figure(figsize=(20,14))\n",
        "    \n",
        "    # summarize history for accuracy\n",
        "    plt.subplot(2, 1, 1)\n",
        "    plt.plot(history.history['accuracy'])\n",
        "    plt.plot(history.history['val_accuracy'])\n",
        "    plt.title('model accuracy')\n",
        "    plt.ylabel('accuracy')\n",
        "    plt.xlabel('epoch')\n",
        "    plt.legend(['train', 'test'], loc='upper left')\n",
        "    \n",
        "    # summarize history for loss\n",
        "    plt.subplot(2, 1, 2)\n",
        "    plt.plot(history.history['loss'][10:])  #avoid plotting the first 10 epoch due to large values.\n",
        "    plt.plot(history.history['val_loss'][10:])\n",
        "    plt.title('model loss')\n",
        "    plt.ylabel('loss')\n",
        "    plt.xlabel('epoch')\n",
        "    plt.legend(['train', 'test'], loc='upper right')\n",
        "    plt.show()    \n",
        "\n",
        "display_training_loss_and_accuracy(history)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x1008 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "6v1AgGIIKW4X",
        "outputId": "21e577da-f0ad-4ca8-8808-a8ada92e7769"
      },
      "source": [
        "#------------------------------------------------------------------------------------------\n",
        "\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "from sklearn.metrics import confusion_matrix, classification_report\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "from functools import reduce\n",
        " \n",
        "def display_classification_confusion_matrix(keras_model, x_train, y_train, x_test, y_test, labels):\n",
        "    \n",
        "    '''\n",
        "    x_train = []\n",
        "    x_test = []\n",
        "    y_train = []\n",
        "    y_test = []\n",
        "    '''\n",
        "    pred_y_train = []\n",
        "    pred_y_test = []\n",
        "\n",
        "    print (x_train.shape)\n",
        "    pred_y_train = keras_model.predict(x_train)\n",
        "\n",
        "    x_train = np.array(x_train)\n",
        "    y_train = np.array(y_train)\n",
        "    pred_y_train = np.array(pred_y_train)\n",
        "\n",
        "    pred_y_test = keras_model.predict(x_test)\n",
        "    #print(pred_y_test)\n",
        "    x_test = np.array(x_test)\n",
        "    y_test = np.array(y_test)\n",
        "    pred_y_test = np.array(pred_y_test)\n",
        "\n",
        "    #test_generator.on_epoch_end()\n",
        "\n",
        "    plt.figure(figsize=(20,6))  \n",
        "\n",
        "    labels = np.array(labels).argmax(axis=1)\n",
        "    print(labels,np.unique(labels))\n",
        "\n",
        "    # Convert the target labels into the categorical index\n",
        "    #\n",
        "    y_train_index = y_train.argmax(axis=1)\n",
        "    #print(y_train_index)\n",
        "    pred_y_train_index = pred_y_train.argmax(axis=1)\n",
        "    y_test_index = y_test.argmax(axis=1)\n",
        "    pred_y_test_index = pred_y_test.argmax(axis=1)\n",
        "    #print(pred_y_test_index)\n",
        "    \n",
        "    # Print the first Confusion Matrix for the training data\n",
        "    #\n",
        "    cm = confusion_matrix(y_train_index, pred_y_train_index)\n",
        "    #print(cm)\n",
        "    #print(labels)\n",
        "    cm_df = pd.DataFrame(cm, labels, labels)\n",
        "    #print(cm_df)          \n",
        "    plt.subplot(1, 2, 1)\n",
        "    plt.title('Confusion Matrix (Train Data)')\n",
        "    sns.heatmap(cm_df, annot=True)\n",
        "    plt.ylabel('Actual')\n",
        "    plt.xlabel('Predicted')        \n",
        "    \n",
        "    # Print the second Confusion Matrix for the test data\n",
        "    #    \n",
        "    cm = confusion_matrix(y_test_index, pred_y_test_index)\n",
        "    \n",
        "    cm_df = pd.DataFrame(cm, labels, labels)          \n",
        "    plt.subplot(1, 2, 2)\n",
        "    plt.title('Confusion Matrix (Test Data)')\n",
        "    sns.heatmap(cm_df, annot=True)\n",
        "    plt.ylabel('Actual')\n",
        "    plt.xlabel('Predicted')        \n",
        "    \n",
        "    plt.show()\n",
        "    #need to convert target names in classification report to string as the class numbers are integer. \n",
        "    #Confusing as to whether its integer or string.\n",
        "    target_names=list(map(str,labels))\n",
        "\n",
        "    # Finally display the classification reports\n",
        "    #\n",
        "    print (\"Train Data:\")\n",
        "    print (\"--------------------------------------------------------\")\n",
        "    print(classification_report(y_train_index, pred_y_train_index, target_names=target_names))\n",
        "    print (\"\")\n",
        "    print (\"Test Data:\")\n",
        "    print (\"--------------------------------------------------------\")\n",
        "    print(classification_report(y_test_index, pred_y_test_index, target_names=target_names))\n",
        "    \n",
        "\n",
        "#Exclude the O tags from the confusion matrix.\n",
        "\n",
        "display_classification_confusion_matrix(model, X_train, y_train, X_test, y_test, labels)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(633, 95, 36)\n",
            "[43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20\n",
            " 19 18 17 16 15 14 13 12 11 10  9  8  7  6  5  4  3  2  1  0] [ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23\n",
            " 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x432 with 4 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Train Data:\n",
            "--------------------------------------------------------\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "          43       1.00      1.00      1.00        14\n",
            "          42       1.00      1.00      1.00        14\n",
            "          41       1.00      0.80      0.89        15\n",
            "          40       1.00      0.80      0.89        15\n",
            "          39       0.82      1.00      0.90        14\n",
            "          38       1.00      1.00      1.00        14\n",
            "          37       1.00      1.00      1.00        15\n",
            "          36       1.00      1.00      1.00        15\n",
            "          35       1.00      1.00      1.00        14\n",
            "          34       1.00      1.00      1.00        15\n",
            "          33       1.00      1.00      1.00        14\n",
            "          32       1.00      1.00      1.00        14\n",
            "          31       0.88      1.00      0.94        15\n",
            "          30       1.00      0.93      0.96        14\n",
            "          29       1.00      1.00      1.00        14\n",
            "          28       1.00      1.00      1.00        14\n",
            "          27       1.00      1.00      1.00        15\n",
            "          26       1.00      1.00      1.00        14\n",
            "          25       0.83      1.00      0.91        15\n",
            "          24       1.00      1.00      1.00        15\n",
            "          23       1.00      1.00      1.00        14\n",
            "          22       1.00      1.00      1.00        14\n",
            "          21       1.00      1.00      1.00        14\n",
            "          20       0.93      1.00      0.97        14\n",
            "          19       1.00      1.00      1.00        15\n",
            "          18       1.00      1.00      1.00        14\n",
            "          17       1.00      1.00      1.00        15\n",
            "          16       1.00      0.86      0.92        14\n",
            "          15       1.00      1.00      1.00        14\n",
            "          14       1.00      1.00      1.00        14\n",
            "          13       1.00      1.00      1.00        15\n",
            "          12       1.00      1.00      1.00        15\n",
            "          11       1.00      0.93      0.96        14\n",
            "          10       0.88      1.00      0.93        14\n",
            "           9       1.00      1.00      1.00        14\n",
            "           8       1.00      1.00      1.00        15\n",
            "           7       1.00      1.00      1.00        14\n",
            "           6       1.00      1.00      1.00        15\n",
            "           5       1.00      1.00      1.00        14\n",
            "           4       1.00      0.87      0.93        15\n",
            "           3       1.00      1.00      1.00        15\n",
            "           2       1.00      1.00      1.00        14\n",
            "           1       1.00      1.00      1.00        14\n",
            "           0       0.93      1.00      0.97        14\n",
            "\n",
            "    accuracy                           0.98       633\n",
            "   macro avg       0.98      0.98      0.98       633\n",
            "weighted avg       0.98      0.98      0.98       633\n",
            "\n",
            "\n",
            "Test Data:\n",
            "--------------------------------------------------------\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "          43       1.00      1.00      1.00         4\n",
            "          42       0.60      0.75      0.67         4\n",
            "          41       1.00      0.67      0.80         3\n",
            "          40       1.00      0.67      0.80         3\n",
            "          39       0.75      0.75      0.75         4\n",
            "          38       0.67      1.00      0.80         4\n",
            "          37       1.00      0.67      0.80         3\n",
            "          36       1.00      0.67      0.80         3\n",
            "          35       0.67      1.00      0.80         4\n",
            "          34       0.20      0.33      0.25         3\n",
            "          33       1.00      1.00      1.00         4\n",
            "          32       1.00      0.75      0.86         4\n",
            "          31       1.00      0.67      0.80         3\n",
            "          30       0.60      0.75      0.67         4\n",
            "          29       0.67      0.50      0.57         4\n",
            "          28       1.00      0.75      0.86         4\n",
            "          27       1.00      0.67      0.80         3\n",
            "          26       0.75      0.75      0.75         4\n",
            "          25       0.50      1.00      0.67         3\n",
            "          24       0.67      0.67      0.67         3\n",
            "          23       1.00      0.75      0.86         4\n",
            "          22       0.75      0.75      0.75         4\n",
            "          21       0.67      0.50      0.57         4\n",
            "          20       0.50      0.75      0.60         4\n",
            "          19       1.00      1.00      1.00         3\n",
            "          18       1.00      0.75      0.86         4\n",
            "          17       0.50      0.67      0.57         3\n",
            "          16       0.80      1.00      0.89         4\n",
            "          15       1.00      0.50      0.67         4\n",
            "          14       1.00      1.00      1.00         4\n",
            "          13       0.43      1.00      0.60         3\n",
            "          12       0.50      1.00      0.67         3\n",
            "          11       1.00      0.50      0.67         4\n",
            "          10       0.40      0.50      0.44         4\n",
            "           9       1.00      1.00      1.00         4\n",
            "           8       0.50      0.67      0.57         3\n",
            "           7       1.00      0.25      0.40         4\n",
            "           6       1.00      1.00      1.00         3\n",
            "           5       1.00      1.00      1.00         4\n",
            "           4       1.00      0.33      0.50         3\n",
            "           3       0.50      0.67      0.57         3\n",
            "           2       1.00      1.00      1.00         4\n",
            "           1       1.00      0.25      0.40         4\n",
            "           0       0.67      0.50      0.57         4\n",
            "\n",
            "    accuracy                           0.74       159\n",
            "   macro avg       0.80      0.73      0.73       159\n",
            "weighted avg       0.81      0.74      0.74       159\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AoADA2XcN1L5"
      },
      "source": [
        "#### Reshape the data for input to ConvLSTM"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YhDlDtpPJFc-",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "275f0b4b-0559-44df-dc55-ab93bf6fe39b"
      },
      "source": [
        "X_train, X_test, y_train, y_test"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([[[-2.70267967e-01,  2.70021121e-01,  0.00000000e+00, ...,\n",
              "           0.00000000e+00, -1.01092160e-01,  2.92874494e-01],\n",
              "         [-3.53200000e+01,  5.09440000e+01,  0.00000000e+00, ...,\n",
              "           0.00000000e+00, -5.98000000e+00,  5.09670000e+01],\n",
              "         [-2.06441799e-01,  3.40148656e-01,  0.00000000e+00, ...,\n",
              "           0.00000000e+00, -2.41086340e-02,  3.52357912e-01],\n",
              "         ...,\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00, -1.78894530e-01,  2.80302742e-01],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00]],\n",
              " \n",
              "        [[-4.02925978e-02, -3.83824392e-01, -1.12835728e-01, ...,\n",
              "           2.49809993e-01,  1.61458343e-02,  1.24904996e-01],\n",
              "         [-5.13245758e-02,  1.88183319e-01, -1.02608929e-01, ...,\n",
              "           2.52234548e-01,  3.96196765e-03,  1.26117274e-01],\n",
              "         [-8.15919706e-02, -1.22022394e+00, -6.02080385e-02, ...,\n",
              "           2.74691507e-01, -2.13620224e-02,  1.37345754e-01],\n",
              "         ...,\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00]],\n",
              " \n",
              "        [[ 2.41450788e-01,  3.07236240e-01,  0.00000000e+00, ...,\n",
              "           3.51318351e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 2.53712473e-01,  4.34741843e-01,  0.00000000e+00, ...,\n",
              "           4.22880822e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 2.11706270e-01,  4.86301137e-01,  0.00000000e+00, ...,\n",
              "           4.73930967e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         ...,\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00]],\n",
              " \n",
              "        ...,\n",
              " \n",
              "        [[ 1.14524533e+01,  4.45328306e+01,  0.00000000e+00, ...,\n",
              "           2.15360000e+01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 5.71392318e-02, -1.26069428e-03, -4.75567497e-02, ...,\n",
              "          -2.70149264e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 6.20114609e-02, -7.08782964e-01, -3.78028539e-02, ...,\n",
              "           1.23949117e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         ...,\n",
              "         [ 9.65954203e-02, -9.90745355e-01,  6.39106362e-02, ...,\n",
              "          -5.72580925e-03,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 1.03773361e-01, -4.35727763e-01,  5.69103041e-02, ...,\n",
              "          -2.37301549e-04,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 8.03101508e-02,  1.32826092e-01,  3.75745390e-02, ...,\n",
              "          -1.87939195e-02,  0.00000000e+00,  0.00000000e+00]],\n",
              " \n",
              "        [[-1.22300099e-01,  3.27522581e-01,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  8.19500945e-02,  3.41058236e-01],\n",
              "         [-1.19814391e-02,  3.94415525e-01,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  1.55758708e-01,  3.95301907e-01],\n",
              "         [-1.06134044e-01,  3.32118435e-01,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  9.24785807e-02,  3.58165216e-01],\n",
              "         ...,\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00]],\n",
              " \n",
              "        [[-1.64979163e-01,  3.60736532e-01,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  1.59607749e-02,  4.05196499e-01],\n",
              "         [-2.17942027e-01,  3.13064858e-01,  0.00000000e+00, ...,\n",
              "           0.00000000e+00, -5.42943296e-02,  3.40941973e-01],\n",
              "         [-1.53504387e-01,  2.91870846e-01,  0.00000000e+00, ...,\n",
              "           0.00000000e+00, -4.23606937e-02,  3.33190232e-01],\n",
              "         ...,\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00, -2.16468985e-01,  3.04342021e-01],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00, -1.71108934e-01,  3.27943633e-01],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00, -1.84266535e-01,  3.51703325e-01]]]),\n",
              " array([[[-5.22333333e+00,  5.88477879e+01, -1.56666667e+01, ...,\n",
              "           4.17800000e+01,  3.89333333e+00,  2.08900000e+01],\n",
              "         [-6.04042933e-02, -1.37275477e+00, -1.04594046e-01, ...,\n",
              "           2.48928084e-01, -1.02623672e-04,  1.24464042e-01],\n",
              "         [-6.31889943e-02,  1.28336627e-01, -1.18456631e-01, ...,\n",
              "           2.53297539e-01, -1.95609039e-02,  1.26648770e-01],\n",
              "         ...,\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  8.30083970e-02, ...,\n",
              "          -4.28825833e-01, -7.47045879e-02, -3.59258273e-04],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  7.38199135e-02, ...,\n",
              "          -5.29356377e-01, -6.95264515e-02, -7.23222760e-01],\n",
              "         [-1.09091216e-01, -1.19579037e+00,  7.03745278e-02, ...,\n",
              "          -2.38354375e+00, -7.84298437e-02, -8.86689137e-01]],\n",
              " \n",
              "        [[ 3.01188234e-01,  4.64637840e-01,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 1.86774147e-01,  3.49771602e-01,  0.00000000e+00, ...,\n",
              "           3.61097517e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 2.15066104e-01,  3.73590556e-01,  0.00000000e+00, ...,\n",
              "           3.62921664e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         ...,\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           2.92355815e-01, -3.18905528e-02,  3.13625380e-01],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           2.73277309e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           2.75628089e-01,  1.08244385e-04,  2.86587833e-01]],\n",
              " \n",
              "        [[ 6.22391000e+00,  2.27980334e+01,  0.00000000e+00, ...,\n",
              "           1.96193333e+01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 8.38500154e-02, -2.28708203e-01,  4.13462534e-03, ...,\n",
              "           8.18590679e-02,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 3.33508500e-02, -5.01109889e-01, -5.70071907e-02, ...,\n",
              "          -1.08304753e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         ...,\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00]],\n",
              " \n",
              "        ...,\n",
              " \n",
              "        [[ 5.76615624e-02,  7.64001438e-02, -7.08222088e-02, ...,\n",
              "           1.28425629e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 8.33791445e-02, -1.53948027e+00, -8.74816831e-02, ...,\n",
              "          -9.61331916e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 5.13522240e-02, -1.70423162e+00, -1.28084317e-01, ...,\n",
              "          -6.63380920e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         ...,\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  6.50736741e-02, ...,\n",
              "          -8.72019152e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00]],\n",
              " \n",
              "        [[-4.09254961e-02,  1.63480131e-01, -8.17673855e-02, ...,\n",
              "           2.45828879e-01,  4.18033668e-03,  1.22914439e-01],\n",
              "         [-2.36379712e-02,  2.54081836e-01, -1.41948634e-01, ...,\n",
              "           1.68377912e-01,  1.58257626e-02,  1.22335561e-01],\n",
              "         [-4.23776019e-02, -1.32731189e+00, -1.48335146e-01, ...,\n",
              "           2.86142233e-01,  5.44274631e-03,  1.43071117e-01],\n",
              "         ...,\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00]],\n",
              " \n",
              "        [[ 0.00000000e+00,  0.00000000e+00,  0.00000000e+00, ...,\n",
              "           0.00000000e+00,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 6.86718533e-02, -7.19279736e-01,  9.75534987e-05, ...,\n",
              "           1.16919486e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 6.26028162e-02,  2.25351458e-01, -8.35230184e-02, ...,\n",
              "          -5.52675666e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         ...,\n",
              "         [ 4.54081575e-02, -6.53245425e-01, -2.99639920e-02, ...,\n",
              "          -4.10372854e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 7.48746478e-02, -1.13245901e+00,  1.27121643e-04, ...,\n",
              "          -8.56040839e-01,  0.00000000e+00,  0.00000000e+00],\n",
              "         [ 0.00000000e+00,  0.00000000e+00, -1.52170346e-02, ...,\n",
              "          -7.36380405e-01,  0.00000000e+00,  0.00000000e+00]]]),\n",
              " array([[0, 0, 0, ..., 0, 0, 0],\n",
              "        [0, 0, 0, ..., 0, 0, 0],\n",
              "        [0, 0, 0, ..., 0, 0, 0],\n",
              "        ...,\n",
              "        [0, 0, 0, ..., 0, 0, 0],\n",
              "        [0, 0, 0, ..., 0, 0, 0],\n",
              "        [0, 0, 0, ..., 0, 0, 0]]),\n",
              " array([[0, 0, 0, ..., 0, 0, 0],\n",
              "        [1, 0, 0, ..., 0, 0, 0],\n",
              "        [0, 0, 0, ..., 0, 0, 0],\n",
              "        ...,\n",
              "        [0, 0, 0, ..., 0, 0, 0],\n",
              "        [0, 0, 0, ..., 0, 0, 0],\n",
              "        [0, 0, 0, ..., 0, 0, 0]]))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rjCPMAeoOWAH",
        "outputId": "5630cd91-b888-492b-a297-64a965d696de"
      },
      "source": [
        "print(X_train.shape, y_train.shape ,X_test.shape ,y_test.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(633, 60, 36) (633, 44) (159, 60, 36) (159, 44)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GOZ7etuPJP5J"
      },
      "source": [
        "X_train=X_train.reshape(633,60,18,1,2)\n",
        "y_train=y_train.reshape(633,44)\n",
        "X_test=X_test.reshape(159,60,18,1,2)\n",
        "y_test=y_test.reshape(159,44)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tY8iJIRuJoT0",
        "outputId": "fd3c6f78-517d-4fab-a62c-1d8a10492086"
      },
      "source": [
        "print(X_train.shape, X_test.shape, y_train.shape, y_test.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(633, 60, 18, 1, 2) (159, 60, 18, 1, 2) (633, 44) (159, 44)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "twkfcM8wM9K6",
        "outputId": "1be5d453-26c8-4912-8c49-99d6751c0c20"
      },
      "source": [
        "print(X_test[0][0][0][0][0])\n",
        "print(X_train[0][0][0][0][0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "-5.223333333333358\n",
            "-0.2702679674097743\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Bk66132eZvQH"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vGb0kASnR__H"
      },
      "source": [
        "from tensorflow.keras import Model\n",
        "cnn = Sequential()\n",
        "cnn.add(Conv2D(16, (5, 5), input_shape=(None,60, 36)))\n",
        "\n",
        "cnn.add(Conv2D(16, (4, 4)))\n",
        "\n",
        "cnn.add(Flatten()) # Not sure if this if the proper way to do this.\n",
        "\n",
        "rnn = Sequential()\n",
        "\n",
        "rnn = LSTM(64, return_sequences=False, input_shape=(None,60, 36))\n",
        "\n",
        "dense = Sequential()\n",
        "dense.add(Dense(128))\n",
        "dense.add(Dense(64))\n",
        "\n",
        "dense.add(Dense(1)) # Model output\n",
        "\n",
        "adm=optimizers.Adam(lr=0.0001, beta_1=0.9, beta_2=0.999, epsilon=None, decay=0.01, amsgrad=False) \n",
        "main_input = Input(shape=(211, 60, 36)) # Data has been reshaped to (800, 5, 120, 60, 1)\n",
        "\n",
        "model = TimeDistributed(cnn)(main_input) # this should make the cnn 'run' 5 times?\n",
        "model = rnn(model) # combine timedistributed cnn with rnn\n",
        "model = dense(model) # add dense\n",
        "\n",
        "final_model = Model(inputs=main_input, outputs=model)\n",
        "\n",
        "final_model.compile(loss='categorical_crossentropy', optimizer=adm, metrics=['accuracy'])\n",
        "final_model.fit(x=X_train, y=y_train, batch_size=50, epochs=2000, verbose=1, callbacks=callbacks_list, validation_data=(X_test, y_test))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sCR7PXe0MMZU"
      },
      "source": [
        "### ConvLSTM2D\n",
        "\n",
        "network structure able to capture spatiotemporal correlations. In Keras, this is reflected in the ConvLSTM2D class, which computes convolutional operations in both the input and the recurrent transformations.\n",
        "\n",
        "ConvLSTM is a LSTM in which the gates (input to state and state to state transitions) are convolution operations.\n",
        "\n",
        "Input of Keras's ConvLSTM layer is a 5D tensor with Shape:\n",
        "`[samples,time,rows,cols,channels]`\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "https://stackoverflow.com/questions/49603498/convolution2d-lstm-versus-convlstm2d"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "56bBxnMNcOzO",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f5704784-23a6-41e5-e459-dcf83032380d"
      },
      "source": [
        " # We create a layer which take as input movies of shape\n",
        "# (n_frames, width, height, channels) and returns a movie\n",
        "# of identical shape.\n",
        "l2_rate=0.01\n",
        "dropout_rate=0.2\n",
        "spatial_dropout_rate=0.01\n",
        "\n",
        "def create_model():\n",
        "  model = Sequential()\n",
        "  model.add(ConvLSTM2D(filters=16, kernel_size=(3,3),\n",
        "                    input_shape=(60,18,1,2),\n",
        "                    padding='same', activation='elu', return_sequences=True))\n",
        "  model.add(BatchNormalization())\n",
        "\n",
        "  model.add(ConvLSTM2D(filters=16, kernel_size=(3,3),\n",
        "                     padding='same',  activation='elu',  return_sequences=True))\n",
        "  model.add(BatchNormalization())\n",
        "\n",
        "  # model.add(ConvLSTM2D(filters=128, kernel_size=(3,3),\n",
        "  #                   padding='same',  activation='elu',  return_sequences=True))\n",
        "  # model.add(BatchNormalization())\n",
        "\n",
        "  # model.add(ConvLSTM2D(filters=32, kernel_size=(3,3),\n",
        "  #                   padding='same',  activation='elu',  return_sequences=True))\n",
        "  #model.add(Reshape((60,36,72)))\n",
        "  #model.add(SpatialDropout2D(spatial_dropout_rate))\n",
        "  #model.add(BatchNormalization())\n",
        "  #model.add(GlobalAveragePooling2D())\n",
        "\n",
        "  # model.add(Conv2D(filters=44, kernel_size=(3,3),\n",
        "  #               activation='elu',\n",
        "  #               padding='same', data_format='channels_last'))\n",
        "  #model.add(BatchNormalization())\n",
        "  \n",
        "  #model.add(GlobalAveragePooling2D())\n",
        "  \n",
        "  #model.add(GlobalAveragePooling2D())\n",
        "  # model.add(Bidirectional(LSTM(72,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='mul'))\n",
        "  # model.add(BatchNormalization())\n",
        "  # model.add(Dropout(dropout_rate))  \n",
        "  # #model.add(RepeatVector(5))\n",
        "  # model.add(Bidirectional(LSTM(72,return_sequences=True,kernel_regularizer=l2(l2_rate)),merge_mode='ave'))\n",
        "  # model.add(BatchNormalization())\n",
        "  # model.add(Dropout(dropout_rate))\n",
        "\n",
        "  model.add(TimeDistributed(Dense(16)))\n",
        "  model.add(Flatten())\n",
        "  \n",
        "  model.add(Dense(16, activation='elu'))\n",
        "  model.add(Dropout(0.1))\n",
        "  model.add(BatchNormalization())\n",
        "  # Softmax output\n",
        "  model.add(Dense(44, activation='softmax'))\n",
        "\n",
        "  adm=Adam(lr=0.001, beta_1=0.9, beta_2=0.999, epsilon=1e-7, decay=0.01, amsgrad=False, clipnorm=1.0) \n",
        "  model.compile(loss='categorical_crossentropy', optimizer=adm, metrics=['accuracy'])\n",
        "  model.summary()\n",
        "\n",
        "  return model\n",
        "\n",
        "\n",
        "model = create_model()\n",
        "\n",
        "\n",
        "# Create the training folder\n",
        "#\n",
        "training_session_id = datetime.now().strftime(\"%Y-%m-%d %H-%M-%S\")\n",
        "training_session_folder = 'train_%s' % (training_session_id)\n",
        "os.makedirs(training_session_folder, exist_ok=True)\n",
        "\n",
        "# Configure the checkpoint and stop point.\n",
        "# This allows the training to save the best models and also stop the\n",
        "# training early if it detects that there are no improvements after\n",
        "# a long time.\n",
        "#\n",
        "callbacks_list = [\n",
        "    keras.callbacks.ModelCheckpoint(\n",
        "        filepath=training_session_folder + '/model.{epoch:04d}-val_acc-{val_accuracy:4.2f}-loss-{val_loss:4.2f}.h5',\n",
        "        monitor='val_loss', save_best_only=True),\n",
        "    keras.callbacks.EarlyStopping(monitor='val_loss', patience=300, restore_best_weights=True)\n",
        "]\n",
        "\n",
        "\n",
        "# Train our model\n",
        "#\n",
        "history = model.fit(x=X_train, y=y_train,batch_size=100, epochs=2000, verbose=1, callbacks=callbacks_list, validation_data=(X_test, y_test))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_9\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv_lst_m2d_27 (ConvLSTM2D) (None, 60, 18, 1, 16)     10432     \n",
            "_________________________________________________________________\n",
            "batch_normalization_10 (Batc (None, 60, 18, 1, 16)     64        \n",
            "_________________________________________________________________\n",
            "conv_lst_m2d_28 (ConvLSTM2D) (None, 60, 18, 1, 16)     18496     \n",
            "_________________________________________________________________\n",
            "batch_normalization_11 (Batc (None, 60, 18, 1, 16)     64        \n",
            "_________________________________________________________________\n",
            "time_distributed_9 (TimeDist (None, 60, 18, 1, 16)     272       \n",
            "_________________________________________________________________\n",
            "flatten_9 (Flatten)          (None, 17280)             0         \n",
            "_________________________________________________________________\n",
            "dense_24 (Dense)             (None, 16)                276496    \n",
            "_________________________________________________________________\n",
            "dropout_5 (Dropout)          (None, 16)                0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_12 (Batc (None, 16)                64        \n",
            "_________________________________________________________________\n",
            "dense_25 (Dense)             (None, 44)                748       \n",
            "=================================================================\n",
            "Total params: 306,636\n",
            "Trainable params: 306,540\n",
            "Non-trainable params: 96\n",
            "_________________________________________________________________\n",
            "Epoch 1/2000\n",
            "7/7 [==============================] - 10s 667ms/step - loss: 3.9898 - accuracy: 0.0266 - val_loss: 4.5228 - val_accuracy: 0.0629\n",
            "Epoch 2/2000\n",
            "7/7 [==============================] - 4s 529ms/step - loss: 3.6554 - accuracy: 0.0766 - val_loss: 4.0362 - val_accuracy: 0.0440\n",
            "Epoch 3/2000\n",
            "7/7 [==============================] - 4s 558ms/step - loss: 3.5580 - accuracy: 0.1080 - val_loss: 3.8112 - val_accuracy: 0.0377\n",
            "Epoch 4/2000\n",
            "7/7 [==============================] - 4s 542ms/step - loss: 3.4820 - accuracy: 0.1160 - val_loss: 3.7901 - val_accuracy: 0.0440\n",
            "Epoch 5/2000\n",
            "7/7 [==============================] - 4s 643ms/step - loss: 3.3622 - accuracy: 0.1569 - val_loss: 3.7403 - val_accuracy: 0.0377\n",
            "Epoch 6/2000\n",
            "7/7 [==============================] - 5s 646ms/step - loss: 3.2452 - accuracy: 0.2081 - val_loss: 3.7368 - val_accuracy: 0.0440\n",
            "Epoch 7/2000\n",
            "7/7 [==============================] - 4s 619ms/step - loss: 3.2275 - accuracy: 0.1936 - val_loss: 3.7071 - val_accuracy: 0.0440\n",
            "Epoch 8/2000\n",
            "7/7 [==============================] - 4s 552ms/step - loss: 3.1609 - accuracy: 0.2106 - val_loss: 3.7253 - val_accuracy: 0.0377\n",
            "Epoch 9/2000\n",
            "7/7 [==============================] - 4s 526ms/step - loss: 3.1115 - accuracy: 0.2234 - val_loss: 3.7192 - val_accuracy: 0.0440\n",
            "Epoch 10/2000\n",
            "7/7 [==============================] - 4s 604ms/step - loss: 3.0746 - accuracy: 0.2641 - val_loss: 3.7416 - val_accuracy: 0.0440\n",
            "Epoch 11/2000\n",
            "7/7 [==============================] - 4s 564ms/step - loss: 3.0270 - accuracy: 0.2732 - val_loss: 3.7382 - val_accuracy: 0.0314\n",
            "Epoch 12/2000\n",
            "7/7 [==============================] - 4s 548ms/step - loss: 2.9635 - accuracy: 0.2931 - val_loss: 3.7438 - val_accuracy: 0.0252\n",
            "Epoch 13/2000\n",
            "7/7 [==============================] - 4s 556ms/step - loss: 2.9420 - accuracy: 0.3236 - val_loss: 3.7326 - val_accuracy: 0.0377\n",
            "Epoch 14/2000\n",
            "7/7 [==============================] - 4s 569ms/step - loss: 2.9138 - accuracy: 0.3242 - val_loss: 3.7308 - val_accuracy: 0.0503\n",
            "Epoch 15/2000\n",
            "7/7 [==============================] - 4s 591ms/step - loss: 2.8057 - accuracy: 0.3799 - val_loss: 3.7323 - val_accuracy: 0.0566\n",
            "Epoch 16/2000\n",
            "7/7 [==============================] - 4s 609ms/step - loss: 2.8529 - accuracy: 0.3848 - val_loss: 3.7266 - val_accuracy: 0.0503\n",
            "Epoch 17/2000\n",
            "7/7 [==============================] - 4s 549ms/step - loss: 2.7990 - accuracy: 0.3664 - val_loss: 3.7319 - val_accuracy: 0.0629\n",
            "Epoch 18/2000\n",
            "7/7 [==============================] - 4s 602ms/step - loss: 2.6796 - accuracy: 0.4193 - val_loss: 3.7151 - val_accuracy: 0.0629\n",
            "Epoch 19/2000\n",
            "7/7 [==============================] - 4s 621ms/step - loss: 2.6462 - accuracy: 0.4283 - val_loss: 3.6910 - val_accuracy: 0.0692\n",
            "Epoch 20/2000\n",
            "7/7 [==============================] - 4s 572ms/step - loss: 2.6087 - accuracy: 0.4604 - val_loss: 3.6873 - val_accuracy: 0.0755\n",
            "Epoch 21/2000\n",
            "7/7 [==============================] - 4s 520ms/step - loss: 2.5985 - accuracy: 0.4442 - val_loss: 3.6891 - val_accuracy: 0.0566\n",
            "Epoch 22/2000\n",
            "7/7 [==============================] - 4s 616ms/step - loss: 2.5728 - accuracy: 0.4535 - val_loss: 3.6920 - val_accuracy: 0.0566\n",
            "Epoch 23/2000\n",
            "7/7 [==============================] - 4s 618ms/step - loss: 2.5168 - accuracy: 0.4961 - val_loss: 3.6806 - val_accuracy: 0.0566\n",
            "Epoch 24/2000\n",
            "7/7 [==============================] - 4s 530ms/step - loss: 2.5156 - accuracy: 0.4852 - val_loss: 3.6698 - val_accuracy: 0.0692\n",
            "Epoch 25/2000\n",
            "7/7 [==============================] - 4s 526ms/step - loss: 2.4679 - accuracy: 0.5262 - val_loss: 3.6654 - val_accuracy: 0.0566\n",
            "Epoch 26/2000\n",
            "7/7 [==============================] - 4s 530ms/step - loss: 2.3847 - accuracy: 0.5724 - val_loss: 3.6470 - val_accuracy: 0.0818\n",
            "Epoch 27/2000\n",
            "7/7 [==============================] - 4s 585ms/step - loss: 2.3769 - accuracy: 0.5457 - val_loss: 3.6477 - val_accuracy: 0.0818\n",
            "Epoch 28/2000\n",
            "7/7 [==============================] - 4s 550ms/step - loss: 2.3670 - accuracy: 0.5611 - val_loss: 3.6429 - val_accuracy: 0.0755\n",
            "Epoch 29/2000\n",
            "7/7 [==============================] - 4s 525ms/step - loss: 2.3558 - accuracy: 0.5842 - val_loss: 3.6313 - val_accuracy: 0.0881\n",
            "Epoch 30/2000\n",
            "7/7 [==============================] - 4s 521ms/step - loss: 2.3048 - accuracy: 0.5947 - val_loss: 3.6320 - val_accuracy: 0.0943\n",
            "Epoch 31/2000\n",
            "7/7 [==============================] - 4s 535ms/step - loss: 2.2447 - accuracy: 0.6086 - val_loss: 3.6272 - val_accuracy: 0.0943\n",
            "Epoch 32/2000\n",
            "7/7 [==============================] - 4s 572ms/step - loss: 2.2326 - accuracy: 0.5976 - val_loss: 3.6082 - val_accuracy: 0.0943\n",
            "Epoch 33/2000\n",
            "7/7 [==============================] - 4s 517ms/step - loss: 2.2202 - accuracy: 0.6030 - val_loss: 3.5949 - val_accuracy: 0.1006\n",
            "Epoch 34/2000\n",
            "7/7 [==============================] - 4s 557ms/step - loss: 2.1997 - accuracy: 0.6293 - val_loss: 3.5777 - val_accuracy: 0.1132\n",
            "Epoch 35/2000\n",
            "7/7 [==============================] - 4s 538ms/step - loss: 2.2045 - accuracy: 0.6417 - val_loss: 3.5687 - val_accuracy: 0.1132\n",
            "Epoch 36/2000\n",
            "7/7 [==============================] - 4s 531ms/step - loss: 2.1570 - accuracy: 0.6688 - val_loss: 3.5611 - val_accuracy: 0.1132\n",
            "Epoch 37/2000\n",
            "7/7 [==============================] - 4s 544ms/step - loss: 2.1337 - accuracy: 0.6963 - val_loss: 3.5335 - val_accuracy: 0.1195\n",
            "Epoch 38/2000\n",
            "7/7 [==============================] - 4s 625ms/step - loss: 2.1216 - accuracy: 0.6674 - val_loss: 3.5245 - val_accuracy: 0.1321\n",
            "Epoch 39/2000\n",
            "7/7 [==============================] - 4s 544ms/step - loss: 2.0788 - accuracy: 0.6721 - val_loss: 3.5104 - val_accuracy: 0.1321\n",
            "Epoch 40/2000\n",
            "7/7 [==============================] - 4s 537ms/step - loss: 2.0505 - accuracy: 0.6988 - val_loss: 3.5006 - val_accuracy: 0.1384\n",
            "Epoch 41/2000\n",
            "7/7 [==============================] - 4s 523ms/step - loss: 2.0344 - accuracy: 0.6789 - val_loss: 3.4915 - val_accuracy: 0.1384\n",
            "Epoch 42/2000\n",
            "7/7 [==============================] - 4s 528ms/step - loss: 2.0149 - accuracy: 0.6933 - val_loss: 3.4726 - val_accuracy: 0.1509\n",
            "Epoch 43/2000\n",
            "7/7 [==============================] - 4s 573ms/step - loss: 2.0062 - accuracy: 0.7129 - val_loss: 3.4638 - val_accuracy: 0.1698\n",
            "Epoch 44/2000\n",
            "7/7 [==============================] - 4s 594ms/step - loss: 2.0497 - accuracy: 0.6916 - val_loss: 3.4440 - val_accuracy: 0.1698\n",
            "Epoch 45/2000\n",
            "7/7 [==============================] - 4s 522ms/step - loss: 1.9857 - accuracy: 0.7402 - val_loss: 3.4255 - val_accuracy: 0.1635\n",
            "Epoch 46/2000\n",
            "7/7 [==============================] - 4s 581ms/step - loss: 2.0144 - accuracy: 0.6935 - val_loss: 3.4194 - val_accuracy: 0.1698\n",
            "Epoch 47/2000\n",
            "7/7 [==============================] - 4s 535ms/step - loss: 2.0071 - accuracy: 0.6902 - val_loss: 3.4095 - val_accuracy: 0.1824\n",
            "Epoch 48/2000\n",
            "7/7 [==============================] - 4s 528ms/step - loss: 2.0200 - accuracy: 0.6812 - val_loss: 3.4085 - val_accuracy: 0.1887\n",
            "Epoch 49/2000\n",
            "7/7 [==============================] - 4s 552ms/step - loss: 1.9479 - accuracy: 0.7348 - val_loss: 3.4005 - val_accuracy: 0.2264\n",
            "Epoch 50/2000\n",
            "7/7 [==============================] - 4s 544ms/step - loss: 1.9802 - accuracy: 0.7268 - val_loss: 3.3777 - val_accuracy: 0.2138\n",
            "Epoch 51/2000\n",
            "7/7 [==============================] - 4s 590ms/step - loss: 1.9463 - accuracy: 0.7095 - val_loss: 3.3643 - val_accuracy: 0.2075\n",
            "Epoch 52/2000\n",
            "7/7 [==============================] - 4s 524ms/step - loss: 1.8995 - accuracy: 0.7492 - val_loss: 3.3401 - val_accuracy: 0.2327\n",
            "Epoch 53/2000\n",
            "7/7 [==============================] - 4s 531ms/step - loss: 1.9175 - accuracy: 0.7341 - val_loss: 3.3430 - val_accuracy: 0.2390\n",
            "Epoch 54/2000\n",
            "7/7 [==============================] - 4s 576ms/step - loss: 1.9551 - accuracy: 0.7139 - val_loss: 3.3156 - val_accuracy: 0.2704\n",
            "Epoch 55/2000\n",
            "7/7 [==============================] - 4s 540ms/step - loss: 1.8885 - accuracy: 0.7396 - val_loss: 3.2843 - val_accuracy: 0.2893\n",
            "Epoch 56/2000\n",
            "7/7 [==============================] - 4s 533ms/step - loss: 1.8732 - accuracy: 0.7645 - val_loss: 3.2412 - val_accuracy: 0.2956\n",
            "Epoch 57/2000\n",
            "7/7 [==============================] - 4s 597ms/step - loss: 1.8538 - accuracy: 0.7450 - val_loss: 3.2209 - val_accuracy: 0.2893\n",
            "Epoch 58/2000\n",
            "7/7 [==============================] - 4s 540ms/step - loss: 1.8218 - accuracy: 0.7441 - val_loss: 3.1795 - val_accuracy: 0.3208\n",
            "Epoch 59/2000\n",
            "7/7 [==============================] - 4s 575ms/step - loss: 1.8100 - accuracy: 0.7695 - val_loss: 3.1470 - val_accuracy: 0.3270\n",
            "Epoch 60/2000\n",
            "7/7 [==============================] - 4s 608ms/step - loss: 1.8069 - accuracy: 0.7500 - val_loss: 3.1479 - val_accuracy: 0.3270\n",
            "Epoch 61/2000\n",
            "7/7 [==============================] - 4s 564ms/step - loss: 1.7686 - accuracy: 0.7761 - val_loss: 3.1254 - val_accuracy: 0.3396\n",
            "Epoch 62/2000\n",
            "7/7 [==============================] - 4s 532ms/step - loss: 1.7796 - accuracy: 0.7532 - val_loss: 3.0814 - val_accuracy: 0.3396\n",
            "Epoch 63/2000\n",
            "7/7 [==============================] - 4s 543ms/step - loss: 1.8097 - accuracy: 0.7677 - val_loss: 3.0784 - val_accuracy: 0.3711\n",
            "Epoch 64/2000\n",
            "7/7 [==============================] - 4s 567ms/step - loss: 1.8316 - accuracy: 0.7386 - val_loss: 3.0665 - val_accuracy: 0.3774\n",
            "Epoch 65/2000\n",
            "7/7 [==============================] - 4s 541ms/step - loss: 1.7677 - accuracy: 0.7804 - val_loss: 3.0355 - val_accuracy: 0.3774\n",
            "Epoch 66/2000\n",
            "7/7 [==============================] - 4s 600ms/step - loss: 1.7997 - accuracy: 0.7462 - val_loss: 3.0330 - val_accuracy: 0.3774\n",
            "Epoch 67/2000\n",
            "7/7 [==============================] - 4s 630ms/step - loss: 1.7249 - accuracy: 0.7896 - val_loss: 3.0234 - val_accuracy: 0.3899\n",
            "Epoch 68/2000\n",
            "7/7 [==============================] - 4s 586ms/step - loss: 1.7073 - accuracy: 0.7754 - val_loss: 2.9932 - val_accuracy: 0.4214\n",
            "Epoch 69/2000\n",
            "7/7 [==============================] - 4s 579ms/step - loss: 1.7353 - accuracy: 0.7479 - val_loss: 2.9560 - val_accuracy: 0.4465\n",
            "Epoch 70/2000\n",
            "7/7 [==============================] - 4s 573ms/step - loss: 1.7084 - accuracy: 0.7806 - val_loss: 2.9535 - val_accuracy: 0.4528\n",
            "Epoch 71/2000\n",
            "7/7 [==============================] - 4s 536ms/step - loss: 1.7207 - accuracy: 0.7904 - val_loss: 2.9237 - val_accuracy: 0.4214\n",
            "Epoch 72/2000\n",
            "7/7 [==============================] - 5s 643ms/step - loss: 1.7108 - accuracy: 0.7805 - val_loss: 2.8840 - val_accuracy: 0.4340\n",
            "Epoch 73/2000\n",
            "7/7 [==============================] - 4s 593ms/step - loss: 1.7111 - accuracy: 0.8140 - val_loss: 2.8720 - val_accuracy: 0.4528\n",
            "Epoch 74/2000\n",
            "7/7 [==============================] - 4s 568ms/step - loss: 1.7039 - accuracy: 0.7743 - val_loss: 2.8334 - val_accuracy: 0.5094\n",
            "Epoch 75/2000\n",
            "7/7 [==============================] - 4s 581ms/step - loss: 1.6786 - accuracy: 0.8326 - val_loss: 2.8302 - val_accuracy: 0.4654\n",
            "Epoch 76/2000\n",
            "7/7 [==============================] - 4s 545ms/step - loss: 1.7106 - accuracy: 0.7881 - val_loss: 2.7993 - val_accuracy: 0.4465\n",
            "Epoch 77/2000\n",
            "7/7 [==============================] - 4s 609ms/step - loss: 1.6999 - accuracy: 0.7853 - val_loss: 2.7984 - val_accuracy: 0.4591\n",
            "Epoch 78/2000\n",
            "7/7 [==============================] - 4s 551ms/step - loss: 1.6872 - accuracy: 0.7885 - val_loss: 2.7920 - val_accuracy: 0.4906\n",
            "Epoch 79/2000\n",
            "7/7 [==============================] - 4s 591ms/step - loss: 1.6934 - accuracy: 0.7865 - val_loss: 2.7445 - val_accuracy: 0.4969\n",
            "Epoch 80/2000\n",
            "7/7 [==============================] - 4s 570ms/step - loss: 1.6919 - accuracy: 0.7814 - val_loss: 2.7389 - val_accuracy: 0.4843\n",
            "Epoch 81/2000\n",
            "7/7 [==============================] - 4s 575ms/step - loss: 1.6222 - accuracy: 0.8061 - val_loss: 2.7248 - val_accuracy: 0.4780\n",
            "Epoch 82/2000\n",
            "7/7 [==============================] - 4s 573ms/step - loss: 1.6034 - accuracy: 0.8124 - val_loss: 2.6866 - val_accuracy: 0.4969\n",
            "Epoch 83/2000\n",
            "7/7 [==============================] - 4s 595ms/step - loss: 1.6423 - accuracy: 0.8127 - val_loss: 2.6763 - val_accuracy: 0.5157\n",
            "Epoch 84/2000\n",
            "7/7 [==============================] - 4s 563ms/step - loss: 1.6269 - accuracy: 0.8176 - val_loss: 2.6746 - val_accuracy: 0.5346\n",
            "Epoch 85/2000\n",
            "7/7 [==============================] - 4s 537ms/step - loss: 1.5974 - accuracy: 0.8098 - val_loss: 2.6467 - val_accuracy: 0.4969\n",
            "Epoch 86/2000\n",
            "7/7 [==============================] - 4s 537ms/step - loss: 1.6513 - accuracy: 0.7938 - val_loss: 2.5858 - val_accuracy: 0.5094\n",
            "Epoch 87/2000\n",
            "7/7 [==============================] - 4s 527ms/step - loss: 1.6203 - accuracy: 0.8170 - val_loss: 2.5998 - val_accuracy: 0.5283\n",
            "Epoch 88/2000\n",
            "7/7 [==============================] - 4s 560ms/step - loss: 1.5810 - accuracy: 0.8288 - val_loss: 2.6037 - val_accuracy: 0.5094\n",
            "Epoch 89/2000\n",
            "7/7 [==============================] - 4s 626ms/step - loss: 1.5950 - accuracy: 0.8021 - val_loss: 2.5727 - val_accuracy: 0.5409\n",
            "Epoch 90/2000\n",
            "7/7 [==============================] - 4s 628ms/step - loss: 1.5935 - accuracy: 0.8073 - val_loss: 2.5874 - val_accuracy: 0.5283\n",
            "Epoch 91/2000\n",
            "7/7 [==============================] - 4s 528ms/step - loss: 1.5911 - accuracy: 0.8245 - val_loss: 2.5750 - val_accuracy: 0.5409\n",
            "Epoch 92/2000\n",
            "7/7 [==============================] - 4s 628ms/step - loss: 1.6119 - accuracy: 0.7944 - val_loss: 2.5554 - val_accuracy: 0.5472\n",
            "Epoch 93/2000\n",
            "7/7 [==============================] - 4s 594ms/step - loss: 1.5734 - accuracy: 0.8117 - val_loss: 2.5248 - val_accuracy: 0.5535\n",
            "Epoch 94/2000\n",
            "7/7 [==============================] - 4s 635ms/step - loss: 1.5770 - accuracy: 0.8040 - val_loss: 2.5070 - val_accuracy: 0.5597\n",
            "Epoch 95/2000\n",
            "7/7 [==============================] - 4s 588ms/step - loss: 1.5574 - accuracy: 0.7999 - val_loss: 2.4891 - val_accuracy: 0.5472\n",
            "Epoch 96/2000\n",
            "7/7 [==============================] - 4s 605ms/step - loss: 1.6018 - accuracy: 0.8128 - val_loss: 2.4990 - val_accuracy: 0.5660\n",
            "Epoch 97/2000\n",
            "7/7 [==============================] - 4s 576ms/step - loss: 1.5624 - accuracy: 0.8115 - val_loss: 2.5190 - val_accuracy: 0.5535\n",
            "Epoch 98/2000\n",
            "7/7 [==============================] - 4s 610ms/step - loss: 1.5798 - accuracy: 0.7907 - val_loss: 2.5202 - val_accuracy: 0.5472\n",
            "Epoch 99/2000\n",
            "7/7 [==============================] - 4s 561ms/step - loss: 1.5927 - accuracy: 0.8100 - val_loss: 2.5240 - val_accuracy: 0.5409\n",
            "Epoch 100/2000\n",
            "7/7 [==============================] - 4s 557ms/step - loss: 1.4993 - accuracy: 0.8317 - val_loss: 2.4870 - val_accuracy: 0.5535\n",
            "Epoch 101/2000\n",
            "7/7 [==============================] - 4s 634ms/step - loss: 1.5740 - accuracy: 0.8118 - val_loss: 2.4434 - val_accuracy: 0.5660\n",
            "Epoch 102/2000\n",
            "7/7 [==============================] - 4s 551ms/step - loss: 1.5631 - accuracy: 0.7948 - val_loss: 2.4613 - val_accuracy: 0.5660\n",
            "Epoch 103/2000\n",
            "7/7 [==============================] - 4s 588ms/step - loss: 1.5666 - accuracy: 0.7888 - val_loss: 2.4417 - val_accuracy: 0.5912\n",
            "Epoch 104/2000\n",
            "7/7 [==============================] - 4s 551ms/step - loss: 1.5003 - accuracy: 0.8317 - val_loss: 2.4416 - val_accuracy: 0.5786\n",
            "Epoch 105/2000\n",
            "7/7 [==============================] - 4s 596ms/step - loss: 1.5867 - accuracy: 0.8010 - val_loss: 2.5017 - val_accuracy: 0.5535\n",
            "Epoch 106/2000\n",
            "7/7 [==============================] - 4s 537ms/step - loss: 1.5255 - accuracy: 0.8237 - val_loss: 2.4567 - val_accuracy: 0.5723\n",
            "Epoch 107/2000\n",
            "7/7 [==============================] - 4s 547ms/step - loss: 1.4781 - accuracy: 0.8291 - val_loss: 2.4997 - val_accuracy: 0.5660\n",
            "Epoch 108/2000\n",
            "7/7 [==============================] - 4s 628ms/step - loss: 1.5814 - accuracy: 0.8257 - val_loss: 2.4874 - val_accuracy: 0.5660\n",
            "Epoch 109/2000\n",
            "7/7 [==============================] - 4s 594ms/step - loss: 1.5244 - accuracy: 0.8077 - val_loss: 2.5187 - val_accuracy: 0.5535\n",
            "Epoch 110/2000\n",
            "7/7 [==============================] - 4s 557ms/step - loss: 1.5156 - accuracy: 0.8067 - val_loss: 2.4477 - val_accuracy: 0.5535\n",
            "Epoch 111/2000\n",
            "7/7 [==============================] - 4s 564ms/step - loss: 1.5685 - accuracy: 0.8162 - val_loss: 2.4602 - val_accuracy: 0.5660\n",
            "Epoch 112/2000\n",
            "7/7 [==============================] - 4s 559ms/step - loss: 1.5624 - accuracy: 0.7914 - val_loss: 2.4610 - val_accuracy: 0.5849\n",
            "Epoch 113/2000\n",
            "7/7 [==============================] - 4s 629ms/step - loss: 1.4467 - accuracy: 0.8214 - val_loss: 2.4364 - val_accuracy: 0.5786\n",
            "Epoch 114/2000\n",
            "7/7 [==============================] - 4s 561ms/step - loss: 1.5309 - accuracy: 0.8180 - val_loss: 2.4049 - val_accuracy: 0.5723\n",
            "Epoch 115/2000\n",
            "7/7 [==============================] - 4s 536ms/step - loss: 1.5087 - accuracy: 0.8179 - val_loss: 2.3955 - val_accuracy: 0.5472\n",
            "Epoch 116/2000\n",
            "7/7 [==============================] - 4s 562ms/step - loss: 1.4493 - accuracy: 0.8359 - val_loss: 2.3863 - val_accuracy: 0.5786\n",
            "Epoch 117/2000\n",
            "7/7 [==============================] - 4s 573ms/step - loss: 1.4675 - accuracy: 0.8275 - val_loss: 2.3736 - val_accuracy: 0.5912\n",
            "Epoch 118/2000\n",
            "7/7 [==============================] - 4s 615ms/step - loss: 1.4300 - accuracy: 0.8267 - val_loss: 2.3786 - val_accuracy: 0.5597\n",
            "Epoch 119/2000\n",
            "7/7 [==============================] - 4s 593ms/step - loss: 1.4366 - accuracy: 0.8495 - val_loss: 2.3725 - val_accuracy: 0.5660\n",
            "Epoch 120/2000\n",
            "7/7 [==============================] - 4s 568ms/step - loss: 1.3600 - accuracy: 0.8580 - val_loss: 2.3573 - val_accuracy: 0.5597\n",
            "Epoch 121/2000\n",
            "7/7 [==============================] - 4s 546ms/step - loss: 1.4165 - accuracy: 0.8428 - val_loss: 2.3095 - val_accuracy: 0.5849\n",
            "Epoch 122/2000\n",
            "7/7 [==============================] - 4s 535ms/step - loss: 1.4278 - accuracy: 0.8626 - val_loss: 2.3302 - val_accuracy: 0.5660\n",
            "Epoch 123/2000\n",
            "7/7 [==============================] - 4s 557ms/step - loss: 1.4140 - accuracy: 0.8510 - val_loss: 2.3328 - val_accuracy: 0.5660\n",
            "Epoch 124/2000\n",
            "7/7 [==============================] - 4s 601ms/step - loss: 1.4800 - accuracy: 0.8396 - val_loss: 2.3445 - val_accuracy: 0.5975\n",
            "Epoch 125/2000\n",
            "7/7 [==============================] - 5s 636ms/step - loss: 1.3843 - accuracy: 0.8390 - val_loss: 2.3076 - val_accuracy: 0.5849\n",
            "Epoch 126/2000\n",
            "7/7 [==============================] - 4s 610ms/step - loss: 1.3880 - accuracy: 0.8490 - val_loss: 2.3106 - val_accuracy: 0.5849\n",
            "Epoch 127/2000\n",
            "7/7 [==============================] - 4s 574ms/step - loss: 1.4066 - accuracy: 0.8592 - val_loss: 2.3701 - val_accuracy: 0.5723\n",
            "Epoch 128/2000\n",
            "7/7 [==============================] - 4s 566ms/step - loss: 1.3786 - accuracy: 0.8420 - val_loss: 2.3394 - val_accuracy: 0.5535\n",
            "Epoch 129/2000\n",
            "7/7 [==============================] - 4s 592ms/step - loss: 1.3106 - accuracy: 0.8531 - val_loss: 2.3568 - val_accuracy: 0.5723\n",
            "Epoch 130/2000\n",
            "7/7 [==============================] - 4s 540ms/step - loss: 1.3513 - accuracy: 0.8522 - val_loss: 2.3491 - val_accuracy: 0.5660\n",
            "Epoch 131/2000\n",
            "7/7 [==============================] - 4s 617ms/step - loss: 1.3152 - accuracy: 0.8585 - val_loss: 2.3393 - val_accuracy: 0.5597\n",
            "Epoch 132/2000\n",
            "7/7 [==============================] - 4s 540ms/step - loss: 1.3539 - accuracy: 0.8481 - val_loss: 2.3826 - val_accuracy: 0.5660\n",
            "Epoch 133/2000\n",
            "7/7 [==============================] - 4s 623ms/step - loss: 1.4216 - accuracy: 0.8085 - val_loss: 2.3776 - val_accuracy: 0.5660\n",
            "Epoch 134/2000\n",
            "7/7 [==============================] - 4s 585ms/step - loss: 1.3765 - accuracy: 0.8638 - val_loss: 2.3857 - val_accuracy: 0.5786\n",
            "Epoch 135/2000\n",
            "7/7 [==============================] - 4s 535ms/step - loss: 1.3027 - accuracy: 0.8713 - val_loss: 2.3820 - val_accuracy: 0.5849\n",
            "Epoch 136/2000\n",
            "7/7 [==============================] - 4s 525ms/step - loss: 1.3347 - accuracy: 0.8320 - val_loss: 2.3778 - val_accuracy: 0.5535\n",
            "Epoch 137/2000\n",
            "7/7 [==============================] - 4s 554ms/step - loss: 1.3010 - accuracy: 0.8500 - val_loss: 2.3503 - val_accuracy: 0.5472\n",
            "Epoch 138/2000\n",
            "7/7 [==============================] - 4s 607ms/step - loss: 1.2862 - accuracy: 0.8685 - val_loss: 2.3283 - val_accuracy: 0.5660\n",
            "Epoch 139/2000\n",
            "7/7 [==============================] - 4s 546ms/step - loss: 1.3295 - accuracy: 0.8263 - val_loss: 2.3037 - val_accuracy: 0.5786\n",
            "Epoch 140/2000\n",
            "7/7 [==============================] - 4s 556ms/step - loss: 1.3096 - accuracy: 0.8632 - val_loss: 2.2992 - val_accuracy: 0.5975\n",
            "Epoch 141/2000\n",
            "7/7 [==============================] - 4s 538ms/step - loss: 1.3089 - accuracy: 0.8616 - val_loss: 2.2873 - val_accuracy: 0.5912\n",
            "Epoch 142/2000\n",
            "7/7 [==============================] - 4s 542ms/step - loss: 1.2706 - accuracy: 0.8723 - val_loss: 2.2598 - val_accuracy: 0.6101\n",
            "Epoch 143/2000\n",
            "7/7 [==============================] - 4s 539ms/step - loss: 1.2749 - accuracy: 0.8603 - val_loss: 2.2538 - val_accuracy: 0.6101\n",
            "Epoch 144/2000\n",
            "7/7 [==============================] - 4s 535ms/step - loss: 1.3600 - accuracy: 0.8362 - val_loss: 2.2669 - val_accuracy: 0.6038\n",
            "Epoch 145/2000\n",
            "7/7 [==============================] - 4s 583ms/step - loss: 1.4364 - accuracy: 0.8234 - val_loss: 2.2729 - val_accuracy: 0.6101\n",
            "Epoch 146/2000\n",
            "7/7 [==============================] - 4s 535ms/step - loss: 1.3418 - accuracy: 0.8588 - val_loss: 2.2532 - val_accuracy: 0.5975\n",
            "Epoch 147/2000\n",
            "7/7 [==============================] - 4s 538ms/step - loss: 1.2798 - accuracy: 0.8625 - val_loss: 2.2897 - val_accuracy: 0.5975\n",
            "Epoch 148/2000\n",
            "7/7 [==============================] - 4s 620ms/step - loss: 1.3283 - accuracy: 0.8333 - val_loss: 2.2765 - val_accuracy: 0.6038\n",
            "Epoch 149/2000\n",
            "7/7 [==============================] - 4s 545ms/step - loss: 1.3609 - accuracy: 0.8285 - val_loss: 2.2503 - val_accuracy: 0.5975\n",
            "Epoch 150/2000\n",
            "7/7 [==============================] - 4s 596ms/step - loss: 1.3627 - accuracy: 0.8509 - val_loss: 2.2973 - val_accuracy: 0.5849\n",
            "Epoch 151/2000\n",
            "7/7 [==============================] - 4s 555ms/step - loss: 1.2913 - accuracy: 0.8586 - val_loss: 2.3025 - val_accuracy: 0.5723\n",
            "Epoch 152/2000\n",
            "7/7 [==============================] - 4s 589ms/step - loss: 1.3125 - accuracy: 0.8553 - val_loss: 2.2574 - val_accuracy: 0.5786\n",
            "Epoch 153/2000\n",
            "7/7 [==============================] - 4s 538ms/step - loss: 1.2787 - accuracy: 0.8627 - val_loss: 2.2526 - val_accuracy: 0.5975\n",
            "Epoch 154/2000\n",
            "7/7 [==============================] - 4s 597ms/step - loss: 1.4100 - accuracy: 0.8058 - val_loss: 2.2646 - val_accuracy: 0.5912\n",
            "Epoch 155/2000\n",
            "7/7 [==============================] - 4s 631ms/step - loss: 1.2580 - accuracy: 0.8717 - val_loss: 2.2602 - val_accuracy: 0.5849\n",
            "Epoch 156/2000\n",
            "7/7 [==============================] - 4s 537ms/step - loss: 1.3322 - accuracy: 0.8490 - val_loss: 2.2063 - val_accuracy: 0.5849\n",
            "Epoch 157/2000\n",
            "7/7 [==============================] - 4s 571ms/step - loss: 1.3126 - accuracy: 0.8564 - val_loss: 2.2619 - val_accuracy: 0.5597\n",
            "Epoch 158/2000\n",
            "7/7 [==============================] - 4s 588ms/step - loss: 1.2547 - accuracy: 0.8794 - val_loss: 2.1947 - val_accuracy: 0.5723\n",
            "Epoch 159/2000\n",
            "7/7 [==============================] - 4s 584ms/step - loss: 1.2450 - accuracy: 0.8665 - val_loss: 2.2185 - val_accuracy: 0.5786\n",
            "Epoch 160/2000\n",
            "7/7 [==============================] - 4s 556ms/step - loss: 1.2714 - accuracy: 0.8523 - val_loss: 2.2179 - val_accuracy: 0.5975\n",
            "Epoch 161/2000\n",
            "7/7 [==============================] - 4s 530ms/step - loss: 1.2129 - accuracy: 0.8958 - val_loss: 2.2513 - val_accuracy: 0.5912\n",
            "Epoch 162/2000\n",
            "7/7 [==============================] - 4s 541ms/step - loss: 1.2408 - accuracy: 0.8629 - val_loss: 2.2363 - val_accuracy: 0.5849\n",
            "Epoch 163/2000\n",
            "7/7 [==============================] - 4s 551ms/step - loss: 1.2589 - accuracy: 0.8706 - val_loss: 2.2507 - val_accuracy: 0.5786\n",
            "Epoch 164/2000\n",
            "7/7 [==============================] - 4s 610ms/step - loss: 1.2389 - accuracy: 0.8620 - val_loss: 2.2197 - val_accuracy: 0.5786\n",
            "Epoch 165/2000\n",
            "7/7 [==============================] - 4s 582ms/step - loss: 1.2957 - accuracy: 0.8504 - val_loss: 2.2279 - val_accuracy: 0.5786\n",
            "Epoch 166/2000\n",
            "7/7 [==============================] - 4s 552ms/step - loss: 1.3148 - accuracy: 0.8533 - val_loss: 2.2373 - val_accuracy: 0.5912\n",
            "Epoch 167/2000\n",
            "7/7 [==============================] - 4s 541ms/step - loss: 1.3405 - accuracy: 0.8543 - val_loss: 2.2684 - val_accuracy: 0.5660\n",
            "Epoch 168/2000\n",
            "7/7 [==============================] - 4s 549ms/step - loss: 1.2370 - accuracy: 0.8786 - val_loss: 2.2691 - val_accuracy: 0.5723\n",
            "Epoch 169/2000\n",
            "7/7 [==============================] - 4s 531ms/step - loss: 1.3052 - accuracy: 0.8759 - val_loss: 2.2626 - val_accuracy: 0.5849\n",
            "Epoch 170/2000\n",
            "7/7 [==============================] - 4s 603ms/step - loss: 1.3459 - accuracy: 0.8639 - val_loss: 2.3126 - val_accuracy: 0.5912\n",
            "Epoch 171/2000\n",
            "7/7 [==============================] - 4s 592ms/step - loss: 1.3148 - accuracy: 0.8650 - val_loss: 2.3418 - val_accuracy: 0.5535\n",
            "Epoch 172/2000\n",
            "7/7 [==============================] - 4s 546ms/step - loss: 1.2939 - accuracy: 0.8523 - val_loss: 2.3598 - val_accuracy: 0.5535\n",
            "Epoch 173/2000\n",
            "7/7 [==============================] - 4s 568ms/step - loss: 1.2265 - accuracy: 0.8755 - val_loss: 2.3705 - val_accuracy: 0.5409\n",
            "Epoch 174/2000\n",
            "7/7 [==============================] - 4s 572ms/step - loss: 1.3123 - accuracy: 0.8556 - val_loss: 2.3408 - val_accuracy: 0.5660\n",
            "Epoch 175/2000\n",
            "7/7 [==============================] - 4s 578ms/step - loss: 1.2851 - accuracy: 0.8529 - val_loss: 2.3193 - val_accuracy: 0.5912\n",
            "Epoch 176/2000\n",
            "7/7 [==============================] - 4s 559ms/step - loss: 1.2996 - accuracy: 0.8439 - val_loss: 2.4316 - val_accuracy: 0.5660\n",
            "Epoch 177/2000\n",
            "7/7 [==============================] - 4s 531ms/step - loss: 1.3523 - accuracy: 0.8444 - val_loss: 2.4361 - val_accuracy: 0.5849\n",
            "Epoch 178/2000\n",
            "7/7 [==============================] - 4s 548ms/step - loss: 1.3982 - accuracy: 0.8435 - val_loss: 2.4127 - val_accuracy: 0.5660\n",
            "Epoch 179/2000\n",
            "7/7 [==============================] - 4s 583ms/step - loss: 1.2617 - accuracy: 0.8596 - val_loss: 2.4033 - val_accuracy: 0.5597\n",
            "Epoch 180/2000\n",
            "7/7 [==============================] - 4s 547ms/step - loss: 1.3922 - accuracy: 0.8328 - val_loss: 2.4209 - val_accuracy: 0.5346\n",
            "Epoch 181/2000\n",
            "7/7 [==============================] - 4s 571ms/step - loss: 1.2988 - accuracy: 0.8363 - val_loss: 2.4986 - val_accuracy: 0.5220\n",
            "Epoch 182/2000\n",
            "7/7 [==============================] - 4s 583ms/step - loss: 1.2727 - accuracy: 0.8714 - val_loss: 2.3893 - val_accuracy: 0.5535\n",
            "Epoch 183/2000\n",
            "7/7 [==============================] - 4s 548ms/step - loss: 1.4795 - accuracy: 0.8480 - val_loss: 2.4642 - val_accuracy: 0.5472\n",
            "Epoch 184/2000\n",
            "7/7 [==============================] - 4s 564ms/step - loss: 1.2717 - accuracy: 0.8578 - val_loss: 2.4827 - val_accuracy: 0.5157\n",
            "Epoch 185/2000\n",
            "7/7 [==============================] - 4s 530ms/step - loss: 1.2924 - accuracy: 0.8732 - val_loss: 2.4355 - val_accuracy: 0.5472\n",
            "Epoch 186/2000\n",
            "7/7 [==============================] - 4s 576ms/step - loss: 1.2956 - accuracy: 0.8430 - val_loss: 2.4268 - val_accuracy: 0.5220\n",
            "Epoch 187/2000\n",
            "7/7 [==============================] - 4s 527ms/step - loss: 1.4328 - accuracy: 0.8108 - val_loss: 2.4387 - val_accuracy: 0.5346\n",
            "Epoch 188/2000\n",
            "7/7 [==============================] - 4s 532ms/step - loss: 1.4767 - accuracy: 0.8177 - val_loss: 2.5846 - val_accuracy: 0.4843\n",
            "Epoch 189/2000\n",
            "7/7 [==============================] - 4s 552ms/step - loss: 1.4559 - accuracy: 0.8324 - val_loss: 2.4930 - val_accuracy: 0.4780\n",
            "Epoch 190/2000\n",
            "7/7 [==============================] - 4s 576ms/step - loss: 1.4157 - accuracy: 0.8452 - val_loss: 2.7714 - val_accuracy: 0.4528\n",
            "Epoch 191/2000\n",
            "7/7 [==============================] - 4s 537ms/step - loss: 1.6320 - accuracy: 0.7941 - val_loss: 2.7363 - val_accuracy: 0.3774\n",
            "Epoch 192/2000\n",
            "7/7 [==============================] - 4s 562ms/step - loss: 1.9170 - accuracy: 0.7535 - val_loss: 2.8245 - val_accuracy: 0.3522\n",
            "Epoch 193/2000\n",
            "7/7 [==============================] - 4s 539ms/step - loss: 1.7047 - accuracy: 0.7781 - val_loss: 2.8328 - val_accuracy: 0.3522\n",
            "Epoch 194/2000\n",
            "7/7 [==============================] - 4s 595ms/step - loss: 1.4489 - accuracy: 0.8052 - val_loss: 2.7619 - val_accuracy: 0.3711\n",
            "Epoch 195/2000\n",
            "7/7 [==============================] - 4s 543ms/step - loss: 1.5418 - accuracy: 0.8148 - val_loss: 3.0497 - val_accuracy: 0.2767\n",
            "Epoch 196/2000\n",
            "7/7 [==============================] - 4s 586ms/step - loss: 1.8385 - accuracy: 0.7508 - val_loss: 3.0683 - val_accuracy: 0.2767\n",
            "Epoch 197/2000\n",
            "7/7 [==============================] - 4s 530ms/step - loss: 1.6359 - accuracy: 0.7918 - val_loss: 3.0459 - val_accuracy: 0.2516\n",
            "Epoch 198/2000\n",
            "7/7 [==============================] - 4s 554ms/step - loss: 1.8574 - accuracy: 0.7689 - val_loss: 3.1476 - val_accuracy: 0.2327\n",
            "Epoch 199/2000\n",
            "7/7 [==============================] - 4s 601ms/step - loss: 1.8503 - accuracy: 0.7312 - val_loss: 3.2420 - val_accuracy: 0.2013\n",
            "Epoch 200/2000\n",
            "7/7 [==============================] - 4s 566ms/step - loss: 1.7414 - accuracy: 0.7563 - val_loss: 3.2060 - val_accuracy: 0.2201\n",
            "Epoch 201/2000\n",
            "7/7 [==============================] - 4s 580ms/step - loss: 1.9861 - accuracy: 0.6922 - val_loss: 3.2638 - val_accuracy: 0.1887\n",
            "Epoch 202/2000\n",
            "7/7 [==============================] - 4s 535ms/step - loss: 1.5183 - accuracy: 0.8322 - val_loss: 3.2563 - val_accuracy: 0.1887\n",
            "Epoch 203/2000\n",
            "7/7 [==============================] - 4s 527ms/step - loss: 1.6291 - accuracy: 0.8115 - val_loss: 3.2158 - val_accuracy: 0.2075\n",
            "Epoch 204/2000\n",
            "7/7 [==============================] - 4s 595ms/step - loss: 1.4798 - accuracy: 0.8166 - val_loss: 3.1910 - val_accuracy: 0.2453\n",
            "Epoch 205/2000\n",
            "7/7 [==============================] - 4s 566ms/step - loss: 1.4975 - accuracy: 0.8042 - val_loss: 3.1846 - val_accuracy: 0.2390\n",
            "Epoch 206/2000\n",
            "7/7 [==============================] - 4s 554ms/step - loss: 1.5771 - accuracy: 0.8273 - val_loss: 3.1218 - val_accuracy: 0.2767\n",
            "Epoch 207/2000\n",
            "7/7 [==============================] - 4s 541ms/step - loss: 1.3353 - accuracy: 0.8437 - val_loss: 3.1048 - val_accuracy: 0.2767\n",
            "Epoch 208/2000\n",
            "7/7 [==============================] - 4s 628ms/step - loss: 1.6309 - accuracy: 0.8143 - val_loss: 3.0952 - val_accuracy: 0.2767\n",
            "Epoch 209/2000\n",
            "7/7 [==============================] - 4s 572ms/step - loss: 1.7597 - accuracy: 0.7371 - val_loss: 3.1121 - val_accuracy: 0.2704\n",
            "Epoch 210/2000\n",
            "7/7 [==============================] - 4s 584ms/step - loss: 1.6765 - accuracy: 0.7895 - val_loss: 3.0625 - val_accuracy: 0.3082\n",
            "Epoch 211/2000\n",
            "7/7 [==============================] - 4s 548ms/step - loss: 1.7268 - accuracy: 0.7654 - val_loss: 3.0933 - val_accuracy: 0.3019\n",
            "Epoch 212/2000\n",
            "7/7 [==============================] - 4s 525ms/step - loss: 1.8488 - accuracy: 0.7773 - val_loss: 3.1414 - val_accuracy: 0.2830\n",
            "Epoch 213/2000\n",
            "7/7 [==============================] - 4s 595ms/step - loss: 1.6989 - accuracy: 0.7871 - val_loss: 3.1369 - val_accuracy: 0.2893\n",
            "Epoch 214/2000\n",
            "7/7 [==============================] - 4s 589ms/step - loss: 1.6016 - accuracy: 0.7982 - val_loss: 3.1329 - val_accuracy: 0.2830\n",
            "Epoch 215/2000\n",
            "7/7 [==============================] - 4s 537ms/step - loss: 1.7549 - accuracy: 0.7597 - val_loss: 3.2055 - val_accuracy: 0.2390\n",
            "Epoch 216/2000\n",
            "7/7 [==============================] - 4s 564ms/step - loss: 2.1112 - accuracy: 0.6876 - val_loss: 3.3196 - val_accuracy: 0.2138\n",
            "Epoch 217/2000\n",
            "7/7 [==============================] - 4s 557ms/step - loss: 1.8499 - accuracy: 0.7074 - val_loss: 3.4326 - val_accuracy: 0.1635\n",
            "Epoch 218/2000\n",
            "7/7 [==============================] - 4s 555ms/step - loss: 1.9941 - accuracy: 0.6685 - val_loss: 3.4564 - val_accuracy: 0.1509\n",
            "Epoch 219/2000\n",
            "7/7 [==============================] - 4s 604ms/step - loss: 2.0614 - accuracy: 0.6734 - val_loss: 3.5408 - val_accuracy: 0.1321\n",
            "Epoch 220/2000\n",
            "7/7 [==============================] - 4s 583ms/step - loss: 2.1490 - accuracy: 0.6734 - val_loss: 3.5629 - val_accuracy: 0.1195\n",
            "Epoch 221/2000\n",
            "7/7 [==============================] - 4s 557ms/step - loss: 2.2748 - accuracy: 0.5989 - val_loss: 3.7039 - val_accuracy: 0.1006\n",
            "Epoch 222/2000\n",
            "7/7 [==============================] - 4s 552ms/step - loss: 1.9992 - accuracy: 0.7043 - val_loss: 3.7676 - val_accuracy: 0.0881\n",
            "Epoch 223/2000\n",
            "7/7 [==============================] - 4s 574ms/step - loss: 2.4397 - accuracy: 0.6243 - val_loss: 3.7994 - val_accuracy: 0.0818\n",
            "Epoch 224/2000\n",
            "7/7 [==============================] - 4s 553ms/step - loss: 2.0627 - accuracy: 0.6626 - val_loss: 3.8164 - val_accuracy: 0.0818\n",
            "Epoch 225/2000\n",
            "7/7 [==============================] - 4s 576ms/step - loss: 1.9625 - accuracy: 0.7226 - val_loss: 3.7449 - val_accuracy: 0.0943\n",
            "Epoch 226/2000\n",
            "7/7 [==============================] - 4s 566ms/step - loss: 1.8023 - accuracy: 0.7584 - val_loss: 3.8584 - val_accuracy: 0.0881\n",
            "Epoch 227/2000\n",
            "7/7 [==============================] - 4s 615ms/step - loss: 2.0879 - accuracy: 0.6728 - val_loss: 3.8289 - val_accuracy: 0.0943\n",
            "Epoch 228/2000\n",
            "7/7 [==============================] - 4s 591ms/step - loss: 1.8127 - accuracy: 0.7230 - val_loss: 3.8711 - val_accuracy: 0.0818\n",
            "Epoch 229/2000\n",
            "7/7 [==============================] - 4s 589ms/step - loss: 2.2108 - accuracy: 0.6274 - val_loss: 3.8449 - val_accuracy: 0.0818\n",
            "Epoch 230/2000\n",
            "7/7 [==============================] - 4s 618ms/step - loss: 2.2040 - accuracy: 0.6181 - val_loss: 3.8747 - val_accuracy: 0.0755\n",
            "Epoch 231/2000\n",
            "7/7 [==============================] - 4s 533ms/step - loss: 2.1869 - accuracy: 0.6578 - val_loss: 3.8920 - val_accuracy: 0.0755\n",
            "Epoch 232/2000\n",
            "7/7 [==============================] - 4s 612ms/step - loss: 2.0145 - accuracy: 0.7092 - val_loss: 3.8103 - val_accuracy: 0.0818\n",
            "Epoch 233/2000\n",
            "7/7 [==============================] - 4s 591ms/step - loss: 2.1679 - accuracy: 0.6639 - val_loss: 3.7864 - val_accuracy: 0.0755\n",
            "Epoch 234/2000\n",
            "7/7 [==============================] - 4s 532ms/step - loss: 2.0549 - accuracy: 0.6699 - val_loss: 3.8215 - val_accuracy: 0.0755\n",
            "Epoch 235/2000\n",
            "7/7 [==============================] - 5s 648ms/step - loss: 2.2351 - accuracy: 0.6189 - val_loss: 3.8119 - val_accuracy: 0.0818\n",
            "Epoch 236/2000\n",
            "7/7 [==============================] - 4s 530ms/step - loss: 2.1180 - accuracy: 0.6461 - val_loss: 3.7705 - val_accuracy: 0.0943\n",
            "Epoch 237/2000\n",
            "7/7 [==============================] - 4s 615ms/step - loss: 1.9322 - accuracy: 0.6928 - val_loss: 3.7099 - val_accuracy: 0.0881\n",
            "Epoch 238/2000\n",
            "7/7 [==============================] - 4s 590ms/step - loss: 1.9145 - accuracy: 0.6981 - val_loss: 3.6968 - val_accuracy: 0.0881\n",
            "Epoch 239/2000\n",
            "7/7 [==============================] - 4s 570ms/step - loss: 1.8272 - accuracy: 0.7368 - val_loss: 3.6727 - val_accuracy: 0.0881\n",
            "Epoch 240/2000\n",
            "7/7 [==============================] - 4s 554ms/step - loss: 1.8316 - accuracy: 0.7170 - val_loss: 3.6365 - val_accuracy: 0.0881\n",
            "Epoch 241/2000\n",
            "7/7 [==============================] - 4s 627ms/step - loss: 1.8594 - accuracy: 0.7168 - val_loss: 3.7071 - val_accuracy: 0.0881\n",
            "Epoch 242/2000\n",
            "7/7 [==============================] - 4s 558ms/step - loss: 1.9799 - accuracy: 0.6980 - val_loss: 3.7109 - val_accuracy: 0.0818\n",
            "Epoch 243/2000\n",
            "7/7 [==============================] - 4s 586ms/step - loss: 1.7326 - accuracy: 0.7609 - val_loss: 3.6754 - val_accuracy: 0.0943\n",
            "Epoch 244/2000\n",
            "7/7 [==============================] - 4s 645ms/step - loss: 1.5693 - accuracy: 0.7582 - val_loss: 3.6462 - val_accuracy: 0.0943\n",
            "Epoch 245/2000\n",
            "7/7 [==============================] - 4s 565ms/step - loss: 1.6846 - accuracy: 0.7405 - val_loss: 3.6756 - val_accuracy: 0.0881\n",
            "Epoch 246/2000\n",
            "7/7 [==============================] - 4s 566ms/step - loss: 1.5088 - accuracy: 0.7832 - val_loss: 3.6702 - val_accuracy: 0.0881\n",
            "Epoch 247/2000\n",
            "7/7 [==============================] - 4s 543ms/step - loss: 1.6707 - accuracy: 0.7816 - val_loss: 3.6561 - val_accuracy: 0.1132\n",
            "Epoch 248/2000\n",
            "7/7 [==============================] - 4s 586ms/step - loss: 1.6589 - accuracy: 0.7667 - val_loss: 3.6814 - val_accuracy: 0.0943\n",
            "Epoch 249/2000\n",
            "7/7 [==============================] - 4s 577ms/step - loss: 1.6238 - accuracy: 0.7754 - val_loss: 3.6511 - val_accuracy: 0.1006\n",
            "Epoch 250/2000\n",
            "7/7 [==============================] - 4s 537ms/step - loss: 1.5346 - accuracy: 0.8103 - val_loss: 3.6625 - val_accuracy: 0.1006\n",
            "Epoch 251/2000\n",
            "7/7 [==============================] - 4s 603ms/step - loss: 1.7508 - accuracy: 0.7711 - val_loss: 3.6859 - val_accuracy: 0.1006\n",
            "Epoch 252/2000\n",
            "7/7 [==============================] - 4s 540ms/step - loss: 1.5583 - accuracy: 0.7688 - val_loss: 3.7031 - val_accuracy: 0.1069\n",
            "Epoch 253/2000\n",
            "7/7 [==============================] - 4s 581ms/step - loss: 1.6602 - accuracy: 0.7672 - val_loss: 3.6670 - val_accuracy: 0.1132\n",
            "Epoch 254/2000\n",
            "7/7 [==============================] - 4s 565ms/step - loss: 1.6928 - accuracy: 0.7681 - val_loss: 3.6906 - val_accuracy: 0.1132\n",
            "Epoch 255/2000\n",
            "7/7 [==============================] - 4s 547ms/step - loss: 1.7392 - accuracy: 0.7548 - val_loss: 3.6629 - val_accuracy: 0.1006\n",
            "Epoch 256/2000\n",
            "7/7 [==============================] - 4s 549ms/step - loss: 1.6870 - accuracy: 0.7672 - val_loss: 3.6636 - val_accuracy: 0.1132\n",
            "Epoch 257/2000\n",
            "7/7 [==============================] - 4s 557ms/step - loss: 1.5875 - accuracy: 0.7676 - val_loss: 3.6487 - val_accuracy: 0.1006\n",
            "Epoch 258/2000\n",
            "7/7 [==============================] - 4s 626ms/step - loss: 1.6782 - accuracy: 0.7447 - val_loss: 3.6777 - val_accuracy: 0.1069\n",
            "Epoch 259/2000\n",
            "7/7 [==============================] - 4s 628ms/step - loss: 1.5994 - accuracy: 0.7517 - val_loss: 3.6144 - val_accuracy: 0.1132\n",
            "Epoch 260/2000\n",
            "7/7 [==============================] - 4s 576ms/step - loss: 1.6467 - accuracy: 0.7483 - val_loss: 3.6210 - val_accuracy: 0.1132\n",
            "Epoch 261/2000\n",
            "7/7 [==============================] - 4s 560ms/step - loss: 1.6552 - accuracy: 0.7951 - val_loss: 3.5708 - val_accuracy: 0.1258\n",
            "Epoch 262/2000\n",
            "7/7 [==============================] - 4s 563ms/step - loss: 1.6948 - accuracy: 0.7522 - val_loss: 3.5700 - val_accuracy: 0.1258\n",
            "Epoch 263/2000\n",
            "7/7 [==============================] - 4s 597ms/step - loss: 1.5863 - accuracy: 0.7883 - val_loss: 3.5611 - val_accuracy: 0.1321\n",
            "Epoch 264/2000\n",
            "7/7 [==============================] - 4s 611ms/step - loss: 1.7705 - accuracy: 0.7025 - val_loss: 3.5445 - val_accuracy: 0.1321\n",
            "Epoch 265/2000\n",
            "7/7 [==============================] - 4s 541ms/step - loss: 1.5986 - accuracy: 0.7665 - val_loss: 3.5202 - val_accuracy: 0.1195\n",
            "Epoch 266/2000\n",
            "7/7 [==============================] - 4s 607ms/step - loss: 1.6653 - accuracy: 0.7536 - val_loss: 3.4876 - val_accuracy: 0.1195\n",
            "Epoch 267/2000\n",
            "7/7 [==============================] - 4s 568ms/step - loss: 1.7122 - accuracy: 0.7635 - val_loss: 3.4540 - val_accuracy: 0.1258\n",
            "Epoch 268/2000\n",
            "7/7 [==============================] - 4s 583ms/step - loss: 1.6028 - accuracy: 0.7723 - val_loss: 3.4587 - val_accuracy: 0.1195\n",
            "Epoch 269/2000\n",
            "7/7 [==============================] - 4s 604ms/step - loss: 1.6277 - accuracy: 0.7552 - val_loss: 3.4351 - val_accuracy: 0.1258\n",
            "Epoch 270/2000\n",
            "7/7 [==============================] - 4s 586ms/step - loss: 1.4933 - accuracy: 0.7992 - val_loss: 3.5140 - val_accuracy: 0.1258\n",
            "Epoch 271/2000\n",
            "7/7 [==============================] - 4s 595ms/step - loss: 1.4759 - accuracy: 0.7944 - val_loss: 3.4721 - val_accuracy: 0.1195\n",
            "Epoch 272/2000\n",
            "7/7 [==============================] - 4s 546ms/step - loss: 1.6000 - accuracy: 0.7945 - val_loss: 3.4804 - val_accuracy: 0.1258\n",
            "Epoch 273/2000\n",
            "7/7 [==============================] - 4s 540ms/step - loss: 1.5088 - accuracy: 0.7806 - val_loss: 3.4827 - val_accuracy: 0.1258\n",
            "Epoch 274/2000\n",
            "7/7 [==============================] - 4s 622ms/step - loss: 1.5165 - accuracy: 0.8125 - val_loss: 3.4527 - val_accuracy: 0.1195\n",
            "Epoch 275/2000\n",
            "7/7 [==============================] - 4s 540ms/step - loss: 1.6002 - accuracy: 0.7782 - val_loss: 3.4405 - val_accuracy: 0.1509\n",
            "Epoch 276/2000\n",
            "7/7 [==============================] - 4s 557ms/step - loss: 1.7073 - accuracy: 0.7683 - val_loss: 3.4657 - val_accuracy: 0.1384\n",
            "Epoch 277/2000\n",
            "7/7 [==============================] - 4s 552ms/step - loss: 1.5909 - accuracy: 0.7799 - val_loss: 3.5235 - val_accuracy: 0.1069\n",
            "Epoch 278/2000\n",
            "7/7 [==============================] - 4s 600ms/step - loss: 1.4812 - accuracy: 0.8012 - val_loss: 3.4460 - val_accuracy: 0.1321\n",
            "Epoch 279/2000\n",
            "7/7 [==============================] - 4s 646ms/step - loss: 1.3795 - accuracy: 0.8170 - val_loss: 3.4408 - val_accuracy: 0.1258\n",
            "Epoch 280/2000\n",
            "7/7 [==============================] - 4s 567ms/step - loss: 1.6118 - accuracy: 0.7671 - val_loss: 3.4509 - val_accuracy: 0.1321\n",
            "Epoch 281/2000\n",
            "7/7 [==============================] - 4s 596ms/step - loss: 1.4013 - accuracy: 0.8246 - val_loss: 3.4777 - val_accuracy: 0.1321\n",
            "Epoch 282/2000\n",
            "7/7 [==============================] - 4s 587ms/step - loss: 1.3896 - accuracy: 0.8324 - val_loss: 3.5388 - val_accuracy: 0.1321\n",
            "Epoch 283/2000\n",
            "7/7 [==============================] - 4s 584ms/step - loss: 1.3388 - accuracy: 0.8184 - val_loss: 3.5075 - val_accuracy: 0.1509\n",
            "Epoch 284/2000\n",
            "7/7 [==============================] - 4s 600ms/step - loss: 1.3654 - accuracy: 0.8338 - val_loss: 3.5031 - val_accuracy: 0.1258\n",
            "Epoch 285/2000\n",
            "7/7 [==============================] - 4s 545ms/step - loss: 1.7426 - accuracy: 0.7415 - val_loss: 3.4252 - val_accuracy: 0.1447\n",
            "Epoch 286/2000\n",
            "7/7 [==============================] - 4s 542ms/step - loss: 1.4401 - accuracy: 0.8268 - val_loss: 3.3921 - val_accuracy: 0.1635\n",
            "Epoch 287/2000\n",
            "7/7 [==============================] - 4s 599ms/step - loss: 1.5121 - accuracy: 0.7870 - val_loss: 3.3701 - val_accuracy: 0.1572\n",
            "Epoch 288/2000\n",
            "7/7 [==============================] - 4s 571ms/step - loss: 1.3862 - accuracy: 0.8275 - val_loss: 3.3736 - val_accuracy: 0.1572\n",
            "Epoch 289/2000\n",
            "7/7 [==============================] - 4s 591ms/step - loss: 1.3177 - accuracy: 0.8625 - val_loss: 3.3543 - val_accuracy: 0.1572\n",
            "Epoch 290/2000\n",
            "7/7 [==============================] - 4s 566ms/step - loss: 1.2645 - accuracy: 0.8406 - val_loss: 3.3575 - val_accuracy: 0.1698\n",
            "Epoch 291/2000\n",
            "7/7 [==============================] - 4s 589ms/step - loss: 1.4152 - accuracy: 0.8064 - val_loss: 3.3516 - val_accuracy: 0.1572\n",
            "Epoch 292/2000\n",
            "7/7 [==============================] - 4s 552ms/step - loss: 1.6359 - accuracy: 0.7414 - val_loss: 3.3851 - val_accuracy: 0.1509\n",
            "Epoch 293/2000\n",
            "7/7 [==============================] - 4s 539ms/step - loss: 1.5822 - accuracy: 0.7603 - val_loss: 3.3899 - val_accuracy: 0.1509\n",
            "Epoch 294/2000\n",
            "7/7 [==============================] - 4s 618ms/step - loss: 1.3783 - accuracy: 0.8082 - val_loss: 3.3851 - val_accuracy: 0.1572\n",
            "Epoch 295/2000\n",
            "7/7 [==============================] - 4s 628ms/step - loss: 1.4870 - accuracy: 0.8262 - val_loss: 3.3559 - val_accuracy: 0.1572\n",
            "Epoch 296/2000\n",
            "7/7 [==============================] - 4s 600ms/step - loss: 1.5362 - accuracy: 0.7902 - val_loss: 3.3158 - val_accuracy: 0.1447\n",
            "Epoch 297/2000\n",
            "7/7 [==============================] - 4s 554ms/step - loss: 1.5046 - accuracy: 0.7867 - val_loss: 3.3287 - val_accuracy: 0.1447\n",
            "Epoch 298/2000\n",
            "7/7 [==============================] - 4s 632ms/step - loss: 1.3710 - accuracy: 0.8261 - val_loss: 3.2913 - val_accuracy: 0.1509\n",
            "Epoch 299/2000\n",
            "7/7 [==============================] - 4s 556ms/step - loss: 1.4904 - accuracy: 0.7789 - val_loss: 3.3225 - val_accuracy: 0.1635\n",
            "Epoch 300/2000\n",
            "7/7 [==============================] - 4s 587ms/step - loss: 1.4158 - accuracy: 0.8137 - val_loss: 3.3950 - val_accuracy: 0.1509\n",
            "Epoch 301/2000\n",
            "7/7 [==============================] - 4s 584ms/step - loss: 1.6582 - accuracy: 0.7668 - val_loss: 3.4215 - val_accuracy: 0.1509\n",
            "Epoch 302/2000\n",
            "7/7 [==============================] - 4s 571ms/step - loss: 1.6821 - accuracy: 0.7313 - val_loss: 3.4244 - val_accuracy: 0.1447\n",
            "Epoch 303/2000\n",
            "7/7 [==============================] - 4s 552ms/step - loss: 1.6776 - accuracy: 0.7720 - val_loss: 3.3867 - val_accuracy: 0.1447\n",
            "Epoch 304/2000\n",
            "7/7 [==============================] - 4s 580ms/step - loss: 1.8076 - accuracy: 0.7318 - val_loss: 3.3945 - val_accuracy: 0.1761\n",
            "Epoch 305/2000\n",
            "7/7 [==============================] - 4s 626ms/step - loss: 1.5933 - accuracy: 0.7735 - val_loss: 3.4225 - val_accuracy: 0.1509\n",
            "Epoch 306/2000\n",
            "7/7 [==============================] - 4s 577ms/step - loss: 1.4133 - accuracy: 0.8129 - val_loss: 3.4145 - val_accuracy: 0.1635\n",
            "Epoch 307/2000\n",
            "7/7 [==============================] - 4s 607ms/step - loss: 1.4477 - accuracy: 0.7942 - val_loss: 3.5042 - val_accuracy: 0.1195\n",
            "Epoch 308/2000\n",
            "7/7 [==============================] - 4s 629ms/step - loss: 1.4419 - accuracy: 0.8003 - val_loss: 3.5454 - val_accuracy: 0.1132\n",
            "Epoch 309/2000\n",
            "7/7 [==============================] - 4s 576ms/step - loss: 1.5060 - accuracy: 0.8093 - val_loss: 3.5310 - val_accuracy: 0.1069\n",
            "Epoch 310/2000\n",
            "7/7 [==============================] - 4s 600ms/step - loss: 1.4094 - accuracy: 0.8337 - val_loss: 3.4904 - val_accuracy: 0.1258\n",
            "Epoch 311/2000\n",
            "7/7 [==============================] - 4s 537ms/step - loss: 1.6133 - accuracy: 0.8005 - val_loss: 3.4581 - val_accuracy: 0.1258\n",
            "Epoch 312/2000\n",
            "7/7 [==============================] - 4s 540ms/step - loss: 1.5405 - accuracy: 0.8123 - val_loss: 3.4244 - val_accuracy: 0.1321\n",
            "Epoch 313/2000\n",
            "7/7 [==============================] - 4s 564ms/step - loss: 1.6846 - accuracy: 0.7694 - val_loss: 3.4235 - val_accuracy: 0.1447\n",
            "Epoch 314/2000\n",
            "7/7 [==============================] - 4s 585ms/step - loss: 1.7246 - accuracy: 0.7451 - val_loss: 3.4190 - val_accuracy: 0.1509\n",
            "Epoch 315/2000\n",
            "7/7 [==============================] - 5s 663ms/step - loss: 1.5775 - accuracy: 0.7818 - val_loss: 3.4213 - val_accuracy: 0.1384\n",
            "Epoch 316/2000\n",
            "7/7 [==============================] - 4s 566ms/step - loss: 1.5116 - accuracy: 0.7998 - val_loss: 3.4021 - val_accuracy: 0.1635\n",
            "Epoch 317/2000\n",
            "7/7 [==============================] - 4s 599ms/step - loss: 1.6471 - accuracy: 0.7749 - val_loss: 3.3963 - val_accuracy: 0.1447\n",
            "Epoch 318/2000\n",
            "7/7 [==============================] - 4s 614ms/step - loss: 1.5254 - accuracy: 0.8070 - val_loss: 3.4297 - val_accuracy: 0.1572\n",
            "Epoch 319/2000\n",
            "7/7 [==============================] - 4s 565ms/step - loss: 1.6611 - accuracy: 0.7711 - val_loss: 3.4798 - val_accuracy: 0.1509\n",
            "Epoch 320/2000\n",
            "7/7 [==============================] - 4s 608ms/step - loss: 1.7539 - accuracy: 0.7438 - val_loss: 3.4589 - val_accuracy: 0.1572\n",
            "Epoch 321/2000\n",
            "7/7 [==============================] - 4s 561ms/step - loss: 1.5362 - accuracy: 0.7982 - val_loss: 3.4579 - val_accuracy: 0.1509\n",
            "Epoch 322/2000\n",
            "7/7 [==============================] - 4s 549ms/step - loss: 1.7661 - accuracy: 0.7415 - val_loss: 3.4425 - val_accuracy: 0.1509\n",
            "Epoch 323/2000\n",
            "7/7 [==============================] - 4s 588ms/step - loss: 1.7068 - accuracy: 0.7601 - val_loss: 3.4854 - val_accuracy: 0.1384\n",
            "Epoch 324/2000\n",
            "7/7 [==============================] - 4s 559ms/step - loss: 1.8627 - accuracy: 0.7252 - val_loss: 3.5493 - val_accuracy: 0.1447\n",
            "Epoch 325/2000\n",
            "7/7 [==============================] - 4s 566ms/step - loss: 1.6408 - accuracy: 0.7738 - val_loss: 3.6229 - val_accuracy: 0.1321\n",
            "Epoch 326/2000\n",
            "7/7 [==============================] - 4s 616ms/step - loss: 1.7291 - accuracy: 0.7148 - val_loss: 3.6192 - val_accuracy: 0.1384\n",
            "Epoch 327/2000\n",
            "7/7 [==============================] - 4s 540ms/step - loss: 1.9134 - accuracy: 0.6877 - val_loss: 3.6223 - val_accuracy: 0.1258\n",
            "Epoch 328/2000\n",
            "7/7 [==============================] - 4s 562ms/step - loss: 1.7410 - accuracy: 0.7327 - val_loss: 3.6162 - val_accuracy: 0.1195\n",
            "Epoch 329/2000\n",
            "7/7 [==============================] - 4s 565ms/step - loss: 1.5767 - accuracy: 0.7724 - val_loss: 3.5753 - val_accuracy: 0.1321\n",
            "Epoch 330/2000\n",
            "7/7 [==============================] - 4s 582ms/step - loss: 1.6005 - accuracy: 0.7808 - val_loss: 3.6452 - val_accuracy: 0.1132\n",
            "Epoch 331/2000\n",
            "7/7 [==============================] - 4s 564ms/step - loss: 1.5770 - accuracy: 0.7915 - val_loss: 3.6064 - val_accuracy: 0.1321\n",
            "Epoch 332/2000\n",
            "7/7 [==============================] - 4s 638ms/step - loss: 1.6455 - accuracy: 0.7246 - val_loss: 3.6113 - val_accuracy: 0.1321\n",
            "Epoch 333/2000\n",
            "7/7 [==============================] - 4s 556ms/step - loss: 1.7583 - accuracy: 0.7413 - val_loss: 3.7881 - val_accuracy: 0.1132\n",
            "Epoch 334/2000\n",
            "7/7 [==============================] - 4s 597ms/step - loss: 1.5252 - accuracy: 0.7775 - val_loss: 3.7630 - val_accuracy: 0.1195\n",
            "Epoch 335/2000\n",
            "7/7 [==============================] - 4s 551ms/step - loss: 1.5935 - accuracy: 0.7547 - val_loss: 3.6991 - val_accuracy: 0.1132\n",
            "Epoch 336/2000\n",
            "7/7 [==============================] - 4s 590ms/step - loss: 1.3910 - accuracy: 0.8011 - val_loss: 3.6730 - val_accuracy: 0.1195\n",
            "Epoch 337/2000\n",
            "7/7 [==============================] - 4s 625ms/step - loss: 1.8204 - accuracy: 0.7054 - val_loss: 3.8357 - val_accuracy: 0.1132\n",
            "Epoch 338/2000\n",
            "7/7 [==============================] - 4s 622ms/step - loss: 1.6308 - accuracy: 0.7564 - val_loss: 3.8609 - val_accuracy: 0.1258\n",
            "Epoch 339/2000\n",
            "7/7 [==============================] - 4s 552ms/step - loss: 1.5648 - accuracy: 0.7586 - val_loss: 3.8844 - val_accuracy: 0.1069\n",
            "Epoch 340/2000\n",
            "7/7 [==============================] - 4s 565ms/step - loss: 1.5829 - accuracy: 0.7494 - val_loss: 3.8866 - val_accuracy: 0.1069\n",
            "Epoch 341/2000\n",
            "7/7 [==============================] - 4s 648ms/step - loss: 1.7251 - accuracy: 0.7236 - val_loss: 3.9328 - val_accuracy: 0.1069\n",
            "Epoch 342/2000\n",
            "7/7 [==============================] - 4s 632ms/step - loss: 1.5079 - accuracy: 0.7902 - val_loss: 3.8645 - val_accuracy: 0.1132\n",
            "Epoch 343/2000\n",
            "7/7 [==============================] - 4s 635ms/step - loss: 1.3746 - accuracy: 0.8147 - val_loss: 3.8486 - val_accuracy: 0.1132\n",
            "Epoch 344/2000\n",
            "7/7 [==============================] - 4s 569ms/step - loss: 1.4782 - accuracy: 0.7918 - val_loss: 3.8594 - val_accuracy: 0.1069\n",
            "Epoch 345/2000\n",
            "7/7 [==============================] - 4s 564ms/step - loss: 1.4962 - accuracy: 0.7687 - val_loss: 3.8298 - val_accuracy: 0.1006\n",
            "Epoch 346/2000\n",
            "7/7 [==============================] - 4s 574ms/step - loss: 1.5096 - accuracy: 0.7877 - val_loss: 3.8387 - val_accuracy: 0.1132\n",
            "Epoch 347/2000\n",
            "7/7 [==============================] - 4s 558ms/step - loss: 1.3884 - accuracy: 0.7956 - val_loss: 3.8618 - val_accuracy: 0.1195\n",
            "Epoch 348/2000\n",
            "7/7 [==============================] - 4s 615ms/step - loss: 1.4697 - accuracy: 0.7889 - val_loss: 3.8378 - val_accuracy: 0.1195\n",
            "Epoch 349/2000\n",
            "7/7 [==============================] - 4s 598ms/step - loss: 1.3271 - accuracy: 0.8188 - val_loss: 3.8091 - val_accuracy: 0.1195\n",
            "Epoch 350/2000\n",
            "7/7 [==============================] - 4s 549ms/step - loss: 1.2860 - accuracy: 0.8435 - val_loss: 3.8628 - val_accuracy: 0.1132\n",
            "Epoch 351/2000\n",
            "7/7 [==============================] - 4s 583ms/step - loss: 1.3136 - accuracy: 0.8293 - val_loss: 3.8140 - val_accuracy: 0.1069\n",
            "Epoch 352/2000\n",
            "7/7 [==============================] - 4s 567ms/step - loss: 1.4370 - accuracy: 0.8031 - val_loss: 3.8159 - val_accuracy: 0.1006\n",
            "Epoch 353/2000\n",
            "7/7 [==============================] - 4s 636ms/step - loss: 1.4185 - accuracy: 0.7880 - val_loss: 3.8347 - val_accuracy: 0.1195\n",
            "Epoch 354/2000\n",
            "7/7 [==============================] - 4s 598ms/step - loss: 1.3412 - accuracy: 0.8146 - val_loss: 3.8722 - val_accuracy: 0.1132\n",
            "Epoch 355/2000\n",
            "7/7 [==============================] - 4s 556ms/step - loss: 1.5122 - accuracy: 0.7728 - val_loss: 3.7789 - val_accuracy: 0.1069\n",
            "Epoch 356/2000\n",
            "7/7 [==============================] - 4s 553ms/step - loss: 1.3551 - accuracy: 0.8222 - val_loss: 3.8310 - val_accuracy: 0.1195\n",
            "Epoch 357/2000\n",
            "7/7 [==============================] - 4s 602ms/step - loss: 1.4655 - accuracy: 0.8307 - val_loss: 3.7736 - val_accuracy: 0.1572\n",
            "Epoch 358/2000\n",
            "7/7 [==============================] - 4s 643ms/step - loss: 1.5977 - accuracy: 0.7814 - val_loss: 3.7949 - val_accuracy: 0.1321\n",
            "Epoch 359/2000\n",
            "7/7 [==============================] - 4s 624ms/step - loss: 1.4582 - accuracy: 0.7883 - val_loss: 3.7966 - val_accuracy: 0.1258\n",
            "Epoch 360/2000\n",
            "7/7 [==============================] - 4s 547ms/step - loss: 1.3704 - accuracy: 0.8085 - val_loss: 3.8386 - val_accuracy: 0.1321\n",
            "Epoch 361/2000\n",
            "7/7 [==============================] - 4s 548ms/step - loss: 1.3877 - accuracy: 0.8313 - val_loss: 3.8405 - val_accuracy: 0.1258\n",
            "Epoch 362/2000\n",
            "7/7 [==============================] - 4s 558ms/step - loss: 1.2381 - accuracy: 0.8493 - val_loss: 3.7741 - val_accuracy: 0.1384\n",
            "Epoch 363/2000\n",
            "7/7 [==============================] - 4s 576ms/step - loss: 1.4573 - accuracy: 0.7970 - val_loss: 3.7912 - val_accuracy: 0.1447\n",
            "Epoch 364/2000\n",
            "7/7 [==============================] - 4s 547ms/step - loss: 1.2357 - accuracy: 0.8463 - val_loss: 3.7966 - val_accuracy: 0.1509\n",
            "Epoch 365/2000\n",
            "7/7 [==============================] - 4s 594ms/step - loss: 1.3315 - accuracy: 0.8330 - val_loss: 3.7715 - val_accuracy: 0.1447\n",
            "Epoch 366/2000\n",
            "7/7 [==============================] - 4s 578ms/step - loss: 1.4570 - accuracy: 0.8158 - val_loss: 3.8052 - val_accuracy: 0.1321\n",
            "Epoch 367/2000\n",
            "7/7 [==============================] - 4s 584ms/step - loss: 1.3529 - accuracy: 0.7998 - val_loss: 3.7923 - val_accuracy: 0.1321\n",
            "Epoch 368/2000\n",
            "7/7 [==============================] - 5s 683ms/step - loss: 1.5506 - accuracy: 0.7449 - val_loss: 3.7657 - val_accuracy: 0.1447\n",
            "Epoch 369/2000\n",
            "7/7 [==============================] - 4s 559ms/step - loss: 1.5004 - accuracy: 0.7958 - val_loss: 3.7644 - val_accuracy: 0.1384\n",
            "Epoch 370/2000\n",
            "7/7 [==============================] - 4s 577ms/step - loss: 1.2814 - accuracy: 0.8204 - val_loss: 3.8701 - val_accuracy: 0.1195\n",
            "Epoch 371/2000\n",
            "7/7 [==============================] - 4s 600ms/step - loss: 1.2376 - accuracy: 0.8456 - val_loss: 3.9025 - val_accuracy: 0.1132\n",
            "Epoch 372/2000\n",
            "7/7 [==============================] - 4s 576ms/step - loss: 1.5660 - accuracy: 0.8029 - val_loss: 3.9689 - val_accuracy: 0.0943\n",
            "Epoch 373/2000\n",
            "7/7 [==============================] - 4s 562ms/step - loss: 1.3634 - accuracy: 0.8146 - val_loss: 3.9758 - val_accuracy: 0.0943\n",
            "Epoch 374/2000\n",
            "7/7 [==============================] - 4s 544ms/step - loss: 1.4223 - accuracy: 0.8086 - val_loss: 4.0048 - val_accuracy: 0.0943\n",
            "Epoch 375/2000\n",
            "7/7 [==============================] - 4s 620ms/step - loss: 1.2477 - accuracy: 0.8383 - val_loss: 3.9543 - val_accuracy: 0.1006\n",
            "Epoch 376/2000\n",
            "7/7 [==============================] - 4s 583ms/step - loss: 1.5294 - accuracy: 0.7559 - val_loss: 3.9873 - val_accuracy: 0.0881\n",
            "Epoch 377/2000\n",
            "7/7 [==============================] - 5s 684ms/step - loss: 1.4264 - accuracy: 0.8033 - val_loss: 4.0511 - val_accuracy: 0.0755\n",
            "Epoch 378/2000\n",
            "7/7 [==============================] - 4s 627ms/step - loss: 1.3479 - accuracy: 0.8205 - val_loss: 4.0351 - val_accuracy: 0.0881\n",
            "Epoch 379/2000\n",
            "7/7 [==============================] - 4s 651ms/step - loss: 1.2900 - accuracy: 0.8401 - val_loss: 4.0303 - val_accuracy: 0.1006\n",
            "Epoch 380/2000\n",
            "7/7 [==============================] - 4s 576ms/step - loss: 1.2069 - accuracy: 0.8298 - val_loss: 3.9505 - val_accuracy: 0.1069\n",
            "Epoch 381/2000\n",
            "7/7 [==============================] - 4s 622ms/step - loss: 1.2413 - accuracy: 0.8240 - val_loss: 3.9030 - val_accuracy: 0.1132\n",
            "Epoch 382/2000\n",
            "7/7 [==============================] - 4s 607ms/step - loss: 1.5019 - accuracy: 0.7794 - val_loss: 3.9004 - val_accuracy: 0.1069\n",
            "Epoch 383/2000\n",
            "7/7 [==============================] - 4s 577ms/step - loss: 1.5098 - accuracy: 0.7889 - val_loss: 3.8887 - val_accuracy: 0.1132\n",
            "Epoch 384/2000\n",
            "7/7 [==============================] - 4s 552ms/step - loss: 1.3331 - accuracy: 0.8201 - val_loss: 3.9268 - val_accuracy: 0.1006\n",
            "Epoch 385/2000\n",
            "7/7 [==============================] - 4s 550ms/step - loss: 1.2743 - accuracy: 0.8340 - val_loss: 3.9057 - val_accuracy: 0.1132\n",
            "Epoch 386/2000\n",
            "7/7 [==============================] - 4s 593ms/step - loss: 1.4747 - accuracy: 0.7989 - val_loss: 3.8666 - val_accuracy: 0.1006\n",
            "Epoch 387/2000\n",
            "7/7 [==============================] - 4s 571ms/step - loss: 1.2931 - accuracy: 0.8175 - val_loss: 3.8807 - val_accuracy: 0.1006\n",
            "Epoch 388/2000\n",
            "7/7 [==============================] - 4s 549ms/step - loss: 1.2441 - accuracy: 0.8350 - val_loss: 3.8506 - val_accuracy: 0.0943\n",
            "Epoch 389/2000\n",
            "7/7 [==============================] - 4s 582ms/step - loss: 1.2893 - accuracy: 0.8268 - val_loss: 3.8389 - val_accuracy: 0.1069\n",
            "Epoch 390/2000\n",
            "7/7 [==============================] - 4s 566ms/step - loss: 1.1871 - accuracy: 0.8493 - val_loss: 3.8033 - val_accuracy: 0.1195\n",
            "Epoch 391/2000\n",
            "7/7 [==============================] - 4s 578ms/step - loss: 1.1123 - accuracy: 0.8755 - val_loss: 3.7829 - val_accuracy: 0.1321\n",
            "Epoch 392/2000\n",
            "7/7 [==============================] - 5s 723ms/step - loss: 1.1349 - accuracy: 0.8583 - val_loss: 3.7752 - val_accuracy: 0.1384\n",
            "Epoch 393/2000\n",
            "7/7 [==============================] - 4s 583ms/step - loss: 1.1954 - accuracy: 0.8553 - val_loss: 3.7608 - val_accuracy: 0.1321\n",
            "Epoch 394/2000\n",
            "7/7 [==============================] - 4s 595ms/step - loss: 1.2348 - accuracy: 0.8157 - val_loss: 3.7407 - val_accuracy: 0.1447\n",
            "Epoch 395/2000\n",
            "7/7 [==============================] - 5s 677ms/step - loss: 1.4507 - accuracy: 0.7913 - val_loss: 3.7519 - val_accuracy: 0.1447\n",
            "Epoch 396/2000\n",
            "7/7 [==============================] - 4s 560ms/step - loss: 1.1636 - accuracy: 0.8566 - val_loss: 3.7269 - val_accuracy: 0.1321\n",
            "Epoch 397/2000\n",
            "7/7 [==============================] - 5s 688ms/step - loss: 1.2468 - accuracy: 0.8340 - val_loss: 3.7252 - val_accuracy: 0.1321\n",
            "Epoch 398/2000\n",
            "7/7 [==============================] - 4s 564ms/step - loss: 1.1935 - accuracy: 0.8251 - val_loss: 3.9042 - val_accuracy: 0.1132\n",
            "Epoch 399/2000\n",
            "7/7 [==============================] - 4s 613ms/step - loss: 1.2684 - accuracy: 0.8122 - val_loss: 3.8760 - val_accuracy: 0.1069\n",
            "Epoch 400/2000\n",
            "7/7 [==============================] - 5s 682ms/step - loss: 1.0830 - accuracy: 0.8556 - val_loss: 3.8545 - val_accuracy: 0.1132\n",
            "Epoch 401/2000\n",
            "7/7 [==============================] - 4s 608ms/step - loss: 1.2887 - accuracy: 0.8335 - val_loss: 3.9259 - val_accuracy: 0.0943\n",
            "Epoch 402/2000\n",
            "7/7 [==============================] - 4s 600ms/step - loss: 1.1731 - accuracy: 0.8480 - val_loss: 3.9112 - val_accuracy: 0.1006\n",
            "Epoch 403/2000\n",
            "7/7 [==============================] - 4s 562ms/step - loss: 1.1169 - accuracy: 0.8562 - val_loss: 3.9051 - val_accuracy: 0.1006\n",
            "Epoch 404/2000\n",
            "7/7 [==============================] - 4s 564ms/step - loss: 1.0974 - accuracy: 0.8783 - val_loss: 3.8173 - val_accuracy: 0.0881\n",
            "Epoch 405/2000\n",
            "7/7 [==============================] - 4s 614ms/step - loss: 1.2276 - accuracy: 0.8520 - val_loss: 3.7919 - val_accuracy: 0.0943\n",
            "Epoch 406/2000\n",
            "7/7 [==============================] - 4s 585ms/step - loss: 1.1618 - accuracy: 0.8604 - val_loss: 3.7549 - val_accuracy: 0.1195\n",
            "Epoch 407/2000\n",
            "7/7 [==============================] - 4s 563ms/step - loss: 1.1703 - accuracy: 0.8571 - val_loss: 3.6989 - val_accuracy: 0.1321\n",
            "Epoch 408/2000\n",
            "7/7 [==============================] - 4s 551ms/step - loss: 1.1555 - accuracy: 0.8766 - val_loss: 3.7376 - val_accuracy: 0.1195\n",
            "Epoch 409/2000\n",
            "7/7 [==============================] - 4s 550ms/step - loss: 1.2249 - accuracy: 0.8457 - val_loss: 3.7226 - val_accuracy: 0.1069\n",
            "Epoch 410/2000\n",
            "7/7 [==============================] - 4s 557ms/step - loss: 1.2479 - accuracy: 0.8399 - val_loss: 3.7307 - val_accuracy: 0.1132\n",
            "Epoch 411/2000\n",
            "7/7 [==============================] - 4s 560ms/step - loss: 1.2226 - accuracy: 0.8332 - val_loss: 3.7136 - val_accuracy: 0.1195\n",
            "Epoch 412/2000\n",
            "7/7 [==============================] - 4s 551ms/step - loss: 1.2223 - accuracy: 0.8399 - val_loss: 3.7109 - val_accuracy: 0.1384\n",
            "Epoch 413/2000\n",
            "7/7 [==============================] - 4s 585ms/step - loss: 1.2713 - accuracy: 0.8345 - val_loss: 3.6943 - val_accuracy: 0.1258\n",
            "Epoch 414/2000\n",
            "7/7 [==============================] - 4s 605ms/step - loss: 1.1711 - accuracy: 0.8574 - val_loss: 3.6913 - val_accuracy: 0.1258\n",
            "Epoch 415/2000\n",
            "7/7 [==============================] - 4s 558ms/step - loss: 1.1156 - accuracy: 0.8522 - val_loss: 3.6295 - val_accuracy: 0.1321\n",
            "Epoch 416/2000\n",
            "7/7 [==============================] - 5s 667ms/step - loss: 1.1856 - accuracy: 0.8530 - val_loss: 3.5903 - val_accuracy: 0.1321\n",
            "Epoch 417/2000\n",
            "7/7 [==============================] - 4s 559ms/step - loss: 1.2656 - accuracy: 0.8437 - val_loss: 3.5459 - val_accuracy: 0.1321\n",
            "Epoch 418/2000\n",
            "7/7 [==============================] - 4s 557ms/step - loss: 1.1845 - accuracy: 0.8457 - val_loss: 3.4758 - val_accuracy: 0.1384\n",
            "Epoch 419/2000\n",
            "7/7 [==============================] - 4s 593ms/step - loss: 1.2945 - accuracy: 0.8222 - val_loss: 3.5182 - val_accuracy: 0.1258\n",
            "Epoch 420/2000\n",
            "7/7 [==============================] - 4s 553ms/step - loss: 1.0731 - accuracy: 0.8703 - val_loss: 3.4940 - val_accuracy: 0.1321\n",
            "Epoch 421/2000\n",
            "7/7 [==============================] - 4s 611ms/step - loss: 1.3521 - accuracy: 0.8363 - val_loss: 3.5194 - val_accuracy: 0.1447\n",
            "Epoch 422/2000\n",
            "7/7 [==============================] - 4s 556ms/step - loss: 1.1266 - accuracy: 0.8685 - val_loss: 3.5324 - val_accuracy: 0.1384\n",
            "Epoch 423/2000\n",
            "7/7 [==============================] - 4s 624ms/step - loss: 1.1184 - accuracy: 0.8826 - val_loss: 3.5607 - val_accuracy: 0.1447\n",
            "Epoch 424/2000\n",
            "7/7 [==============================] - 4s 617ms/step - loss: 1.2832 - accuracy: 0.8360 - val_loss: 3.5353 - val_accuracy: 0.1384\n",
            "Epoch 425/2000\n",
            "7/7 [==============================] - 4s 636ms/step - loss: 1.2435 - accuracy: 0.8191 - val_loss: 3.5235 - val_accuracy: 0.1447\n",
            "Epoch 426/2000\n",
            "7/7 [==============================] - 4s 538ms/step - loss: 1.1935 - accuracy: 0.8400 - val_loss: 3.5099 - val_accuracy: 0.1447\n",
            "Epoch 427/2000\n",
            "7/7 [==============================] - 4s 541ms/step - loss: 1.0435 - accuracy: 0.8759 - val_loss: 3.4310 - val_accuracy: 0.1509\n",
            "Epoch 428/2000\n",
            "7/7 [==============================] - 4s 623ms/step - loss: 1.0426 - accuracy: 0.8793 - val_loss: 3.4054 - val_accuracy: 0.1761\n",
            "Epoch 429/2000\n",
            "7/7 [==============================] - 4s 643ms/step - loss: 1.5237 - accuracy: 0.7768 - val_loss: 3.5416 - val_accuracy: 0.1447\n",
            "Epoch 430/2000\n",
            "7/7 [==============================] - 5s 628ms/step - loss: 1.1717 - accuracy: 0.8496 - val_loss: 3.5311 - val_accuracy: 0.1384\n",
            "Epoch 431/2000\n",
            "7/7 [==============================] - 4s 575ms/step - loss: 1.1970 - accuracy: 0.8561 - val_loss: 3.5851 - val_accuracy: 0.1384\n",
            "Epoch 432/2000\n",
            "7/7 [==============================] - 5s 643ms/step - loss: 1.1885 - accuracy: 0.8544 - val_loss: 3.5452 - val_accuracy: 0.1509\n",
            "Epoch 433/2000\n",
            "7/7 [==============================] - 4s 623ms/step - loss: 1.1867 - accuracy: 0.8309 - val_loss: 3.4963 - val_accuracy: 0.1635\n",
            "Epoch 434/2000\n",
            "7/7 [==============================] - 4s 564ms/step - loss: 1.1864 - accuracy: 0.8607 - val_loss: 3.5074 - val_accuracy: 0.1447\n",
            "Epoch 435/2000\n",
            "7/7 [==============================] - 4s 591ms/step - loss: 1.1841 - accuracy: 0.8440 - val_loss: 3.4655 - val_accuracy: 0.1572\n",
            "Epoch 436/2000\n",
            "7/7 [==============================] - 4s 541ms/step - loss: 1.2331 - accuracy: 0.8256 - val_loss: 3.4090 - val_accuracy: 0.1635\n",
            "Epoch 437/2000\n",
            "7/7 [==============================] - 4s 590ms/step - loss: 1.1461 - accuracy: 0.8443 - val_loss: 3.3932 - val_accuracy: 0.1761\n",
            "Epoch 438/2000\n",
            "7/7 [==============================] - 4s 559ms/step - loss: 1.0985 - accuracy: 0.8831 - val_loss: 3.3320 - val_accuracy: 0.1887\n",
            "Epoch 439/2000\n",
            "7/7 [==============================] - 4s 545ms/step - loss: 1.1520 - accuracy: 0.8507 - val_loss: 3.3281 - val_accuracy: 0.1824\n",
            "Epoch 440/2000\n",
            "7/7 [==============================] - 4s 608ms/step - loss: 1.0592 - accuracy: 0.8564 - val_loss: 3.4576 - val_accuracy: 0.1698\n",
            "Epoch 441/2000\n",
            "7/7 [==============================] - 5s 634ms/step - loss: 1.1280 - accuracy: 0.8592 - val_loss: 3.4101 - val_accuracy: 0.1824\n",
            "Epoch 442/2000\n",
            "7/7 [==============================] - 4s 589ms/step - loss: 1.1702 - accuracy: 0.8256 - val_loss: 3.3752 - val_accuracy: 0.1761\n",
            "Epoch 443/2000\n",
            "7/7 [==============================] - 4s 596ms/step - loss: 1.4272 - accuracy: 0.8268 - val_loss: 3.4051 - val_accuracy: 0.1761\n",
            "Epoch 444/2000\n",
            "7/7 [==============================] - 4s 563ms/step - loss: 1.0976 - accuracy: 0.8847 - val_loss: 3.3725 - val_accuracy: 0.1887\n",
            "Epoch 445/2000\n",
            "7/7 [==============================] - 4s 627ms/step - loss: 1.1340 - accuracy: 0.8483 - val_loss: 3.3531 - val_accuracy: 0.1698\n",
            "Epoch 446/2000\n",
            "7/7 [==============================] - 4s 565ms/step - loss: 1.2454 - accuracy: 0.8419 - val_loss: 3.3570 - val_accuracy: 0.1698\n",
            "Epoch 447/2000\n",
            "7/7 [==============================] - 5s 673ms/step - loss: 1.1723 - accuracy: 0.8620 - val_loss: 3.3315 - val_accuracy: 0.1761\n",
            "Epoch 448/2000\n",
            "7/7 [==============================] - 5s 700ms/step - loss: 1.6591 - accuracy: 0.7746 - val_loss: 3.4371 - val_accuracy: 0.1698\n",
            "Epoch 449/2000\n",
            "7/7 [==============================] - 5s 700ms/step - loss: 1.3852 - accuracy: 0.8093 - val_loss: 3.4386 - val_accuracy: 0.1635\n",
            "Epoch 450/2000\n",
            "7/7 [==============================] - 4s 546ms/step - loss: 1.3832 - accuracy: 0.8142 - val_loss: 3.4152 - val_accuracy: 0.1824\n",
            "Epoch 451/2000\n",
            "7/7 [==============================] - 4s 559ms/step - loss: 1.4620 - accuracy: 0.8077 - val_loss: 3.4172 - val_accuracy: 0.1761\n",
            "Epoch 452/2000\n",
            "7/7 [==============================] - 4s 547ms/step - loss: 1.1594 - accuracy: 0.8493 - val_loss: 3.3972 - val_accuracy: 0.1698\n",
            "Epoch 453/2000\n",
            "7/7 [==============================] - 4s 551ms/step - loss: 1.2190 - accuracy: 0.8303 - val_loss: 3.4287 - val_accuracy: 0.1824\n",
            "Epoch 454/2000\n",
            "7/7 [==============================] - 4s 597ms/step - loss: 1.4537 - accuracy: 0.8207 - val_loss: 3.4513 - val_accuracy: 0.1698\n",
            "Epoch 455/2000\n",
            "7/7 [==============================] - 4s 553ms/step - loss: 1.3576 - accuracy: 0.7972 - val_loss: 3.4900 - val_accuracy: 0.1572\n",
            "Epoch 456/2000\n",
            "7/7 [==============================] - 4s 557ms/step - loss: 1.4837 - accuracy: 0.8020 - val_loss: 3.5736 - val_accuracy: 0.1635\n",
            "Epoch 457/2000\n",
            "7/7 [==============================] - 4s 560ms/step - loss: 1.6415 - accuracy: 0.7556 - val_loss: 3.7032 - val_accuracy: 0.1509\n",
            "Epoch 458/2000\n",
            "7/7 [==============================] - 4s 609ms/step - loss: 1.2366 - accuracy: 0.8319 - val_loss: 3.6680 - val_accuracy: 0.1447\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8S9jhj0JAGNx"
      },
      "source": [
        "## Baseline Model Using CNN"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NEd-WNGxAFcw"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uYT-Z8E6KbqT",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "635fb132-8214-4ec6-b492-4a7dd83bf9fa"
      },
      "source": [
        "# X_train=X_train.reshape(211,60,18,1,2)\n",
        "# y_train=y_train.reshape(211,44)\n",
        "# X_test=X_test.reshape(53,60,18,1,2)\n",
        "# y_test=y_test.reshape(53,44)\n",
        "\n",
        "print(X_train.shape, X_test.shape, y_train.shape, y_test.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(211, 60, 18, 2) (53, 60, 18, 2) (211, 44) (53, 44)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BtrKbQsbWVUg",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "cb36557c-2677-485d-850a-301e0f25cd4c"
      },
      "source": [
        "   # Conv 1\n",
        "   #X_train.shape, X_test.shape, y_train.shape, y_test.shape\n",
        "   #(211, 60, 36, 1) (53, 60, 36, 1) (211, 44) (53, 44)\n",
        "def create_model():\n",
        "    spatial_dropout_rate_1=0.1 \n",
        "    spatial_dropout_rate_2=0.1\n",
        "    l2_rate=0.01\n",
        "\n",
        "    model = Sequential()\n",
        "    model.add(Conv2D(filters=256, \n",
        "                     kernel_size=(3, 3), \n",
        "                     padding='same',\n",
        "                     kernel_regularizer=l2(l2_rate), \n",
        "                     input_shape=(60,18,2)))\n",
        "    model.add(LeakyReLU(alpha=0.1))\n",
        "    model.add(BatchNormalization())\n",
        "\n",
        "    #model.add(SpatialDropout2D(spatial_dropout_rate_1))\n",
        "    #Conv 2\n",
        "    model.add(Conv2D(filters=256, \n",
        "                     kernel_size=(3, 3),\n",
        "                     padding='same', \n",
        "                     kernel_regularizer=l2(l2_rate)))\n",
        "    model.add(LeakyReLU(alpha=0.1))\n",
        "    model.add(BatchNormalization())\n",
        "\n",
        "    #model.add(AveragePooling2D(pool_size=(2,2)))\n",
        "    # Max Pooling #1\n",
        "    model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "    model.add(SpatialDropout2D(spatial_dropout_rate_1))\n",
        "    #Conv 3\n",
        "    model.add(Conv2D(filters=256, \n",
        "                     kernel_size=(5, 5),\n",
        "                     padding='same', \n",
        "                     kernel_regularizer=l2(l2_rate)))\n",
        "    model.add(LeakyReLU(alpha=0.1))\n",
        "    model.add(BatchNormalization())\n",
        "\n",
        "    model.add(SpatialDropout2D(spatial_dropout_rate_1))\n",
        "    # #Conv 4\n",
        "    model.add(Conv2D(filters=256, \n",
        "                     kernel_size=(3,3),\n",
        "                     padding='same', \n",
        "                     kernel_regularizer=l2(l2_rate)))\n",
        "    model.add(LeakyReLU(alpha=0.1))\n",
        "    model.add(BatchNormalization())\n",
        "    \n",
        "    #model.add(AveragePooling2D(pool_size=(2,2)))\n",
        "    model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "    model.add(SpatialDropout2D(spatial_dropout_rate_2))\n",
        "\n",
        "\n",
        "    model.add(GlobalAveragePooling2D())\n",
        "\n",
        "\n",
        "    # Softmax output\n",
        "    model.add(Dense(44, activation='softmax'))\n",
        "    adm=Adam(lr=0.0001, beta_1=0.9, beta_2=0.999, epsilon=None, decay=0.01, amsgrad=False) \n",
        "\n",
        "    #model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "    model.compile(loss='categorical_crossentropy', optimizer=adm, metrics=['accuracy'])\n",
        "    print(model.summary())\n",
        "    return model\n",
        "\n",
        "model = create_model()\n",
        "\n",
        "\n",
        "# Create the training folder\n",
        "#\n",
        "training_session_id = datetime.now().strftime(\"%Y-%m-%d %H-%M-%S\")\n",
        "training_session_folder = 'train_%s' % (training_session_id)\n",
        "os.makedirs(training_session_folder, exist_ok=True)\n",
        "\n",
        "# Configure the checkpoint and stop point.\n",
        "# This allows the training to save the best models and also stop the\n",
        "# training early if it detects that there are no improvements after\n",
        "# a long time.\n",
        "#\n",
        "callbacks_list = [\n",
        "    keras.callbacks.ModelCheckpoint(\n",
        "        filepath=training_session_folder + '/model.{epoch:04d}-val_acc-{val_accuracy:4.2f}-loss-{val_loss:4.2f}.h5',\n",
        "        monitor='val_loss', save_best_only=True),\n",
        "    keras.callbacks.EarlyStopping(monitor='val_loss', patience=300, restore_best_weights=True)\n",
        "]\n",
        "\n",
        "\n",
        "# Train our model\n",
        "#\n",
        "history = model.fit(x=X_train, y=y_train, batch_size=50, epochs=2000, verbose=1, callbacks=callbacks_list, validation_data=(X_test, y_test))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_54\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_45 (Conv2D)           (None, 60, 18, 256)       4864      \n",
            "_________________________________________________________________\n",
            "leaky_re_lu_20 (LeakyReLU)   (None, 60, 18, 256)       0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_46 (Batc (None, 60, 18, 256)       1024      \n",
            "_________________________________________________________________\n",
            "conv2d_46 (Conv2D)           (None, 60, 18, 256)       590080    \n",
            "_________________________________________________________________\n",
            "leaky_re_lu_21 (LeakyReLU)   (None, 60, 18, 256)       0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_47 (Batc (None, 60, 18, 256)       1024      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_10 (MaxPooling (None, 30, 9, 256)        0         \n",
            "_________________________________________________________________\n",
            "spatial_dropout2d_17 (Spatia (None, 30, 9, 256)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_47 (Conv2D)           (None, 30, 9, 256)        1638656   \n",
            "_________________________________________________________________\n",
            "leaky_re_lu_22 (LeakyReLU)   (None, 30, 9, 256)        0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_48 (Batc (None, 30, 9, 256)        1024      \n",
            "_________________________________________________________________\n",
            "spatial_dropout2d_18 (Spatia (None, 30, 9, 256)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_48 (Conv2D)           (None, 30, 9, 256)        590080    \n",
            "_________________________________________________________________\n",
            "leaky_re_lu_23 (LeakyReLU)   (None, 30, 9, 256)        0         \n",
            "_________________________________________________________________\n",
            "batch_normalization_49 (Batc (None, 30, 9, 256)        1024      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_11 (MaxPooling (None, 15, 4, 256)        0         \n",
            "_________________________________________________________________\n",
            "spatial_dropout2d_19 (Spatia (None, 15, 4, 256)        0         \n",
            "_________________________________________________________________\n",
            "global_average_pooling2d_12  (None, 256)               0         \n",
            "_________________________________________________________________\n",
            "dense_54 (Dense)             (None, 44)                11308     \n",
            "=================================================================\n",
            "Total params: 2,839,084\n",
            "Trainable params: 2,837,036\n",
            "Non-trainable params: 2,048\n",
            "_________________________________________________________________\n",
            "None\n",
            "Epoch 1/2000\n",
            "5/5 [==============================] - 1s 85ms/step - loss: 11.9485 - accuracy: 0.0122 - val_loss: 11.4636 - val_accuracy: 0.0566\n",
            "Epoch 2/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 11.4732 - accuracy: 0.0623 - val_loss: 11.4005 - val_accuracy: 0.0377\n",
            "Epoch 3/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 11.2009 - accuracy: 0.1112 - val_loss: 11.3285 - val_accuracy: 0.0566\n",
            "Epoch 4/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 11.0755 - accuracy: 0.1148 - val_loss: 11.2677 - val_accuracy: 0.0377\n",
            "Epoch 5/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 10.9613 - accuracy: 0.1085 - val_loss: 11.2095 - val_accuracy: 0.0566\n",
            "Epoch 6/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 10.7861 - accuracy: 0.1420 - val_loss: 11.1529 - val_accuracy: 0.0377\n",
            "Epoch 7/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 10.7015 - accuracy: 0.1453 - val_loss: 11.0997 - val_accuracy: 0.0566\n",
            "Epoch 8/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 10.5516 - accuracy: 0.1771 - val_loss: 11.0497 - val_accuracy: 0.0755\n",
            "Epoch 9/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 10.4804 - accuracy: 0.2083 - val_loss: 10.9995 - val_accuracy: 0.0755\n",
            "Epoch 10/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 10.4640 - accuracy: 0.1441 - val_loss: 10.9509 - val_accuracy: 0.0943\n",
            "Epoch 11/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 10.2774 - accuracy: 0.1995 - val_loss: 10.9017 - val_accuracy: 0.0943\n",
            "Epoch 12/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 10.1488 - accuracy: 0.2667 - val_loss: 10.8558 - val_accuracy: 0.0943\n",
            "Epoch 13/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 10.1027 - accuracy: 0.2562 - val_loss: 10.8131 - val_accuracy: 0.0943\n",
            "Epoch 14/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 10.0128 - accuracy: 0.2479 - val_loss: 10.7698 - val_accuracy: 0.0943\n",
            "Epoch 15/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 9.9313 - accuracy: 0.2319 - val_loss: 10.7289 - val_accuracy: 0.0943\n",
            "Epoch 16/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 9.8649 - accuracy: 0.2824 - val_loss: 10.6899 - val_accuracy: 0.1321\n",
            "Epoch 17/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 9.8311 - accuracy: 0.2842 - val_loss: 10.6501 - val_accuracy: 0.1132\n",
            "Epoch 18/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 9.7186 - accuracy: 0.2871 - val_loss: 10.6090 - val_accuracy: 0.1321\n",
            "Epoch 19/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 9.6845 - accuracy: 0.3187 - val_loss: 10.5700 - val_accuracy: 0.1321\n",
            "Epoch 20/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 9.6023 - accuracy: 0.2974 - val_loss: 10.5347 - val_accuracy: 0.1321\n",
            "Epoch 21/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 9.5333 - accuracy: 0.3011 - val_loss: 10.4995 - val_accuracy: 0.1321\n",
            "Epoch 22/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 9.4861 - accuracy: 0.3239 - val_loss: 10.4625 - val_accuracy: 0.1321\n",
            "Epoch 23/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 9.4044 - accuracy: 0.3144 - val_loss: 10.4258 - val_accuracy: 0.1321\n",
            "Epoch 24/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 9.3667 - accuracy: 0.3175 - val_loss: 10.3887 - val_accuracy: 0.1321\n",
            "Epoch 25/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 9.2773 - accuracy: 0.3809 - val_loss: 10.3528 - val_accuracy: 0.1321\n",
            "Epoch 26/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 9.2868 - accuracy: 0.3714 - val_loss: 10.3165 - val_accuracy: 0.1321\n",
            "Epoch 27/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 9.1895 - accuracy: 0.4481 - val_loss: 10.2828 - val_accuracy: 0.1321\n",
            "Epoch 28/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 9.1729 - accuracy: 0.3515 - val_loss: 10.2486 - val_accuracy: 0.1132\n",
            "Epoch 29/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 9.1876 - accuracy: 0.3517 - val_loss: 10.2168 - val_accuracy: 0.0755\n",
            "Epoch 30/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 9.1254 - accuracy: 0.3572 - val_loss: 10.1888 - val_accuracy: 0.1132\n",
            "Epoch 31/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 9.0223 - accuracy: 0.4273 - val_loss: 10.1621 - val_accuracy: 0.1132\n",
            "Epoch 32/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 9.0092 - accuracy: 0.4123 - val_loss: 10.1356 - val_accuracy: 0.0943\n",
            "Epoch 33/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.9841 - accuracy: 0.4199 - val_loss: 10.1140 - val_accuracy: 0.0943\n",
            "Epoch 34/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.9337 - accuracy: 0.4019 - val_loss: 10.0888 - val_accuracy: 0.0943\n",
            "Epoch 35/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 8.9114 - accuracy: 0.4159 - val_loss: 10.0630 - val_accuracy: 0.0943\n",
            "Epoch 36/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.9082 - accuracy: 0.3895 - val_loss: 10.0399 - val_accuracy: 0.0755\n",
            "Epoch 37/2000\n",
            "5/5 [==============================] - 0s 88ms/step - loss: 8.8355 - accuracy: 0.4326 - val_loss: 10.0149 - val_accuracy: 0.1132\n",
            "Epoch 38/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 8.7688 - accuracy: 0.5180 - val_loss: 9.9863 - val_accuracy: 0.1132\n",
            "Epoch 39/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.7896 - accuracy: 0.4758 - val_loss: 9.9597 - val_accuracy: 0.1132\n",
            "Epoch 40/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 8.7431 - accuracy: 0.4517 - val_loss: 9.9329 - val_accuracy: 0.0755\n",
            "Epoch 41/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.7131 - accuracy: 0.4757 - val_loss: 9.9059 - val_accuracy: 0.0943\n",
            "Epoch 42/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 8.7186 - accuracy: 0.4558 - val_loss: 9.8767 - val_accuracy: 0.0755\n",
            "Epoch 43/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.6603 - accuracy: 0.4383 - val_loss: 9.8549 - val_accuracy: 0.0566\n",
            "Epoch 44/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.5623 - accuracy: 0.5479 - val_loss: 9.8328 - val_accuracy: 0.0755\n",
            "Epoch 45/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.6343 - accuracy: 0.4899 - val_loss: 9.8050 - val_accuracy: 0.0943\n",
            "Epoch 46/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.5739 - accuracy: 0.4838 - val_loss: 9.7833 - val_accuracy: 0.0755\n",
            "Epoch 47/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.5601 - accuracy: 0.4743 - val_loss: 9.7599 - val_accuracy: 0.0943\n",
            "Epoch 48/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.5685 - accuracy: 0.4631 - val_loss: 9.7361 - val_accuracy: 0.0755\n",
            "Epoch 49/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.5249 - accuracy: 0.4713 - val_loss: 9.7083 - val_accuracy: 0.0755\n",
            "Epoch 50/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.3732 - accuracy: 0.4962 - val_loss: 9.6761 - val_accuracy: 0.0755\n",
            "Epoch 51/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.4457 - accuracy: 0.5112 - val_loss: 9.6475 - val_accuracy: 0.0943\n",
            "Epoch 52/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.4488 - accuracy: 0.4720 - val_loss: 9.6246 - val_accuracy: 0.0755\n",
            "Epoch 53/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.4029 - accuracy: 0.5509 - val_loss: 9.6031 - val_accuracy: 0.0566\n",
            "Epoch 54/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 8.3852 - accuracy: 0.5464 - val_loss: 9.5772 - val_accuracy: 0.0755\n",
            "Epoch 55/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.3632 - accuracy: 0.5248 - val_loss: 9.5457 - val_accuracy: 0.0755\n",
            "Epoch 56/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 8.2732 - accuracy: 0.5312 - val_loss: 9.5226 - val_accuracy: 0.0943\n",
            "Epoch 57/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 8.2969 - accuracy: 0.5203 - val_loss: 9.5001 - val_accuracy: 0.0943\n",
            "Epoch 58/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 8.2264 - accuracy: 0.5568 - val_loss: 9.4795 - val_accuracy: 0.0755\n",
            "Epoch 59/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.2699 - accuracy: 0.5271 - val_loss: 9.4562 - val_accuracy: 0.0755\n",
            "Epoch 60/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.2160 - accuracy: 0.5641 - val_loss: 9.4383 - val_accuracy: 0.0943\n",
            "Epoch 61/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 8.2281 - accuracy: 0.5216 - val_loss: 9.4236 - val_accuracy: 0.0943\n",
            "Epoch 62/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.2247 - accuracy: 0.4967 - val_loss: 9.3997 - val_accuracy: 0.1132\n",
            "Epoch 63/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 8.1514 - accuracy: 0.5312 - val_loss: 9.3701 - val_accuracy: 0.1509\n",
            "Epoch 64/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.1382 - accuracy: 0.5596 - val_loss: 9.3506 - val_accuracy: 0.1321\n",
            "Epoch 65/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.1383 - accuracy: 0.5496 - val_loss: 9.3313 - val_accuracy: 0.1321\n",
            "Epoch 66/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 8.1482 - accuracy: 0.5474 - val_loss: 9.3126 - val_accuracy: 0.1509\n",
            "Epoch 67/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.0515 - accuracy: 0.6009 - val_loss: 9.2974 - val_accuracy: 0.1698\n",
            "Epoch 68/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.1322 - accuracy: 0.5564 - val_loss: 9.2787 - val_accuracy: 0.1321\n",
            "Epoch 69/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.0130 - accuracy: 0.6088 - val_loss: 9.2645 - val_accuracy: 0.1321\n",
            "Epoch 70/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.0804 - accuracy: 0.5526 - val_loss: 9.2489 - val_accuracy: 0.1321\n",
            "Epoch 71/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 8.1147 - accuracy: 0.5339 - val_loss: 9.2325 - val_accuracy: 0.1321\n",
            "Epoch 72/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 8.0121 - accuracy: 0.5552 - val_loss: 9.2148 - val_accuracy: 0.1321\n",
            "Epoch 73/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 8.0037 - accuracy: 0.5425 - val_loss: 9.1875 - val_accuracy: 0.1321\n",
            "Epoch 74/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.9781 - accuracy: 0.5505 - val_loss: 9.1669 - val_accuracy: 0.1509\n",
            "Epoch 75/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 8.0278 - accuracy: 0.5445 - val_loss: 9.1505 - val_accuracy: 0.1321\n",
            "Epoch 76/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.9929 - accuracy: 0.5430 - val_loss: 9.1376 - val_accuracy: 0.1509\n",
            "Epoch 77/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 7.8438 - accuracy: 0.6076 - val_loss: 9.1305 - val_accuracy: 0.1509\n",
            "Epoch 78/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 7.8711 - accuracy: 0.6045 - val_loss: 9.1233 - val_accuracy: 0.1321\n",
            "Epoch 79/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.8790 - accuracy: 0.6093 - val_loss: 9.1106 - val_accuracy: 0.1698\n",
            "Epoch 80/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 7.8805 - accuracy: 0.5657 - val_loss: 9.0923 - val_accuracy: 0.1509\n",
            "Epoch 81/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 7.8628 - accuracy: 0.5560 - val_loss: 9.0688 - val_accuracy: 0.1509\n",
            "Epoch 82/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.8925 - accuracy: 0.5501 - val_loss: 9.0510 - val_accuracy: 0.1509\n",
            "Epoch 83/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.8722 - accuracy: 0.5879 - val_loss: 9.0451 - val_accuracy: 0.1509\n",
            "Epoch 84/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.8271 - accuracy: 0.6078 - val_loss: 9.0382 - val_accuracy: 0.1509\n",
            "Epoch 85/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.8010 - accuracy: 0.6047 - val_loss: 9.0271 - val_accuracy: 0.1698\n",
            "Epoch 86/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.8522 - accuracy: 0.5735 - val_loss: 9.0193 - val_accuracy: 0.1698\n",
            "Epoch 87/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.7444 - accuracy: 0.6470 - val_loss: 9.0122 - val_accuracy: 0.1509\n",
            "Epoch 88/2000\n",
            "5/5 [==============================] - 0s 52ms/step - loss: 7.7405 - accuracy: 0.5858 - val_loss: 8.9948 - val_accuracy: 0.1698\n",
            "Epoch 89/2000\n",
            "5/5 [==============================] - 0s 49ms/step - loss: 7.7649 - accuracy: 0.6327 - val_loss: 8.9733 - val_accuracy: 0.1698\n",
            "Epoch 90/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.7309 - accuracy: 0.6129 - val_loss: 8.9565 - val_accuracy: 0.1698\n",
            "Epoch 91/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.7646 - accuracy: 0.5806 - val_loss: 8.9363 - val_accuracy: 0.1698\n",
            "Epoch 92/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.6796 - accuracy: 0.6303 - val_loss: 8.9148 - val_accuracy: 0.1698\n",
            "Epoch 93/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.7474 - accuracy: 0.5819 - val_loss: 8.8931 - val_accuracy: 0.1698\n",
            "Epoch 94/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.6978 - accuracy: 0.6258 - val_loss: 8.8784 - val_accuracy: 0.1698\n",
            "Epoch 95/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.7122 - accuracy: 0.6077 - val_loss: 8.8618 - val_accuracy: 0.1509\n",
            "Epoch 96/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.6976 - accuracy: 0.6279 - val_loss: 8.8468 - val_accuracy: 0.1509\n",
            "Epoch 97/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.6618 - accuracy: 0.5986 - val_loss: 8.8416 - val_accuracy: 0.1321\n",
            "Epoch 98/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.5812 - accuracy: 0.6566 - val_loss: 8.8275 - val_accuracy: 0.1321\n",
            "Epoch 99/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.6554 - accuracy: 0.6234 - val_loss: 8.8144 - val_accuracy: 0.1132\n",
            "Epoch 100/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.6420 - accuracy: 0.6155 - val_loss: 8.7993 - val_accuracy: 0.1132\n",
            "Epoch 101/2000\n",
            "5/5 [==============================] - 0s 76ms/step - loss: 7.5729 - accuracy: 0.6286 - val_loss: 8.7860 - val_accuracy: 0.1321\n",
            "Epoch 102/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.5560 - accuracy: 0.6342 - val_loss: 8.7775 - val_accuracy: 0.1321\n",
            "Epoch 103/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.5821 - accuracy: 0.6058 - val_loss: 8.7634 - val_accuracy: 0.1509\n",
            "Epoch 104/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.5436 - accuracy: 0.6628 - val_loss: 8.7589 - val_accuracy: 0.1321\n",
            "Epoch 105/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.5153 - accuracy: 0.6438 - val_loss: 8.7448 - val_accuracy: 0.2075\n",
            "Epoch 106/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.6190 - accuracy: 0.6292 - val_loss: 8.7341 - val_accuracy: 0.1887\n",
            "Epoch 107/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.5240 - accuracy: 0.6249 - val_loss: 8.7238 - val_accuracy: 0.1887\n",
            "Epoch 108/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.5558 - accuracy: 0.5908 - val_loss: 8.7151 - val_accuracy: 0.1887\n",
            "Epoch 109/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.5354 - accuracy: 0.6430 - val_loss: 8.7002 - val_accuracy: 0.1698\n",
            "Epoch 110/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.4958 - accuracy: 0.6512 - val_loss: 8.6885 - val_accuracy: 0.1698\n",
            "Epoch 111/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 7.5779 - accuracy: 0.6187 - val_loss: 8.6785 - val_accuracy: 0.1698\n",
            "Epoch 112/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 7.5421 - accuracy: 0.6460 - val_loss: 8.6616 - val_accuracy: 0.2075\n",
            "Epoch 113/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.4144 - accuracy: 0.6477 - val_loss: 8.6538 - val_accuracy: 0.2075\n",
            "Epoch 114/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.4766 - accuracy: 0.6256 - val_loss: 8.6435 - val_accuracy: 0.1698\n",
            "Epoch 115/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.4489 - accuracy: 0.6422 - val_loss: 8.6381 - val_accuracy: 0.1698\n",
            "Epoch 116/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.4548 - accuracy: 0.6496 - val_loss: 8.6291 - val_accuracy: 0.1698\n",
            "Epoch 117/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.4230 - accuracy: 0.6167 - val_loss: 8.6209 - val_accuracy: 0.1887\n",
            "Epoch 118/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.4043 - accuracy: 0.6828 - val_loss: 8.6159 - val_accuracy: 0.1698\n",
            "Epoch 119/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.3907 - accuracy: 0.6470 - val_loss: 8.6030 - val_accuracy: 0.1698\n",
            "Epoch 120/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.4158 - accuracy: 0.6435 - val_loss: 8.5937 - val_accuracy: 0.1698\n",
            "Epoch 121/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.3869 - accuracy: 0.6538 - val_loss: 8.5900 - val_accuracy: 0.1698\n",
            "Epoch 122/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.3334 - accuracy: 0.6567 - val_loss: 8.5837 - val_accuracy: 0.1509\n",
            "Epoch 123/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.3625 - accuracy: 0.6685 - val_loss: 8.5730 - val_accuracy: 0.1698\n",
            "Epoch 124/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 7.3388 - accuracy: 0.6643 - val_loss: 8.5681 - val_accuracy: 0.1698\n",
            "Epoch 125/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.3284 - accuracy: 0.6751 - val_loss: 8.5561 - val_accuracy: 0.1887\n",
            "Epoch 126/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.3972 - accuracy: 0.6719 - val_loss: 8.5504 - val_accuracy: 0.2075\n",
            "Epoch 127/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.2983 - accuracy: 0.7006 - val_loss: 8.5504 - val_accuracy: 0.2075\n",
            "Epoch 128/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.3220 - accuracy: 0.6389 - val_loss: 8.5521 - val_accuracy: 0.2075\n",
            "Epoch 129/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.2657 - accuracy: 0.6799 - val_loss: 8.5473 - val_accuracy: 0.2264\n",
            "Epoch 130/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.2989 - accuracy: 0.6448 - val_loss: 8.5378 - val_accuracy: 0.2075\n",
            "Epoch 131/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 7.2808 - accuracy: 0.6548 - val_loss: 8.5317 - val_accuracy: 0.2075\n",
            "Epoch 132/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.2818 - accuracy: 0.6843 - val_loss: 8.5229 - val_accuracy: 0.1887\n",
            "Epoch 133/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.2471 - accuracy: 0.6918 - val_loss: 8.5228 - val_accuracy: 0.1887\n",
            "Epoch 134/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.2669 - accuracy: 0.6962 - val_loss: 8.5146 - val_accuracy: 0.2075\n",
            "Epoch 135/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.2673 - accuracy: 0.6496 - val_loss: 8.5136 - val_accuracy: 0.1698\n",
            "Epoch 136/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 7.2939 - accuracy: 0.7138 - val_loss: 8.4998 - val_accuracy: 0.1698\n",
            "Epoch 137/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.2916 - accuracy: 0.6530 - val_loss: 8.4893 - val_accuracy: 0.1698\n",
            "Epoch 138/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.2459 - accuracy: 0.6961 - val_loss: 8.4719 - val_accuracy: 0.1887\n",
            "Epoch 139/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.2709 - accuracy: 0.6629 - val_loss: 8.4616 - val_accuracy: 0.1887\n",
            "Epoch 140/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.2208 - accuracy: 0.6587 - val_loss: 8.4619 - val_accuracy: 0.1698\n",
            "Epoch 141/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.1966 - accuracy: 0.6951 - val_loss: 8.4566 - val_accuracy: 0.1509\n",
            "Epoch 142/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.1923 - accuracy: 0.6953 - val_loss: 8.4427 - val_accuracy: 0.1509\n",
            "Epoch 143/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.2745 - accuracy: 0.6574 - val_loss: 8.4260 - val_accuracy: 0.1887\n",
            "Epoch 144/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.2007 - accuracy: 0.6497 - val_loss: 8.4118 - val_accuracy: 0.1887\n",
            "Epoch 145/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.1954 - accuracy: 0.6887 - val_loss: 8.4040 - val_accuracy: 0.1887\n",
            "Epoch 146/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.2037 - accuracy: 0.7103 - val_loss: 8.3977 - val_accuracy: 0.1887\n",
            "Epoch 147/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.2257 - accuracy: 0.6559 - val_loss: 8.3938 - val_accuracy: 0.2075\n",
            "Epoch 148/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.1778 - accuracy: 0.6682 - val_loss: 8.3953 - val_accuracy: 0.2075\n",
            "Epoch 149/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.1883 - accuracy: 0.6632 - val_loss: 8.3932 - val_accuracy: 0.2075\n",
            "Epoch 150/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.1478 - accuracy: 0.7050 - val_loss: 8.3893 - val_accuracy: 0.1698\n",
            "Epoch 151/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.1901 - accuracy: 0.6511 - val_loss: 8.3810 - val_accuracy: 0.1698\n",
            "Epoch 152/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.1790 - accuracy: 0.6785 - val_loss: 8.3753 - val_accuracy: 0.1887\n",
            "Epoch 153/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 7.1499 - accuracy: 0.6766 - val_loss: 8.3696 - val_accuracy: 0.1698\n",
            "Epoch 154/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.1286 - accuracy: 0.6679 - val_loss: 8.3727 - val_accuracy: 0.1698\n",
            "Epoch 155/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.1068 - accuracy: 0.7040 - val_loss: 8.3682 - val_accuracy: 0.1887\n",
            "Epoch 156/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.1027 - accuracy: 0.6849 - val_loss: 8.3608 - val_accuracy: 0.2075\n",
            "Epoch 157/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.0585 - accuracy: 0.7621 - val_loss: 8.3470 - val_accuracy: 0.2075\n",
            "Epoch 158/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.0745 - accuracy: 0.7164 - val_loss: 8.3420 - val_accuracy: 0.1887\n",
            "Epoch 159/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 7.0687 - accuracy: 0.7143 - val_loss: 8.3324 - val_accuracy: 0.1887\n",
            "Epoch 160/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 7.0602 - accuracy: 0.6991 - val_loss: 8.3270 - val_accuracy: 0.1887\n",
            "Epoch 161/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 7.0890 - accuracy: 0.6892 - val_loss: 8.3231 - val_accuracy: 0.1887\n",
            "Epoch 162/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.9505 - accuracy: 0.7462 - val_loss: 8.3145 - val_accuracy: 0.2075\n",
            "Epoch 163/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.1016 - accuracy: 0.6741 - val_loss: 8.3102 - val_accuracy: 0.1887\n",
            "Epoch 164/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.0893 - accuracy: 0.6659 - val_loss: 8.3103 - val_accuracy: 0.1887\n",
            "Epoch 165/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 7.0457 - accuracy: 0.7225 - val_loss: 8.3080 - val_accuracy: 0.2075\n",
            "Epoch 166/2000\n",
            "5/5 [==============================] - 0s 78ms/step - loss: 7.0926 - accuracy: 0.6645 - val_loss: 8.3049 - val_accuracy: 0.2075\n",
            "Epoch 167/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.0750 - accuracy: 0.6920 - val_loss: 8.3018 - val_accuracy: 0.2075\n",
            "Epoch 168/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.0449 - accuracy: 0.6945 - val_loss: 8.2953 - val_accuracy: 0.2264\n",
            "Epoch 169/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.0289 - accuracy: 0.7176 - val_loss: 8.2851 - val_accuracy: 0.2264\n",
            "Epoch 170/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.0313 - accuracy: 0.6887 - val_loss: 8.2731 - val_accuracy: 0.2075\n",
            "Epoch 171/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.0488 - accuracy: 0.6581 - val_loss: 8.2649 - val_accuracy: 0.2075\n",
            "Epoch 172/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.9953 - accuracy: 0.6889 - val_loss: 8.2568 - val_accuracy: 0.1887\n",
            "Epoch 173/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.0121 - accuracy: 0.6927 - val_loss: 8.2479 - val_accuracy: 0.2075\n",
            "Epoch 174/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 7.0252 - accuracy: 0.6888 - val_loss: 8.2413 - val_accuracy: 0.2075\n",
            "Epoch 175/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 7.0526 - accuracy: 0.6899 - val_loss: 8.2296 - val_accuracy: 0.1698\n",
            "Epoch 176/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 7.0189 - accuracy: 0.6554 - val_loss: 8.2214 - val_accuracy: 0.1698\n",
            "Epoch 177/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.9821 - accuracy: 0.6955 - val_loss: 8.2111 - val_accuracy: 0.1887\n",
            "Epoch 178/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 6.9324 - accuracy: 0.7013 - val_loss: 8.1982 - val_accuracy: 0.2075\n",
            "Epoch 179/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.9428 - accuracy: 0.7203 - val_loss: 8.1927 - val_accuracy: 0.2264\n",
            "Epoch 180/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.9326 - accuracy: 0.7129 - val_loss: 8.1898 - val_accuracy: 0.2075\n",
            "Epoch 181/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.9203 - accuracy: 0.7261 - val_loss: 8.1892 - val_accuracy: 0.2075\n",
            "Epoch 182/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.9925 - accuracy: 0.7228 - val_loss: 8.1890 - val_accuracy: 0.2075\n",
            "Epoch 183/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.9487 - accuracy: 0.7270 - val_loss: 8.1928 - val_accuracy: 0.1887\n",
            "Epoch 184/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.9590 - accuracy: 0.7123 - val_loss: 8.1976 - val_accuracy: 0.1887\n",
            "Epoch 185/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.9380 - accuracy: 0.6985 - val_loss: 8.1881 - val_accuracy: 0.1887\n",
            "Epoch 186/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.9640 - accuracy: 0.6964 - val_loss: 8.1741 - val_accuracy: 0.1887\n",
            "Epoch 187/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.8749 - accuracy: 0.7510 - val_loss: 8.1788 - val_accuracy: 0.2075\n",
            "Epoch 188/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.9256 - accuracy: 0.7415 - val_loss: 8.1739 - val_accuracy: 0.2264\n",
            "Epoch 189/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.8261 - accuracy: 0.7682 - val_loss: 8.1695 - val_accuracy: 0.2264\n",
            "Epoch 190/2000\n",
            "5/5 [==============================] - 0s 52ms/step - loss: 6.8745 - accuracy: 0.7388 - val_loss: 8.1649 - val_accuracy: 0.2264\n",
            "Epoch 191/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 6.8854 - accuracy: 0.7344 - val_loss: 8.1635 - val_accuracy: 0.2264\n",
            "Epoch 192/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.8749 - accuracy: 0.7328 - val_loss: 8.1616 - val_accuracy: 0.2075\n",
            "Epoch 193/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.9406 - accuracy: 0.7161 - val_loss: 8.1605 - val_accuracy: 0.2075\n",
            "Epoch 194/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.8339 - accuracy: 0.7453 - val_loss: 8.1523 - val_accuracy: 0.2264\n",
            "Epoch 195/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.8679 - accuracy: 0.7479 - val_loss: 8.1444 - val_accuracy: 0.2264\n",
            "Epoch 196/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 6.8496 - accuracy: 0.7648 - val_loss: 8.1396 - val_accuracy: 0.2264\n",
            "Epoch 197/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.8756 - accuracy: 0.7348 - val_loss: 8.1378 - val_accuracy: 0.2453\n",
            "Epoch 198/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.8515 - accuracy: 0.7223 - val_loss: 8.1326 - val_accuracy: 0.2264\n",
            "Epoch 199/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.8997 - accuracy: 0.7064 - val_loss: 8.1202 - val_accuracy: 0.2264\n",
            "Epoch 200/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.8300 - accuracy: 0.7328 - val_loss: 8.1191 - val_accuracy: 0.2264\n",
            "Epoch 201/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.8381 - accuracy: 0.7037 - val_loss: 8.1129 - val_accuracy: 0.2264\n",
            "Epoch 202/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.8326 - accuracy: 0.7322 - val_loss: 8.1191 - val_accuracy: 0.2075\n",
            "Epoch 203/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.8009 - accuracy: 0.7372 - val_loss: 8.1135 - val_accuracy: 0.2264\n",
            "Epoch 204/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.8466 - accuracy: 0.7013 - val_loss: 8.1011 - val_accuracy: 0.2264\n",
            "Epoch 205/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.7916 - accuracy: 0.7723 - val_loss: 8.0968 - val_accuracy: 0.2264\n",
            "Epoch 206/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.8417 - accuracy: 0.7230 - val_loss: 8.0881 - val_accuracy: 0.2264\n",
            "Epoch 207/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.8138 - accuracy: 0.7356 - val_loss: 8.0855 - val_accuracy: 0.2453\n",
            "Epoch 208/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.8340 - accuracy: 0.6835 - val_loss: 8.0787 - val_accuracy: 0.2453\n",
            "Epoch 209/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.7766 - accuracy: 0.7255 - val_loss: 8.0801 - val_accuracy: 0.2453\n",
            "Epoch 210/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.7432 - accuracy: 0.7514 - val_loss: 8.0771 - val_accuracy: 0.2264\n",
            "Epoch 211/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.8477 - accuracy: 0.7025 - val_loss: 8.0712 - val_accuracy: 0.2264\n",
            "Epoch 212/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.8207 - accuracy: 0.7027 - val_loss: 8.0740 - val_accuracy: 0.2075\n",
            "Epoch 213/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.8158 - accuracy: 0.7329 - val_loss: 8.0725 - val_accuracy: 0.2075\n",
            "Epoch 214/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.8199 - accuracy: 0.7327 - val_loss: 8.0646 - val_accuracy: 0.2075\n",
            "Epoch 215/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.8003 - accuracy: 0.7212 - val_loss: 8.0622 - val_accuracy: 0.2075\n",
            "Epoch 216/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.8107 - accuracy: 0.7289 - val_loss: 8.0661 - val_accuracy: 0.2264\n",
            "Epoch 217/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.6782 - accuracy: 0.7871 - val_loss: 8.0621 - val_accuracy: 0.2264\n",
            "Epoch 218/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.7261 - accuracy: 0.7662 - val_loss: 8.0600 - val_accuracy: 0.2264\n",
            "Epoch 219/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.7201 - accuracy: 0.7440 - val_loss: 8.0490 - val_accuracy: 0.2264\n",
            "Epoch 220/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.7270 - accuracy: 0.7794 - val_loss: 8.0373 - val_accuracy: 0.2264\n",
            "Epoch 221/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.7504 - accuracy: 0.7579 - val_loss: 8.0361 - val_accuracy: 0.2264\n",
            "Epoch 222/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.7322 - accuracy: 0.7492 - val_loss: 8.0392 - val_accuracy: 0.2264\n",
            "Epoch 223/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.7399 - accuracy: 0.7289 - val_loss: 8.0412 - val_accuracy: 0.2075\n",
            "Epoch 224/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.7329 - accuracy: 0.7193 - val_loss: 8.0341 - val_accuracy: 0.2264\n",
            "Epoch 225/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.7106 - accuracy: 0.7472 - val_loss: 8.0364 - val_accuracy: 0.2264\n",
            "Epoch 226/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.7637 - accuracy: 0.7093 - val_loss: 8.0363 - val_accuracy: 0.2264\n",
            "Epoch 227/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 6.6765 - accuracy: 0.7804 - val_loss: 8.0365 - val_accuracy: 0.2075\n",
            "Epoch 228/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.6735 - accuracy: 0.7366 - val_loss: 8.0390 - val_accuracy: 0.2075\n",
            "Epoch 229/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.7106 - accuracy: 0.7623 - val_loss: 8.0303 - val_accuracy: 0.2075\n",
            "Epoch 230/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.6497 - accuracy: 0.7574 - val_loss: 8.0300 - val_accuracy: 0.2075\n",
            "Epoch 231/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.6806 - accuracy: 0.7039 - val_loss: 8.0243 - val_accuracy: 0.1887\n",
            "Epoch 232/2000\n",
            "5/5 [==============================] - 0s 88ms/step - loss: 6.6483 - accuracy: 0.7655 - val_loss: 8.0155 - val_accuracy: 0.1887\n",
            "Epoch 233/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.6922 - accuracy: 0.7128 - val_loss: 8.0131 - val_accuracy: 0.2075\n",
            "Epoch 234/2000\n",
            "5/5 [==============================] - 0s 50ms/step - loss: 6.6595 - accuracy: 0.7189 - val_loss: 8.0010 - val_accuracy: 0.2264\n",
            "Epoch 235/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.6869 - accuracy: 0.7473 - val_loss: 8.0011 - val_accuracy: 0.2453\n",
            "Epoch 236/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.6700 - accuracy: 0.7745 - val_loss: 8.0033 - val_accuracy: 0.2453\n",
            "Epoch 237/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.7102 - accuracy: 0.7539 - val_loss: 8.0000 - val_accuracy: 0.2264\n",
            "Epoch 238/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.6504 - accuracy: 0.7567 - val_loss: 7.9946 - val_accuracy: 0.2453\n",
            "Epoch 239/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.6956 - accuracy: 0.7502 - val_loss: 7.9921 - val_accuracy: 0.2264\n",
            "Epoch 240/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.6148 - accuracy: 0.7620 - val_loss: 7.9837 - val_accuracy: 0.2264\n",
            "Epoch 241/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.6880 - accuracy: 0.7467 - val_loss: 7.9792 - val_accuracy: 0.2264\n",
            "Epoch 242/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.6333 - accuracy: 0.7300 - val_loss: 7.9693 - val_accuracy: 0.2264\n",
            "Epoch 243/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.6618 - accuracy: 0.7493 - val_loss: 7.9542 - val_accuracy: 0.2264\n",
            "Epoch 244/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.6626 - accuracy: 0.7423 - val_loss: 7.9496 - val_accuracy: 0.2075\n",
            "Epoch 245/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.5968 - accuracy: 0.7748 - val_loss: 7.9463 - val_accuracy: 0.2264\n",
            "Epoch 246/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.6359 - accuracy: 0.7297 - val_loss: 7.9467 - val_accuracy: 0.2264\n",
            "Epoch 247/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.6001 - accuracy: 0.7763 - val_loss: 7.9399 - val_accuracy: 0.2264\n",
            "Epoch 248/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.5759 - accuracy: 0.7712 - val_loss: 7.9314 - val_accuracy: 0.2264\n",
            "Epoch 249/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.6443 - accuracy: 0.7578 - val_loss: 7.9302 - val_accuracy: 0.2264\n",
            "Epoch 250/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.5784 - accuracy: 0.7237 - val_loss: 7.9203 - val_accuracy: 0.2264\n",
            "Epoch 251/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.5669 - accuracy: 0.7744 - val_loss: 7.9242 - val_accuracy: 0.2264\n",
            "Epoch 252/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.6209 - accuracy: 0.7371 - val_loss: 7.9274 - val_accuracy: 0.2264\n",
            "Epoch 253/2000\n",
            "5/5 [==============================] - 0s 59ms/step - loss: 6.5545 - accuracy: 0.7657 - val_loss: 7.9203 - val_accuracy: 0.2264\n",
            "Epoch 254/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.6436 - accuracy: 0.7085 - val_loss: 7.9157 - val_accuracy: 0.2264\n",
            "Epoch 255/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.6520 - accuracy: 0.7773 - val_loss: 7.9246 - val_accuracy: 0.2264\n",
            "Epoch 256/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.5742 - accuracy: 0.7452 - val_loss: 7.9207 - val_accuracy: 0.2264\n",
            "Epoch 257/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 6.5636 - accuracy: 0.7964 - val_loss: 7.9176 - val_accuracy: 0.2264\n",
            "Epoch 258/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.5809 - accuracy: 0.7704 - val_loss: 7.9220 - val_accuracy: 0.2453\n",
            "Epoch 259/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.6222 - accuracy: 0.7480 - val_loss: 7.9127 - val_accuracy: 0.2264\n",
            "Epoch 260/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.5738 - accuracy: 0.7597 - val_loss: 7.9129 - val_accuracy: 0.2264\n",
            "Epoch 261/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.5389 - accuracy: 0.7828 - val_loss: 7.9171 - val_accuracy: 0.2264\n",
            "Epoch 262/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.5630 - accuracy: 0.7820 - val_loss: 7.9175 - val_accuracy: 0.2264\n",
            "Epoch 263/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.6102 - accuracy: 0.7380 - val_loss: 7.9199 - val_accuracy: 0.2642\n",
            "Epoch 264/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.5101 - accuracy: 0.8014 - val_loss: 7.9188 - val_accuracy: 0.2642\n",
            "Epoch 265/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.4975 - accuracy: 0.7890 - val_loss: 7.9104 - val_accuracy: 0.2642\n",
            "Epoch 266/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.5398 - accuracy: 0.7669 - val_loss: 7.8959 - val_accuracy: 0.2642\n",
            "Epoch 267/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.5162 - accuracy: 0.7697 - val_loss: 7.8920 - val_accuracy: 0.2453\n",
            "Epoch 268/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.4740 - accuracy: 0.7577 - val_loss: 7.8939 - val_accuracy: 0.2642\n",
            "Epoch 269/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 6.5225 - accuracy: 0.8025 - val_loss: 7.8824 - val_accuracy: 0.2642\n",
            "Epoch 270/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.4880 - accuracy: 0.8100 - val_loss: 7.8753 - val_accuracy: 0.2642\n",
            "Epoch 271/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.5454 - accuracy: 0.7902 - val_loss: 7.8766 - val_accuracy: 0.2642\n",
            "Epoch 272/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.4652 - accuracy: 0.7777 - val_loss: 7.8656 - val_accuracy: 0.2642\n",
            "Epoch 273/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.5619 - accuracy: 0.7250 - val_loss: 7.8593 - val_accuracy: 0.2642\n",
            "Epoch 274/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.4860 - accuracy: 0.7797 - val_loss: 7.8579 - val_accuracy: 0.2642\n",
            "Epoch 275/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.5075 - accuracy: 0.7514 - val_loss: 7.8529 - val_accuracy: 0.2830\n",
            "Epoch 276/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.5331 - accuracy: 0.7548 - val_loss: 7.8461 - val_accuracy: 0.2830\n",
            "Epoch 277/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.5057 - accuracy: 0.7947 - val_loss: 7.8427 - val_accuracy: 0.2830\n",
            "Epoch 278/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.4501 - accuracy: 0.7949 - val_loss: 7.8343 - val_accuracy: 0.2830\n",
            "Epoch 279/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.4926 - accuracy: 0.7909 - val_loss: 7.8396 - val_accuracy: 0.2453\n",
            "Epoch 280/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.5018 - accuracy: 0.7944 - val_loss: 7.8472 - val_accuracy: 0.2453\n",
            "Epoch 281/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.5070 - accuracy: 0.7673 - val_loss: 7.8544 - val_accuracy: 0.2264\n",
            "Epoch 282/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.4661 - accuracy: 0.7763 - val_loss: 7.8445 - val_accuracy: 0.2264\n",
            "Epoch 283/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.5213 - accuracy: 0.7749 - val_loss: 7.8409 - val_accuracy: 0.2453\n",
            "Epoch 284/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.4301 - accuracy: 0.7858 - val_loss: 7.8428 - val_accuracy: 0.2453\n",
            "Epoch 285/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.4680 - accuracy: 0.7712 - val_loss: 7.8409 - val_accuracy: 0.2453\n",
            "Epoch 286/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.5076 - accuracy: 0.7461 - val_loss: 7.8343 - val_accuracy: 0.2830\n",
            "Epoch 287/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.5376 - accuracy: 0.7425 - val_loss: 7.8359 - val_accuracy: 0.2642\n",
            "Epoch 288/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.4317 - accuracy: 0.8151 - val_loss: 7.8273 - val_accuracy: 0.2453\n",
            "Epoch 289/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.4907 - accuracy: 0.7824 - val_loss: 7.8170 - val_accuracy: 0.2453\n",
            "Epoch 290/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.3560 - accuracy: 0.8306 - val_loss: 7.8131 - val_accuracy: 0.2642\n",
            "Epoch 291/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.4855 - accuracy: 0.7516 - val_loss: 7.8166 - val_accuracy: 0.2453\n",
            "Epoch 292/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.4032 - accuracy: 0.8287 - val_loss: 7.8100 - val_accuracy: 0.2453\n",
            "Epoch 293/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.3966 - accuracy: 0.8059 - val_loss: 7.8208 - val_accuracy: 0.2264\n",
            "Epoch 294/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.4356 - accuracy: 0.7906 - val_loss: 7.8160 - val_accuracy: 0.2264\n",
            "Epoch 295/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.4314 - accuracy: 0.7666 - val_loss: 7.8131 - val_accuracy: 0.2642\n",
            "Epoch 296/2000\n",
            "5/5 [==============================] - 0s 78ms/step - loss: 6.4065 - accuracy: 0.8162 - val_loss: 7.8151 - val_accuracy: 0.2264\n",
            "Epoch 297/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.5022 - accuracy: 0.7568 - val_loss: 7.8173 - val_accuracy: 0.2453\n",
            "Epoch 298/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.5142 - accuracy: 0.7791 - val_loss: 7.8145 - val_accuracy: 0.2642\n",
            "Epoch 299/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3439 - accuracy: 0.8239 - val_loss: 7.8170 - val_accuracy: 0.2642\n",
            "Epoch 300/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.4088 - accuracy: 0.7830 - val_loss: 7.8096 - val_accuracy: 0.2642\n",
            "Epoch 301/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.4208 - accuracy: 0.8049 - val_loss: 7.8096 - val_accuracy: 0.2642\n",
            "Epoch 302/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3773 - accuracy: 0.8019 - val_loss: 7.8025 - val_accuracy: 0.2830\n",
            "Epoch 303/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.4622 - accuracy: 0.7525 - val_loss: 7.7951 - val_accuracy: 0.3019\n",
            "Epoch 304/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 6.3826 - accuracy: 0.7833 - val_loss: 7.7920 - val_accuracy: 0.3019\n",
            "Epoch 305/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.3961 - accuracy: 0.7951 - val_loss: 7.7803 - val_accuracy: 0.3019\n",
            "Epoch 306/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.3875 - accuracy: 0.8205 - val_loss: 7.7804 - val_accuracy: 0.3019\n",
            "Epoch 307/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.4113 - accuracy: 0.7894 - val_loss: 7.7736 - val_accuracy: 0.2830\n",
            "Epoch 308/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.3846 - accuracy: 0.8101 - val_loss: 7.7758 - val_accuracy: 0.2642\n",
            "Epoch 309/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3958 - accuracy: 0.7895 - val_loss: 7.7654 - val_accuracy: 0.2642\n",
            "Epoch 310/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.4405 - accuracy: 0.7594 - val_loss: 7.7652 - val_accuracy: 0.2642\n",
            "Epoch 311/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.4289 - accuracy: 0.7402 - val_loss: 7.7671 - val_accuracy: 0.2453\n",
            "Epoch 312/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.3841 - accuracy: 0.7937 - val_loss: 7.7578 - val_accuracy: 0.2453\n",
            "Epoch 313/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.3796 - accuracy: 0.7801 - val_loss: 7.7608 - val_accuracy: 0.2453\n",
            "Epoch 314/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.3598 - accuracy: 0.8071 - val_loss: 7.7561 - val_accuracy: 0.2453\n",
            "Epoch 315/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3605 - accuracy: 0.8166 - val_loss: 7.7663 - val_accuracy: 0.2453\n",
            "Epoch 316/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.3793 - accuracy: 0.7778 - val_loss: 7.7642 - val_accuracy: 0.2453\n",
            "Epoch 317/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.3327 - accuracy: 0.7997 - val_loss: 7.7702 - val_accuracy: 0.2453\n",
            "Epoch 318/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 6.3831 - accuracy: 0.7814 - val_loss: 7.7620 - val_accuracy: 0.2453\n",
            "Epoch 319/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 6.3984 - accuracy: 0.7810 - val_loss: 7.7592 - val_accuracy: 0.2453\n",
            "Epoch 320/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3717 - accuracy: 0.7919 - val_loss: 7.7731 - val_accuracy: 0.2453\n",
            "Epoch 321/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.3048 - accuracy: 0.8187 - val_loss: 7.7779 - val_accuracy: 0.2453\n",
            "Epoch 322/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3662 - accuracy: 0.7980 - val_loss: 7.7755 - val_accuracy: 0.2453\n",
            "Epoch 323/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.4159 - accuracy: 0.7818 - val_loss: 7.7856 - val_accuracy: 0.2453\n",
            "Epoch 324/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.4108 - accuracy: 0.7337 - val_loss: 7.7827 - val_accuracy: 0.2453\n",
            "Epoch 325/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.4151 - accuracy: 0.7783 - val_loss: 7.7884 - val_accuracy: 0.2264\n",
            "Epoch 326/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 6.3625 - accuracy: 0.7622 - val_loss: 7.8039 - val_accuracy: 0.2075\n",
            "Epoch 327/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3567 - accuracy: 0.7586 - val_loss: 7.8018 - val_accuracy: 0.2075\n",
            "Epoch 328/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.3878 - accuracy: 0.7729 - val_loss: 7.8055 - val_accuracy: 0.2075\n",
            "Epoch 329/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 6.3189 - accuracy: 0.8226 - val_loss: 7.7995 - val_accuracy: 0.2075\n",
            "Epoch 330/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3361 - accuracy: 0.7938 - val_loss: 7.7753 - val_accuracy: 0.2075\n",
            "Epoch 331/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.3204 - accuracy: 0.8186 - val_loss: 7.7549 - val_accuracy: 0.2453\n",
            "Epoch 332/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3252 - accuracy: 0.7876 - val_loss: 7.7504 - val_accuracy: 0.2453\n",
            "Epoch 333/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.3354 - accuracy: 0.8013 - val_loss: 7.7545 - val_accuracy: 0.2453\n",
            "Epoch 334/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.3313 - accuracy: 0.7831 - val_loss: 7.7452 - val_accuracy: 0.2453\n",
            "Epoch 335/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.3741 - accuracy: 0.7703 - val_loss: 7.7459 - val_accuracy: 0.2453\n",
            "Epoch 336/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 6.2673 - accuracy: 0.7949 - val_loss: 7.7400 - val_accuracy: 0.2453\n",
            "Epoch 337/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.3851 - accuracy: 0.7405 - val_loss: 7.7447 - val_accuracy: 0.2264\n",
            "Epoch 338/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3018 - accuracy: 0.7956 - val_loss: 7.7395 - val_accuracy: 0.2264\n",
            "Epoch 339/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.3414 - accuracy: 0.7920 - val_loss: 7.7309 - val_accuracy: 0.2264\n",
            "Epoch 340/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3273 - accuracy: 0.7773 - val_loss: 7.7263 - val_accuracy: 0.2642\n",
            "Epoch 341/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3657 - accuracy: 0.7785 - val_loss: 7.7175 - val_accuracy: 0.2642\n",
            "Epoch 342/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2286 - accuracy: 0.8223 - val_loss: 7.7066 - val_accuracy: 0.2453\n",
            "Epoch 343/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.2481 - accuracy: 0.7996 - val_loss: 7.6991 - val_accuracy: 0.2642\n",
            "Epoch 344/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.2701 - accuracy: 0.7937 - val_loss: 7.6858 - val_accuracy: 0.2642\n",
            "Epoch 345/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.3059 - accuracy: 0.7811 - val_loss: 7.6849 - val_accuracy: 0.2642\n",
            "Epoch 346/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 6.2288 - accuracy: 0.8362 - val_loss: 7.6780 - val_accuracy: 0.2830\n",
            "Epoch 347/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 6.2982 - accuracy: 0.8049 - val_loss: 7.6868 - val_accuracy: 0.2830\n",
            "Epoch 348/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.2415 - accuracy: 0.8195 - val_loss: 7.7036 - val_accuracy: 0.2453\n",
            "Epoch 349/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.3302 - accuracy: 0.7631 - val_loss: 7.6961 - val_accuracy: 0.2453\n",
            "Epoch 350/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.2720 - accuracy: 0.7778 - val_loss: 7.6859 - val_accuracy: 0.2453\n",
            "Epoch 351/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2691 - accuracy: 0.8162 - val_loss: 7.6771 - val_accuracy: 0.2075\n",
            "Epoch 352/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.3589 - accuracy: 0.7441 - val_loss: 7.6830 - val_accuracy: 0.2075\n",
            "Epoch 353/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.2499 - accuracy: 0.7904 - val_loss: 7.6748 - val_accuracy: 0.2264\n",
            "Epoch 354/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2834 - accuracy: 0.7999 - val_loss: 7.6755 - val_accuracy: 0.2264\n",
            "Epoch 355/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.2008 - accuracy: 0.8334 - val_loss: 7.6653 - val_accuracy: 0.2453\n",
            "Epoch 356/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 6.2186 - accuracy: 0.8109 - val_loss: 7.6715 - val_accuracy: 0.2264\n",
            "Epoch 357/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.2878 - accuracy: 0.7716 - val_loss: 7.6671 - val_accuracy: 0.2264\n",
            "Epoch 358/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2307 - accuracy: 0.7944 - val_loss: 7.6577 - val_accuracy: 0.2642\n",
            "Epoch 359/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.3112 - accuracy: 0.7820 - val_loss: 7.6642 - val_accuracy: 0.2642\n",
            "Epoch 360/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.2309 - accuracy: 0.8347 - val_loss: 7.6523 - val_accuracy: 0.2830\n",
            "Epoch 361/2000\n",
            "5/5 [==============================] - 0s 78ms/step - loss: 6.2167 - accuracy: 0.8168 - val_loss: 7.6591 - val_accuracy: 0.2453\n",
            "Epoch 362/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2393 - accuracy: 0.8213 - val_loss: 7.6580 - val_accuracy: 0.2642\n",
            "Epoch 363/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.2520 - accuracy: 0.7913 - val_loss: 7.6594 - val_accuracy: 0.2642\n",
            "Epoch 364/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 6.2136 - accuracy: 0.7877 - val_loss: 7.6515 - val_accuracy: 0.2453\n",
            "Epoch 365/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.2110 - accuracy: 0.7842 - val_loss: 7.6431 - val_accuracy: 0.2453\n",
            "Epoch 366/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.2251 - accuracy: 0.8202 - val_loss: 7.6406 - val_accuracy: 0.2453\n",
            "Epoch 367/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2169 - accuracy: 0.8182 - val_loss: 7.6378 - val_accuracy: 0.2453\n",
            "Epoch 368/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 6.2268 - accuracy: 0.7931 - val_loss: 7.6362 - val_accuracy: 0.2642\n",
            "Epoch 369/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.1911 - accuracy: 0.8452 - val_loss: 7.6345 - val_accuracy: 0.2830\n",
            "Epoch 370/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2336 - accuracy: 0.8336 - val_loss: 7.6401 - val_accuracy: 0.2830\n",
            "Epoch 371/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.2512 - accuracy: 0.8078 - val_loss: 7.6486 - val_accuracy: 0.2830\n",
            "Epoch 372/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2398 - accuracy: 0.8061 - val_loss: 7.6414 - val_accuracy: 0.3019\n",
            "Epoch 373/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.2064 - accuracy: 0.7736 - val_loss: 7.6410 - val_accuracy: 0.2830\n",
            "Epoch 374/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2311 - accuracy: 0.8215 - val_loss: 7.6271 - val_accuracy: 0.2830\n",
            "Epoch 375/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2329 - accuracy: 0.7947 - val_loss: 7.6339 - val_accuracy: 0.3019\n",
            "Epoch 376/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2175 - accuracy: 0.8180 - val_loss: 7.6324 - val_accuracy: 0.2830\n",
            "Epoch 377/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.2846 - accuracy: 0.7866 - val_loss: 7.6203 - val_accuracy: 0.2642\n",
            "Epoch 378/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2510 - accuracy: 0.7770 - val_loss: 7.6171 - val_accuracy: 0.3019\n",
            "Epoch 379/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.1842 - accuracy: 0.8242 - val_loss: 7.6042 - val_accuracy: 0.3019\n",
            "Epoch 380/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.1625 - accuracy: 0.8163 - val_loss: 7.5979 - val_accuracy: 0.3019\n",
            "Epoch 381/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2353 - accuracy: 0.7657 - val_loss: 7.5929 - val_accuracy: 0.2830\n",
            "Epoch 382/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.1503 - accuracy: 0.8346 - val_loss: 7.5981 - val_accuracy: 0.3019\n",
            "Epoch 383/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 6.1749 - accuracy: 0.8249 - val_loss: 7.5982 - val_accuracy: 0.3019\n",
            "Epoch 384/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.1929 - accuracy: 0.8308 - val_loss: 7.5970 - val_accuracy: 0.2830\n",
            "Epoch 385/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.1489 - accuracy: 0.8297 - val_loss: 7.5841 - val_accuracy: 0.2642\n",
            "Epoch 386/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.1519 - accuracy: 0.8273 - val_loss: 7.5732 - val_accuracy: 0.2642\n",
            "Epoch 387/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2138 - accuracy: 0.8222 - val_loss: 7.5669 - val_accuracy: 0.2642\n",
            "Epoch 388/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1363 - accuracy: 0.8201 - val_loss: 7.5653 - val_accuracy: 0.2642\n",
            "Epoch 389/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.1684 - accuracy: 0.8282 - val_loss: 7.5709 - val_accuracy: 0.2642\n",
            "Epoch 390/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.2015 - accuracy: 0.8282 - val_loss: 7.5862 - val_accuracy: 0.2642\n",
            "Epoch 391/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1330 - accuracy: 0.8338 - val_loss: 7.5859 - val_accuracy: 0.2453\n",
            "Epoch 392/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1379 - accuracy: 0.8353 - val_loss: 7.5884 - val_accuracy: 0.2453\n",
            "Epoch 393/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.1506 - accuracy: 0.8283 - val_loss: 7.5853 - val_accuracy: 0.2830\n",
            "Epoch 394/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.1878 - accuracy: 0.8047 - val_loss: 7.5846 - val_accuracy: 0.2830\n",
            "Epoch 395/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.2141 - accuracy: 0.7960 - val_loss: 7.5885 - val_accuracy: 0.2830\n",
            "Epoch 396/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.1589 - accuracy: 0.8088 - val_loss: 7.5874 - val_accuracy: 0.2830\n",
            "Epoch 397/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1857 - accuracy: 0.7776 - val_loss: 7.5815 - val_accuracy: 0.2830\n",
            "Epoch 398/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.1471 - accuracy: 0.8208 - val_loss: 7.5684 - val_accuracy: 0.2830\n",
            "Epoch 399/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1096 - accuracy: 0.8428 - val_loss: 7.5682 - val_accuracy: 0.2453\n",
            "Epoch 400/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1279 - accuracy: 0.8213 - val_loss: 7.5822 - val_accuracy: 0.2453\n",
            "Epoch 401/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.0900 - accuracy: 0.8326 - val_loss: 7.5821 - val_accuracy: 0.2453\n",
            "Epoch 402/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1652 - accuracy: 0.8009 - val_loss: 7.5784 - val_accuracy: 0.2453\n",
            "Epoch 403/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1575 - accuracy: 0.7953 - val_loss: 7.5697 - val_accuracy: 0.2453\n",
            "Epoch 404/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1588 - accuracy: 0.7739 - val_loss: 7.5800 - val_accuracy: 0.2642\n",
            "Epoch 405/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.1379 - accuracy: 0.8231 - val_loss: 7.5964 - val_accuracy: 0.2642\n",
            "Epoch 406/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.1404 - accuracy: 0.8115 - val_loss: 7.5865 - val_accuracy: 0.2642\n",
            "Epoch 407/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.1350 - accuracy: 0.8297 - val_loss: 7.5861 - val_accuracy: 0.2453\n",
            "Epoch 408/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.0468 - accuracy: 0.8473 - val_loss: 7.5808 - val_accuracy: 0.2453\n",
            "Epoch 409/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 6.1071 - accuracy: 0.8195 - val_loss: 7.5838 - val_accuracy: 0.2453\n",
            "Epoch 410/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1602 - accuracy: 0.8040 - val_loss: 7.5812 - val_accuracy: 0.2264\n",
            "Epoch 411/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.1048 - accuracy: 0.8390 - val_loss: 7.5699 - val_accuracy: 0.2642\n",
            "Epoch 412/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0530 - accuracy: 0.8644 - val_loss: 7.5602 - val_accuracy: 0.2642\n",
            "Epoch 413/2000\n",
            "5/5 [==============================] - 0s 51ms/step - loss: 6.0866 - accuracy: 0.8257 - val_loss: 7.5637 - val_accuracy: 0.2642\n",
            "Epoch 414/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 6.1009 - accuracy: 0.8401 - val_loss: 7.5613 - val_accuracy: 0.2830\n",
            "Epoch 415/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.1689 - accuracy: 0.8295 - val_loss: 7.5683 - val_accuracy: 0.2830\n",
            "Epoch 416/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1068 - accuracy: 0.7977 - val_loss: 7.5645 - val_accuracy: 0.2642\n",
            "Epoch 417/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.0524 - accuracy: 0.8355 - val_loss: 7.5571 - val_accuracy: 0.2642\n",
            "Epoch 418/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.0690 - accuracy: 0.8346 - val_loss: 7.5538 - val_accuracy: 0.2642\n",
            "Epoch 419/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.1520 - accuracy: 0.7591 - val_loss: 7.5560 - val_accuracy: 0.2830\n",
            "Epoch 420/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0793 - accuracy: 0.8212 - val_loss: 7.5434 - val_accuracy: 0.2642\n",
            "Epoch 421/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 6.0588 - accuracy: 0.8399 - val_loss: 7.5488 - val_accuracy: 0.2830\n",
            "Epoch 422/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.1011 - accuracy: 0.8035 - val_loss: 7.5467 - val_accuracy: 0.2642\n",
            "Epoch 423/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.1455 - accuracy: 0.8212 - val_loss: 7.5419 - val_accuracy: 0.2642\n",
            "Epoch 424/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1057 - accuracy: 0.7967 - val_loss: 7.5418 - val_accuracy: 0.2830\n",
            "Epoch 425/2000\n",
            "5/5 [==============================] - 0s 77ms/step - loss: 6.1331 - accuracy: 0.8095 - val_loss: 7.5412 - val_accuracy: 0.2642\n",
            "Epoch 426/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.1004 - accuracy: 0.8227 - val_loss: 7.5374 - val_accuracy: 0.2642\n",
            "Epoch 427/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.1370 - accuracy: 0.8207 - val_loss: 7.5353 - val_accuracy: 0.2642\n",
            "Epoch 428/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1060 - accuracy: 0.8332 - val_loss: 7.5361 - val_accuracy: 0.2642\n",
            "Epoch 429/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 6.0774 - accuracy: 0.8268 - val_loss: 7.5234 - val_accuracy: 0.2642\n",
            "Epoch 430/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0747 - accuracy: 0.8504 - val_loss: 7.5259 - val_accuracy: 0.2453\n",
            "Epoch 431/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1377 - accuracy: 0.7895 - val_loss: 7.5274 - val_accuracy: 0.2453\n",
            "Epoch 432/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0598 - accuracy: 0.8269 - val_loss: 7.5154 - val_accuracy: 0.2453\n",
            "Epoch 433/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0980 - accuracy: 0.8489 - val_loss: 7.5144 - val_accuracy: 0.2453\n",
            "Epoch 434/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.0957 - accuracy: 0.8459 - val_loss: 7.5120 - val_accuracy: 0.2642\n",
            "Epoch 435/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.0943 - accuracy: 0.8537 - val_loss: 7.5043 - val_accuracy: 0.2642\n",
            "Epoch 436/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.1182 - accuracy: 0.8054 - val_loss: 7.5138 - val_accuracy: 0.2642\n",
            "Epoch 437/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0384 - accuracy: 0.8216 - val_loss: 7.5100 - val_accuracy: 0.2642\n",
            "Epoch 438/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0990 - accuracy: 0.7991 - val_loss: 7.5059 - val_accuracy: 0.2830\n",
            "Epoch 439/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.0119 - accuracy: 0.8532 - val_loss: 7.5030 - val_accuracy: 0.2830\n",
            "Epoch 440/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.1101 - accuracy: 0.7989 - val_loss: 7.5045 - val_accuracy: 0.2830\n",
            "Epoch 441/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.0352 - accuracy: 0.8443 - val_loss: 7.5043 - val_accuracy: 0.2642\n",
            "Epoch 442/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0623 - accuracy: 0.8378 - val_loss: 7.4954 - val_accuracy: 0.2453\n",
            "Epoch 443/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0204 - accuracy: 0.8320 - val_loss: 7.4862 - val_accuracy: 0.2453\n",
            "Epoch 444/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0739 - accuracy: 0.8147 - val_loss: 7.4757 - val_accuracy: 0.2453\n",
            "Epoch 445/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0819 - accuracy: 0.7878 - val_loss: 7.4751 - val_accuracy: 0.2453\n",
            "Epoch 446/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.0675 - accuracy: 0.8393 - val_loss: 7.4711 - val_accuracy: 0.2453\n",
            "Epoch 447/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9996 - accuracy: 0.8332 - val_loss: 7.4613 - val_accuracy: 0.2453\n",
            "Epoch 448/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.0196 - accuracy: 0.8407 - val_loss: 7.4629 - val_accuracy: 0.2453\n",
            "Epoch 449/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9972 - accuracy: 0.8446 - val_loss: 7.4571 - val_accuracy: 0.2642\n",
            "Epoch 450/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.9974 - accuracy: 0.8276 - val_loss: 7.4480 - val_accuracy: 0.2453\n",
            "Epoch 451/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0604 - accuracy: 0.8422 - val_loss: 7.4499 - val_accuracy: 0.2453\n",
            "Epoch 452/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 6.0841 - accuracy: 0.8078 - val_loss: 7.4481 - val_accuracy: 0.2453\n",
            "Epoch 453/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9759 - accuracy: 0.8151 - val_loss: 7.4455 - val_accuracy: 0.2453\n",
            "Epoch 454/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.0682 - accuracy: 0.8300 - val_loss: 7.4356 - val_accuracy: 0.2453\n",
            "Epoch 455/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.0476 - accuracy: 0.8251 - val_loss: 7.4437 - val_accuracy: 0.2453\n",
            "Epoch 456/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.9456 - accuracy: 0.8611 - val_loss: 7.4332 - val_accuracy: 0.2642\n",
            "Epoch 457/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.0247 - accuracy: 0.8186 - val_loss: 7.4360 - val_accuracy: 0.2642\n",
            "Epoch 458/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 6.0006 - accuracy: 0.8697 - val_loss: 7.4303 - val_accuracy: 0.2642\n",
            "Epoch 459/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0811 - accuracy: 0.7983 - val_loss: 7.4229 - val_accuracy: 0.2642\n",
            "Epoch 460/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.1136 - accuracy: 0.7993 - val_loss: 7.4148 - val_accuracy: 0.2830\n",
            "Epoch 461/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.0762 - accuracy: 0.7912 - val_loss: 7.4124 - val_accuracy: 0.2830\n",
            "Epoch 462/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.0035 - accuracy: 0.8541 - val_loss: 7.4106 - val_accuracy: 0.2830\n",
            "Epoch 463/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.9793 - accuracy: 0.8663 - val_loss: 7.4079 - val_accuracy: 0.2830\n",
            "Epoch 464/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.0002 - accuracy: 0.8239 - val_loss: 7.4113 - val_accuracy: 0.2830\n",
            "Epoch 465/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.0241 - accuracy: 0.8425 - val_loss: 7.4155 - val_accuracy: 0.2830\n",
            "Epoch 466/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 6.0131 - accuracy: 0.8348 - val_loss: 7.4221 - val_accuracy: 0.2830\n",
            "Epoch 467/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.9736 - accuracy: 0.8424 - val_loss: 7.4118 - val_accuracy: 0.2642\n",
            "Epoch 468/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.0203 - accuracy: 0.8347 - val_loss: 7.4035 - val_accuracy: 0.2830\n",
            "Epoch 469/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0070 - accuracy: 0.8349 - val_loss: 7.4017 - val_accuracy: 0.2830\n",
            "Epoch 470/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9654 - accuracy: 0.8193 - val_loss: 7.4099 - val_accuracy: 0.2830\n",
            "Epoch 471/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9610 - accuracy: 0.8741 - val_loss: 7.4174 - val_accuracy: 0.2830\n",
            "Epoch 472/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9803 - accuracy: 0.8261 - val_loss: 7.4128 - val_accuracy: 0.2830\n",
            "Epoch 473/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.9564 - accuracy: 0.8550 - val_loss: 7.4230 - val_accuracy: 0.2642\n",
            "Epoch 474/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.9646 - accuracy: 0.8444 - val_loss: 7.4236 - val_accuracy: 0.2642\n",
            "Epoch 475/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9627 - accuracy: 0.8459 - val_loss: 7.4278 - val_accuracy: 0.2642\n",
            "Epoch 476/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0127 - accuracy: 0.8057 - val_loss: 7.4291 - val_accuracy: 0.2830\n",
            "Epoch 477/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.9873 - accuracy: 0.8425 - val_loss: 7.4220 - val_accuracy: 0.2830\n",
            "Epoch 478/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9753 - accuracy: 0.8065 - val_loss: 7.4196 - val_accuracy: 0.2830\n",
            "Epoch 479/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.9872 - accuracy: 0.8506 - val_loss: 7.4167 - val_accuracy: 0.2830\n",
            "Epoch 480/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9871 - accuracy: 0.8364 - val_loss: 7.4161 - val_accuracy: 0.2830\n",
            "Epoch 481/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.9641 - accuracy: 0.8408 - val_loss: 7.4188 - val_accuracy: 0.2642\n",
            "Epoch 482/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9456 - accuracy: 0.8275 - val_loss: 7.4210 - val_accuracy: 0.2642\n",
            "Epoch 483/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9814 - accuracy: 0.8210 - val_loss: 7.4161 - val_accuracy: 0.2830\n",
            "Epoch 484/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 6.0726 - accuracy: 0.7901 - val_loss: 7.4266 - val_accuracy: 0.2830\n",
            "Epoch 485/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9784 - accuracy: 0.8272 - val_loss: 7.4187 - val_accuracy: 0.2830\n",
            "Epoch 486/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9423 - accuracy: 0.8272 - val_loss: 7.4421 - val_accuracy: 0.2453\n",
            "Epoch 487/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9373 - accuracy: 0.8363 - val_loss: 7.4423 - val_accuracy: 0.2453\n",
            "Epoch 488/2000\n",
            "5/5 [==============================] - 0s 48ms/step - loss: 6.0127 - accuracy: 0.8211 - val_loss: 7.4470 - val_accuracy: 0.2075\n",
            "Epoch 489/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9790 - accuracy: 0.8631 - val_loss: 7.4395 - val_accuracy: 0.2264\n",
            "Epoch 490/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 6.0129 - accuracy: 0.8042 - val_loss: 7.4456 - val_accuracy: 0.2075\n",
            "Epoch 491/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 6.0147 - accuracy: 0.8471 - val_loss: 7.4369 - val_accuracy: 0.2264\n",
            "Epoch 492/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.9754 - accuracy: 0.8582 - val_loss: 7.4243 - val_accuracy: 0.2453\n",
            "Epoch 493/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 6.0129 - accuracy: 0.8086 - val_loss: 7.4280 - val_accuracy: 0.2453\n",
            "Epoch 494/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.9612 - accuracy: 0.8587 - val_loss: 7.4163 - val_accuracy: 0.2453\n",
            "Epoch 495/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.9813 - accuracy: 0.8216 - val_loss: 7.4095 - val_accuracy: 0.2453\n",
            "Epoch 496/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.9700 - accuracy: 0.8640 - val_loss: 7.4042 - val_accuracy: 0.2453\n",
            "Epoch 497/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.9852 - accuracy: 0.8349 - val_loss: 7.4018 - val_accuracy: 0.2453\n",
            "Epoch 498/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.9185 - accuracy: 0.8582 - val_loss: 7.3880 - val_accuracy: 0.2453\n",
            "Epoch 499/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9112 - accuracy: 0.8455 - val_loss: 7.3990 - val_accuracy: 0.2453\n",
            "Epoch 500/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.8638 - accuracy: 0.8607 - val_loss: 7.4009 - val_accuracy: 0.2642\n",
            "Epoch 501/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9804 - accuracy: 0.8493 - val_loss: 7.4036 - val_accuracy: 0.2453\n",
            "Epoch 502/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9652 - accuracy: 0.8107 - val_loss: 7.4062 - val_accuracy: 0.2642\n",
            "Epoch 503/2000\n",
            "5/5 [==============================] - 0s 74ms/step - loss: 5.9561 - accuracy: 0.8192 - val_loss: 7.3973 - val_accuracy: 0.2830\n",
            "Epoch 504/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9690 - accuracy: 0.8333 - val_loss: 7.3901 - val_accuracy: 0.2453\n",
            "Epoch 505/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9597 - accuracy: 0.8264 - val_loss: 7.3808 - val_accuracy: 0.2642\n",
            "Epoch 506/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9610 - accuracy: 0.8308 - val_loss: 7.3723 - val_accuracy: 0.2642\n",
            "Epoch 507/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9265 - accuracy: 0.8743 - val_loss: 7.3808 - val_accuracy: 0.2453\n",
            "Epoch 508/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.9539 - accuracy: 0.8495 - val_loss: 7.3822 - val_accuracy: 0.2830\n",
            "Epoch 509/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.8892 - accuracy: 0.8390 - val_loss: 7.3872 - val_accuracy: 0.2642\n",
            "Epoch 510/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.9640 - accuracy: 0.8115 - val_loss: 7.3800 - val_accuracy: 0.2642\n",
            "Epoch 511/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9308 - accuracy: 0.8189 - val_loss: 7.3779 - val_accuracy: 0.2642\n",
            "Epoch 512/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9012 - accuracy: 0.8613 - val_loss: 7.3798 - val_accuracy: 0.2642\n",
            "Epoch 513/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.9185 - accuracy: 0.8348 - val_loss: 7.3722 - val_accuracy: 0.2830\n",
            "Epoch 514/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9389 - accuracy: 0.8077 - val_loss: 7.3698 - val_accuracy: 0.2830\n",
            "Epoch 515/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9355 - accuracy: 0.8315 - val_loss: 7.3772 - val_accuracy: 0.2642\n",
            "Epoch 516/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.9059 - accuracy: 0.8114 - val_loss: 7.3700 - val_accuracy: 0.2642\n",
            "Epoch 517/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.9181 - accuracy: 0.8321 - val_loss: 7.3695 - val_accuracy: 0.2642\n",
            "Epoch 518/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.9332 - accuracy: 0.8314 - val_loss: 7.3617 - val_accuracy: 0.2830\n",
            "Epoch 519/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8405 - accuracy: 0.8769 - val_loss: 7.3635 - val_accuracy: 0.2453\n",
            "Epoch 520/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.9163 - accuracy: 0.8532 - val_loss: 7.3576 - val_accuracy: 0.2642\n",
            "Epoch 521/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9060 - accuracy: 0.8441 - val_loss: 7.3617 - val_accuracy: 0.2453\n",
            "Epoch 522/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.9490 - accuracy: 0.8063 - val_loss: 7.3642 - val_accuracy: 0.2453\n",
            "Epoch 523/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8866 - accuracy: 0.8385 - val_loss: 7.3604 - val_accuracy: 0.2453\n",
            "Epoch 524/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8271 - accuracy: 0.8961 - val_loss: 7.3488 - val_accuracy: 0.2830\n",
            "Epoch 525/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.8975 - accuracy: 0.8264 - val_loss: 7.3488 - val_accuracy: 0.2642\n",
            "Epoch 526/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8949 - accuracy: 0.8323 - val_loss: 7.3534 - val_accuracy: 0.2453\n",
            "Epoch 527/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.9212 - accuracy: 0.8402 - val_loss: 7.3561 - val_accuracy: 0.2453\n",
            "Epoch 528/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.9206 - accuracy: 0.8230 - val_loss: 7.3594 - val_accuracy: 0.2642\n",
            "Epoch 529/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.8975 - accuracy: 0.8505 - val_loss: 7.3605 - val_accuracy: 0.2642\n",
            "Epoch 530/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.9230 - accuracy: 0.8060 - val_loss: 7.3601 - val_accuracy: 0.2642\n",
            "Epoch 531/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9310 - accuracy: 0.7838 - val_loss: 7.3686 - val_accuracy: 0.2642\n",
            "Epoch 532/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8229 - accuracy: 0.8810 - val_loss: 7.3576 - val_accuracy: 0.2830\n",
            "Epoch 533/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.9136 - accuracy: 0.8342 - val_loss: 7.3572 - val_accuracy: 0.2642\n",
            "Epoch 534/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.9208 - accuracy: 0.8268 - val_loss: 7.3482 - val_accuracy: 0.2642\n",
            "Epoch 535/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8879 - accuracy: 0.8609 - val_loss: 7.3528 - val_accuracy: 0.2830\n",
            "Epoch 536/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.9048 - accuracy: 0.8623 - val_loss: 7.3585 - val_accuracy: 0.2830\n",
            "Epoch 537/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.8759 - accuracy: 0.8408 - val_loss: 7.3574 - val_accuracy: 0.2642\n",
            "Epoch 538/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8741 - accuracy: 0.8159 - val_loss: 7.3463 - val_accuracy: 0.3019\n",
            "Epoch 539/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8038 - accuracy: 0.8662 - val_loss: 7.3518 - val_accuracy: 0.3019\n",
            "Epoch 540/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8832 - accuracy: 0.8212 - val_loss: 7.3584 - val_accuracy: 0.3019\n",
            "Epoch 541/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8996 - accuracy: 0.8466 - val_loss: 7.3623 - val_accuracy: 0.2830\n",
            "Epoch 542/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.8487 - accuracy: 0.8708 - val_loss: 7.3557 - val_accuracy: 0.2830\n",
            "Epoch 543/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8700 - accuracy: 0.8285 - val_loss: 7.3506 - val_accuracy: 0.2830\n",
            "Epoch 544/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8756 - accuracy: 0.8273 - val_loss: 7.3573 - val_accuracy: 0.2642\n",
            "Epoch 545/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.8587 - accuracy: 0.8777 - val_loss: 7.3440 - val_accuracy: 0.2642\n",
            "Epoch 546/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8752 - accuracy: 0.8192 - val_loss: 7.3318 - val_accuracy: 0.3019\n",
            "Epoch 547/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.8686 - accuracy: 0.8622 - val_loss: 7.3306 - val_accuracy: 0.3019\n",
            "Epoch 548/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.8605 - accuracy: 0.8461 - val_loss: 7.3251 - val_accuracy: 0.3019\n",
            "Epoch 549/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.8715 - accuracy: 0.8542 - val_loss: 7.3235 - val_accuracy: 0.3019\n",
            "Epoch 550/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8288 - accuracy: 0.8460 - val_loss: 7.3104 - val_accuracy: 0.2830\n",
            "Epoch 551/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8930 - accuracy: 0.8439 - val_loss: 7.3122 - val_accuracy: 0.2830\n",
            "Epoch 552/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7946 - accuracy: 0.8587 - val_loss: 7.3068 - val_accuracy: 0.3019\n",
            "Epoch 553/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.8079 - accuracy: 0.8876 - val_loss: 7.2946 - val_accuracy: 0.3019\n",
            "Epoch 554/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.8697 - accuracy: 0.8337 - val_loss: 7.2986 - val_accuracy: 0.3019\n",
            "Epoch 555/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8709 - accuracy: 0.8371 - val_loss: 7.2866 - val_accuracy: 0.3019\n",
            "Epoch 556/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.8723 - accuracy: 0.8532 - val_loss: 7.2981 - val_accuracy: 0.3208\n",
            "Epoch 557/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8336 - accuracy: 0.8481 - val_loss: 7.2959 - val_accuracy: 0.3208\n",
            "Epoch 558/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8452 - accuracy: 0.8691 - val_loss: 7.3104 - val_accuracy: 0.3019\n",
            "Epoch 559/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8048 - accuracy: 0.8587 - val_loss: 7.3008 - val_accuracy: 0.3019\n",
            "Epoch 560/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8716 - accuracy: 0.8098 - val_loss: 7.2937 - val_accuracy: 0.3019\n",
            "Epoch 561/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8602 - accuracy: 0.8610 - val_loss: 7.2825 - val_accuracy: 0.3396\n",
            "Epoch 562/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7884 - accuracy: 0.8754 - val_loss: 7.2816 - val_accuracy: 0.3396\n",
            "Epoch 563/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.8610 - accuracy: 0.8092 - val_loss: 7.2810 - val_accuracy: 0.3396\n",
            "Epoch 564/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.8416 - accuracy: 0.8345 - val_loss: 7.2820 - val_accuracy: 0.3208\n",
            "Epoch 565/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.8032 - accuracy: 0.8773 - val_loss: 7.2919 - val_accuracy: 0.3208\n",
            "Epoch 566/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.8423 - accuracy: 0.8075 - val_loss: 7.2896 - val_accuracy: 0.3208\n",
            "Epoch 567/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.8695 - accuracy: 0.8329 - val_loss: 7.2918 - val_accuracy: 0.3208\n",
            "Epoch 568/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8416 - accuracy: 0.8520 - val_loss: 7.2913 - val_accuracy: 0.3019\n",
            "Epoch 569/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8217 - accuracy: 0.8679 - val_loss: 7.2894 - val_accuracy: 0.3019\n",
            "Epoch 570/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8606 - accuracy: 0.8318 - val_loss: 7.2899 - val_accuracy: 0.3019\n",
            "Epoch 571/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.8435 - accuracy: 0.8375 - val_loss: 7.2774 - val_accuracy: 0.3019\n",
            "Epoch 572/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.8039 - accuracy: 0.8662 - val_loss: 7.2843 - val_accuracy: 0.2830\n",
            "Epoch 573/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8473 - accuracy: 0.7948 - val_loss: 7.2725 - val_accuracy: 0.2830\n",
            "Epoch 574/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.8262 - accuracy: 0.8705 - val_loss: 7.2706 - val_accuracy: 0.2830\n",
            "Epoch 575/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.8384 - accuracy: 0.8698 - val_loss: 7.2783 - val_accuracy: 0.2830\n",
            "Epoch 576/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.8453 - accuracy: 0.8674 - val_loss: 7.2770 - val_accuracy: 0.2830\n",
            "Epoch 577/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8704 - accuracy: 0.8131 - val_loss: 7.2675 - val_accuracy: 0.2830\n",
            "Epoch 578/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.7513 - accuracy: 0.8795 - val_loss: 7.2551 - val_accuracy: 0.2830\n",
            "Epoch 579/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.8155 - accuracy: 0.8705 - val_loss: 7.2574 - val_accuracy: 0.2642\n",
            "Epoch 580/2000\n",
            "5/5 [==============================] - 0s 79ms/step - loss: 5.8052 - accuracy: 0.8697 - val_loss: 7.2474 - val_accuracy: 0.2642\n",
            "Epoch 581/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7720 - accuracy: 0.8480 - val_loss: 7.2422 - val_accuracy: 0.2642\n",
            "Epoch 582/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.8389 - accuracy: 0.8154 - val_loss: 7.2351 - val_accuracy: 0.2830\n",
            "Epoch 583/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7663 - accuracy: 0.8729 - val_loss: 7.2267 - val_accuracy: 0.2830\n",
            "Epoch 584/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8756 - accuracy: 0.8112 - val_loss: 7.2224 - val_accuracy: 0.3019\n",
            "Epoch 585/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.7647 - accuracy: 0.8716 - val_loss: 7.2247 - val_accuracy: 0.3019\n",
            "Epoch 586/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8111 - accuracy: 0.8413 - val_loss: 7.2179 - val_accuracy: 0.3208\n",
            "Epoch 587/2000\n",
            "5/5 [==============================] - 0s 48ms/step - loss: 5.8471 - accuracy: 0.8266 - val_loss: 7.2107 - val_accuracy: 0.3208\n",
            "Epoch 588/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7755 - accuracy: 0.8471 - val_loss: 7.2047 - val_accuracy: 0.3208\n",
            "Epoch 589/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8040 - accuracy: 0.8689 - val_loss: 7.1991 - val_accuracy: 0.3208\n",
            "Epoch 590/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.8029 - accuracy: 0.8410 - val_loss: 7.1969 - val_accuracy: 0.3208\n",
            "Epoch 591/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.7885 - accuracy: 0.8528 - val_loss: 7.1937 - val_accuracy: 0.3208\n",
            "Epoch 592/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8499 - accuracy: 0.8526 - val_loss: 7.1920 - val_accuracy: 0.3396\n",
            "Epoch 593/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7988 - accuracy: 0.8464 - val_loss: 7.2027 - val_accuracy: 0.3396\n",
            "Epoch 594/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.8667 - accuracy: 0.8268 - val_loss: 7.2069 - val_accuracy: 0.3208\n",
            "Epoch 595/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8313 - accuracy: 0.8414 - val_loss: 7.2105 - val_accuracy: 0.3396\n",
            "Epoch 596/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7697 - accuracy: 0.8666 - val_loss: 7.2065 - val_accuracy: 0.3396\n",
            "Epoch 597/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8227 - accuracy: 0.8294 - val_loss: 7.2070 - val_accuracy: 0.3396\n",
            "Epoch 598/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8079 - accuracy: 0.8468 - val_loss: 7.2087 - val_accuracy: 0.3396\n",
            "Epoch 599/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8075 - accuracy: 0.8440 - val_loss: 7.2168 - val_accuracy: 0.3585\n",
            "Epoch 600/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8004 - accuracy: 0.8357 - val_loss: 7.2136 - val_accuracy: 0.3208\n",
            "Epoch 601/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8761 - accuracy: 0.8239 - val_loss: 7.2245 - val_accuracy: 0.3396\n",
            "Epoch 602/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7842 - accuracy: 0.8580 - val_loss: 7.2323 - val_accuracy: 0.3396\n",
            "Epoch 603/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7547 - accuracy: 0.8791 - val_loss: 7.2370 - val_accuracy: 0.3208\n",
            "Epoch 604/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7897 - accuracy: 0.8435 - val_loss: 7.2485 - val_accuracy: 0.3019\n",
            "Epoch 605/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7795 - accuracy: 0.8458 - val_loss: 7.2331 - val_accuracy: 0.3396\n",
            "Epoch 606/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7468 - accuracy: 0.8615 - val_loss: 7.2279 - val_accuracy: 0.3585\n",
            "Epoch 607/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7363 - accuracy: 0.8555 - val_loss: 7.2188 - val_accuracy: 0.3396\n",
            "Epoch 608/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.7271 - accuracy: 0.8725 - val_loss: 7.2098 - val_accuracy: 0.3396\n",
            "Epoch 609/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8512 - accuracy: 0.8424 - val_loss: 7.2156 - val_accuracy: 0.3208\n",
            "Epoch 610/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.8252 - accuracy: 0.8413 - val_loss: 7.2222 - val_accuracy: 0.3019\n",
            "Epoch 611/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.8210 - accuracy: 0.8080 - val_loss: 7.2242 - val_accuracy: 0.3019\n",
            "Epoch 612/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.7772 - accuracy: 0.8527 - val_loss: 7.2218 - val_accuracy: 0.3019\n",
            "Epoch 613/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.7627 - accuracy: 0.8691 - val_loss: 7.2228 - val_accuracy: 0.3019\n",
            "Epoch 614/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7314 - accuracy: 0.8668 - val_loss: 7.2143 - val_accuracy: 0.3019\n",
            "Epoch 615/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7341 - accuracy: 0.8511 - val_loss: 7.2048 - val_accuracy: 0.3019\n",
            "Epoch 616/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.7029 - accuracy: 0.8538 - val_loss: 7.1969 - val_accuracy: 0.3019\n",
            "Epoch 617/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.8018 - accuracy: 0.8368 - val_loss: 7.2004 - val_accuracy: 0.3208\n",
            "Epoch 618/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.7772 - accuracy: 0.8271 - val_loss: 7.1971 - val_accuracy: 0.3396\n",
            "Epoch 619/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7562 - accuracy: 0.8299 - val_loss: 7.1867 - val_accuracy: 0.3396\n",
            "Epoch 620/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7421 - accuracy: 0.8711 - val_loss: 7.1764 - val_accuracy: 0.3585\n",
            "Epoch 621/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.7873 - accuracy: 0.8277 - val_loss: 7.1820 - val_accuracy: 0.3585\n",
            "Epoch 622/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7223 - accuracy: 0.8764 - val_loss: 7.1781 - val_accuracy: 0.3585\n",
            "Epoch 623/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7199 - accuracy: 0.8565 - val_loss: 7.1878 - val_accuracy: 0.3396\n",
            "Epoch 624/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6718 - accuracy: 0.9239 - val_loss: 7.2004 - val_accuracy: 0.3396\n",
            "Epoch 625/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.7624 - accuracy: 0.8728 - val_loss: 7.1956 - val_accuracy: 0.3396\n",
            "Epoch 626/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7444 - accuracy: 0.8612 - val_loss: 7.1906 - val_accuracy: 0.3208\n",
            "Epoch 627/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7725 - accuracy: 0.8367 - val_loss: 7.2058 - val_accuracy: 0.3019\n",
            "Epoch 628/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.8150 - accuracy: 0.8482 - val_loss: 7.1965 - val_accuracy: 0.2830\n",
            "Epoch 629/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7304 - accuracy: 0.8555 - val_loss: 7.2172 - val_accuracy: 0.3019\n",
            "Epoch 630/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.7230 - accuracy: 0.8550 - val_loss: 7.2220 - val_accuracy: 0.3019\n",
            "Epoch 631/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.7448 - accuracy: 0.8269 - val_loss: 7.2264 - val_accuracy: 0.3019\n",
            "Epoch 632/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7324 - accuracy: 0.8467 - val_loss: 7.2209 - val_accuracy: 0.3019\n",
            "Epoch 633/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.7645 - accuracy: 0.8588 - val_loss: 7.2358 - val_accuracy: 0.3019\n",
            "Epoch 634/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7432 - accuracy: 0.8520 - val_loss: 7.2230 - val_accuracy: 0.3019\n",
            "Epoch 635/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7278 - accuracy: 0.8558 - val_loss: 7.2083 - val_accuracy: 0.3019\n",
            "Epoch 636/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7174 - accuracy: 0.8843 - val_loss: 7.2210 - val_accuracy: 0.3019\n",
            "Epoch 637/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.7475 - accuracy: 0.8597 - val_loss: 7.2105 - val_accuracy: 0.3019\n",
            "Epoch 638/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7568 - accuracy: 0.8453 - val_loss: 7.2364 - val_accuracy: 0.3019\n",
            "Epoch 639/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7837 - accuracy: 0.8175 - val_loss: 7.2267 - val_accuracy: 0.3019\n",
            "Epoch 640/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.7612 - accuracy: 0.8558 - val_loss: 7.2344 - val_accuracy: 0.3019\n",
            "Epoch 641/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7240 - accuracy: 0.8553 - val_loss: 7.2172 - val_accuracy: 0.3019\n",
            "Epoch 642/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7294 - accuracy: 0.8756 - val_loss: 7.2072 - val_accuracy: 0.3019\n",
            "Epoch 643/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.7036 - accuracy: 0.8757 - val_loss: 7.2116 - val_accuracy: 0.3019\n",
            "Epoch 644/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7596 - accuracy: 0.8340 - val_loss: 7.2167 - val_accuracy: 0.3019\n",
            "Epoch 645/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.7695 - accuracy: 0.8752 - val_loss: 7.2353 - val_accuracy: 0.3019\n",
            "Epoch 646/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.7364 - accuracy: 0.8665 - val_loss: 7.2509 - val_accuracy: 0.3019\n",
            "Epoch 647/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6966 - accuracy: 0.8980 - val_loss: 7.2426 - val_accuracy: 0.2830\n",
            "Epoch 648/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7303 - accuracy: 0.8747 - val_loss: 7.2515 - val_accuracy: 0.3019\n",
            "Epoch 649/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.7086 - accuracy: 0.8762 - val_loss: 7.2345 - val_accuracy: 0.3019\n",
            "Epoch 650/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7327 - accuracy: 0.8416 - val_loss: 7.2301 - val_accuracy: 0.3019\n",
            "Epoch 651/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6852 - accuracy: 0.8559 - val_loss: 7.2298 - val_accuracy: 0.3019\n",
            "Epoch 652/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7180 - accuracy: 0.8612 - val_loss: 7.2150 - val_accuracy: 0.3019\n",
            "Epoch 653/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.7659 - accuracy: 0.8097 - val_loss: 7.2007 - val_accuracy: 0.3019\n",
            "Epoch 654/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7057 - accuracy: 0.8531 - val_loss: 7.1838 - val_accuracy: 0.3019\n",
            "Epoch 655/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7032 - accuracy: 0.8570 - val_loss: 7.1712 - val_accuracy: 0.2830\n",
            "Epoch 656/2000\n",
            "5/5 [==============================] - 0s 73ms/step - loss: 5.7419 - accuracy: 0.8536 - val_loss: 7.1803 - val_accuracy: 0.2830\n",
            "Epoch 657/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6935 - accuracy: 0.8205 - val_loss: 7.1841 - val_accuracy: 0.2830\n",
            "Epoch 658/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7359 - accuracy: 0.9051 - val_loss: 7.1750 - val_accuracy: 0.2830\n",
            "Epoch 659/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7003 - accuracy: 0.8556 - val_loss: 7.1850 - val_accuracy: 0.2830\n",
            "Epoch 660/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7191 - accuracy: 0.8591 - val_loss: 7.1841 - val_accuracy: 0.2830\n",
            "Epoch 661/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6846 - accuracy: 0.8664 - val_loss: 7.1908 - val_accuracy: 0.2830\n",
            "Epoch 662/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.7469 - accuracy: 0.8601 - val_loss: 7.1833 - val_accuracy: 0.2830\n",
            "Epoch 663/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.7458 - accuracy: 0.8560 - val_loss: 7.1763 - val_accuracy: 0.2830\n",
            "Epoch 664/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6579 - accuracy: 0.8669 - val_loss: 7.1766 - val_accuracy: 0.2830\n",
            "Epoch 665/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.7603 - accuracy: 0.8360 - val_loss: 7.1789 - val_accuracy: 0.2830\n",
            "Epoch 666/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.6967 - accuracy: 0.8659 - val_loss: 7.1728 - val_accuracy: 0.3019\n",
            "Epoch 667/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7777 - accuracy: 0.8175 - val_loss: 7.1784 - val_accuracy: 0.2830\n",
            "Epoch 668/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6938 - accuracy: 0.8761 - val_loss: 7.2006 - val_accuracy: 0.2830\n",
            "Epoch 669/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.7065 - accuracy: 0.8233 - val_loss: 7.2028 - val_accuracy: 0.2830\n",
            "Epoch 670/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.6844 - accuracy: 0.8514 - val_loss: 7.1907 - val_accuracy: 0.3019\n",
            "Epoch 671/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.6910 - accuracy: 0.8783 - val_loss: 7.2030 - val_accuracy: 0.2830\n",
            "Epoch 672/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6911 - accuracy: 0.8459 - val_loss: 7.1935 - val_accuracy: 0.3019\n",
            "Epoch 673/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.6680 - accuracy: 0.8576 - val_loss: 7.1969 - val_accuracy: 0.3019\n",
            "Epoch 674/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.6544 - accuracy: 0.8846 - val_loss: 7.1982 - val_accuracy: 0.3019\n",
            "Epoch 675/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6390 - accuracy: 0.8992 - val_loss: 7.1944 - val_accuracy: 0.3019\n",
            "Epoch 676/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7127 - accuracy: 0.8542 - val_loss: 7.2026 - val_accuracy: 0.3019\n",
            "Epoch 677/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7104 - accuracy: 0.8540 - val_loss: 7.1996 - val_accuracy: 0.3019\n",
            "Epoch 678/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6079 - accuracy: 0.8977 - val_loss: 7.1977 - val_accuracy: 0.3019\n",
            "Epoch 679/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6869 - accuracy: 0.8636 - val_loss: 7.1999 - val_accuracy: 0.3019\n",
            "Epoch 680/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6943 - accuracy: 0.8342 - val_loss: 7.1999 - val_accuracy: 0.2830\n",
            "Epoch 681/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.6759 - accuracy: 0.8705 - val_loss: 7.2017 - val_accuracy: 0.2642\n",
            "Epoch 682/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6881 - accuracy: 0.8551 - val_loss: 7.1992 - val_accuracy: 0.2830\n",
            "Epoch 683/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.7319 - accuracy: 0.8278 - val_loss: 7.1953 - val_accuracy: 0.2830\n",
            "Epoch 684/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6291 - accuracy: 0.9018 - val_loss: 7.1907 - val_accuracy: 0.3019\n",
            "Epoch 685/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6838 - accuracy: 0.8453 - val_loss: 7.1860 - val_accuracy: 0.3208\n",
            "Epoch 686/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6846 - accuracy: 0.8693 - val_loss: 7.1764 - val_accuracy: 0.3208\n",
            "Epoch 687/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6991 - accuracy: 0.8613 - val_loss: 7.1853 - val_accuracy: 0.3019\n",
            "Epoch 688/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.6723 - accuracy: 0.8537 - val_loss: 7.1903 - val_accuracy: 0.3019\n",
            "Epoch 689/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.7032 - accuracy: 0.8660 - val_loss: 7.2315 - val_accuracy: 0.2453\n",
            "Epoch 690/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6909 - accuracy: 0.8525 - val_loss: 7.2359 - val_accuracy: 0.2453\n",
            "Epoch 691/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6594 - accuracy: 0.8713 - val_loss: 7.2366 - val_accuracy: 0.2642\n",
            "Epoch 692/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6825 - accuracy: 0.8658 - val_loss: 7.2294 - val_accuracy: 0.2642\n",
            "Epoch 693/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.7027 - accuracy: 0.8412 - val_loss: 7.2318 - val_accuracy: 0.2642\n",
            "Epoch 694/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.6470 - accuracy: 0.8617 - val_loss: 7.2158 - val_accuracy: 0.2642\n",
            "Epoch 695/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6654 - accuracy: 0.8704 - val_loss: 7.2059 - val_accuracy: 0.2830\n",
            "Epoch 696/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.7019 - accuracy: 0.8472 - val_loss: 7.1850 - val_accuracy: 0.3019\n",
            "Epoch 697/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6557 - accuracy: 0.8793 - val_loss: 7.1722 - val_accuracy: 0.3019\n",
            "Epoch 698/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6830 - accuracy: 0.8506 - val_loss: 7.1584 - val_accuracy: 0.3396\n",
            "Epoch 699/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6759 - accuracy: 0.8522 - val_loss: 7.1591 - val_accuracy: 0.3208\n",
            "Epoch 700/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6619 - accuracy: 0.8865 - val_loss: 7.1569 - val_accuracy: 0.3208\n",
            "Epoch 701/2000\n",
            "5/5 [==============================] - 0s 47ms/step - loss: 5.6378 - accuracy: 0.8272 - val_loss: 7.1511 - val_accuracy: 0.3396\n",
            "Epoch 702/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.6102 - accuracy: 0.8775 - val_loss: 7.1555 - val_accuracy: 0.3208\n",
            "Epoch 703/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6225 - accuracy: 0.8689 - val_loss: 7.1433 - val_accuracy: 0.3396\n",
            "Epoch 704/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6521 - accuracy: 0.8864 - val_loss: 7.1353 - val_accuracy: 0.3396\n",
            "Epoch 705/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6412 - accuracy: 0.8615 - val_loss: 7.1387 - val_accuracy: 0.3396\n",
            "Epoch 706/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6618 - accuracy: 0.8508 - val_loss: 7.1265 - val_accuracy: 0.3396\n",
            "Epoch 707/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.6548 - accuracy: 0.8924 - val_loss: 7.1362 - val_accuracy: 0.3208\n",
            "Epoch 708/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.6597 - accuracy: 0.8732 - val_loss: 7.1489 - val_accuracy: 0.3208\n",
            "Epoch 709/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6156 - accuracy: 0.8815 - val_loss: 7.1394 - val_accuracy: 0.3208\n",
            "Epoch 710/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6711 - accuracy: 0.8673 - val_loss: 7.1341 - val_accuracy: 0.3208\n",
            "Epoch 711/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.6755 - accuracy: 0.8627 - val_loss: 7.1354 - val_accuracy: 0.3019\n",
            "Epoch 712/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6334 - accuracy: 0.8726 - val_loss: 7.1238 - val_accuracy: 0.3019\n",
            "Epoch 713/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6403 - accuracy: 0.8971 - val_loss: 7.1135 - val_accuracy: 0.3019\n",
            "Epoch 714/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6084 - accuracy: 0.8361 - val_loss: 7.1123 - val_accuracy: 0.3208\n",
            "Epoch 715/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.6363 - accuracy: 0.8727 - val_loss: 7.1077 - val_accuracy: 0.3208\n",
            "Epoch 716/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.6700 - accuracy: 0.8459 - val_loss: 7.1075 - val_accuracy: 0.3019\n",
            "Epoch 717/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6298 - accuracy: 0.8525 - val_loss: 7.1025 - val_accuracy: 0.2830\n",
            "Epoch 718/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.6331 - accuracy: 0.8591 - val_loss: 7.0980 - val_accuracy: 0.2830\n",
            "Epoch 719/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6153 - accuracy: 0.9140 - val_loss: 7.1087 - val_accuracy: 0.2830\n",
            "Epoch 720/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6363 - accuracy: 0.8688 - val_loss: 7.1100 - val_accuracy: 0.3019\n",
            "Epoch 721/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6597 - accuracy: 0.8348 - val_loss: 7.1108 - val_accuracy: 0.3208\n",
            "Epoch 722/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6360 - accuracy: 0.8419 - val_loss: 7.1003 - val_accuracy: 0.3396\n",
            "Epoch 723/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.6121 - accuracy: 0.8824 - val_loss: 7.1009 - val_accuracy: 0.3396\n",
            "Epoch 724/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6575 - accuracy: 0.8273 - val_loss: 7.1061 - val_accuracy: 0.3208\n",
            "Epoch 725/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6187 - accuracy: 0.8678 - val_loss: 7.1095 - val_accuracy: 0.3208\n",
            "Epoch 726/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6373 - accuracy: 0.8668 - val_loss: 7.1026 - val_accuracy: 0.3208\n",
            "Epoch 727/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6153 - accuracy: 0.8972 - val_loss: 7.0922 - val_accuracy: 0.3585\n",
            "Epoch 728/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.6409 - accuracy: 0.8607 - val_loss: 7.0939 - val_accuracy: 0.3585\n",
            "Epoch 729/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6404 - accuracy: 0.8466 - val_loss: 7.0825 - val_accuracy: 0.3396\n",
            "Epoch 730/2000\n",
            "5/5 [==============================] - 0s 75ms/step - loss: 5.5785 - accuracy: 0.9136 - val_loss: 7.0818 - val_accuracy: 0.3396\n",
            "Epoch 731/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.5653 - accuracy: 0.9104 - val_loss: 7.0757 - val_accuracy: 0.3396\n",
            "Epoch 732/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.6684 - accuracy: 0.8615 - val_loss: 7.0670 - val_accuracy: 0.3585\n",
            "Epoch 733/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6854 - accuracy: 0.8536 - val_loss: 7.0650 - val_accuracy: 0.3585\n",
            "Epoch 734/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6166 - accuracy: 0.8774 - val_loss: 7.0669 - val_accuracy: 0.3585\n",
            "Epoch 735/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6178 - accuracy: 0.8650 - val_loss: 7.0695 - val_accuracy: 0.3396\n",
            "Epoch 736/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6486 - accuracy: 0.8655 - val_loss: 7.0765 - val_accuracy: 0.3208\n",
            "Epoch 737/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6006 - accuracy: 0.8838 - val_loss: 7.0758 - val_accuracy: 0.3208\n",
            "Epoch 738/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6409 - accuracy: 0.8487 - val_loss: 7.0663 - val_accuracy: 0.3208\n",
            "Epoch 739/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.6229 - accuracy: 0.8552 - val_loss: 7.0938 - val_accuracy: 0.3585\n",
            "Epoch 740/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6955 - accuracy: 0.8406 - val_loss: 7.0947 - val_accuracy: 0.3585\n",
            "Epoch 741/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5999 - accuracy: 0.8894 - val_loss: 7.0868 - val_accuracy: 0.3396\n",
            "Epoch 742/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5748 - accuracy: 0.8768 - val_loss: 7.0872 - val_accuracy: 0.3396\n",
            "Epoch 743/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5852 - accuracy: 0.8780 - val_loss: 7.0889 - val_accuracy: 0.3208\n",
            "Epoch 744/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6559 - accuracy: 0.8431 - val_loss: 7.0869 - val_accuracy: 0.3396\n",
            "Epoch 745/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6178 - accuracy: 0.8343 - val_loss: 7.0910 - val_accuracy: 0.3396\n",
            "Epoch 746/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6030 - accuracy: 0.8606 - val_loss: 7.0976 - val_accuracy: 0.3396\n",
            "Epoch 747/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5995 - accuracy: 0.8889 - val_loss: 7.0890 - val_accuracy: 0.3396\n",
            "Epoch 748/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5935 - accuracy: 0.8671 - val_loss: 7.0893 - val_accuracy: 0.3396\n",
            "Epoch 749/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.6041 - accuracy: 0.8571 - val_loss: 7.0752 - val_accuracy: 0.3208\n",
            "Epoch 750/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6161 - accuracy: 0.8552 - val_loss: 7.0832 - val_accuracy: 0.3019\n",
            "Epoch 751/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6079 - accuracy: 0.8796 - val_loss: 7.0815 - val_accuracy: 0.3019\n",
            "Epoch 752/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5892 - accuracy: 0.8632 - val_loss: 7.0734 - val_accuracy: 0.3019\n",
            "Epoch 753/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5759 - accuracy: 0.8855 - val_loss: 7.0883 - val_accuracy: 0.3208\n",
            "Epoch 754/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6185 - accuracy: 0.8559 - val_loss: 7.0799 - val_accuracy: 0.3208\n",
            "Epoch 755/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6376 - accuracy: 0.8507 - val_loss: 7.0696 - val_accuracy: 0.3208\n",
            "Epoch 756/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6483 - accuracy: 0.8496 - val_loss: 7.0627 - val_accuracy: 0.3396\n",
            "Epoch 757/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5714 - accuracy: 0.8796 - val_loss: 7.0647 - val_accuracy: 0.3396\n",
            "Epoch 758/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5666 - accuracy: 0.8947 - val_loss: 7.0802 - val_accuracy: 0.3396\n",
            "Epoch 759/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6497 - accuracy: 0.8688 - val_loss: 7.0702 - val_accuracy: 0.3396\n",
            "Epoch 760/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.5700 - accuracy: 0.8820 - val_loss: 7.0779 - val_accuracy: 0.3396\n",
            "Epoch 761/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5434 - accuracy: 0.8849 - val_loss: 7.0903 - val_accuracy: 0.3396\n",
            "Epoch 762/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5335 - accuracy: 0.8795 - val_loss: 7.0866 - val_accuracy: 0.3396\n",
            "Epoch 763/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6085 - accuracy: 0.8786 - val_loss: 7.0748 - val_accuracy: 0.3396\n",
            "Epoch 764/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5682 - accuracy: 0.8636 - val_loss: 7.0794 - val_accuracy: 0.3396\n",
            "Epoch 765/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.6008 - accuracy: 0.8585 - val_loss: 7.0671 - val_accuracy: 0.3396\n",
            "Epoch 766/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5399 - accuracy: 0.8661 - val_loss: 7.0736 - val_accuracy: 0.3396\n",
            "Epoch 767/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.5378 - accuracy: 0.8876 - val_loss: 7.0798 - val_accuracy: 0.3396\n",
            "Epoch 768/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6314 - accuracy: 0.8544 - val_loss: 7.0840 - val_accuracy: 0.3396\n",
            "Epoch 769/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5499 - accuracy: 0.8727 - val_loss: 7.0881 - val_accuracy: 0.3396\n",
            "Epoch 770/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.5653 - accuracy: 0.8659 - val_loss: 7.1053 - val_accuracy: 0.3208\n",
            "Epoch 771/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6365 - accuracy: 0.8497 - val_loss: 7.1117 - val_accuracy: 0.2830\n",
            "Epoch 772/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6321 - accuracy: 0.8526 - val_loss: 7.1155 - val_accuracy: 0.3019\n",
            "Epoch 773/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5449 - accuracy: 0.8594 - val_loss: 7.1133 - val_accuracy: 0.3019\n",
            "Epoch 774/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.6569 - accuracy: 0.8540 - val_loss: 7.1078 - val_accuracy: 0.3019\n",
            "Epoch 775/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5142 - accuracy: 0.9208 - val_loss: 7.0976 - val_accuracy: 0.3019\n",
            "Epoch 776/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6016 - accuracy: 0.8669 - val_loss: 7.1080 - val_accuracy: 0.3019\n",
            "Epoch 777/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.6163 - accuracy: 0.8462 - val_loss: 7.1144 - val_accuracy: 0.3019\n",
            "Epoch 778/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.5579 - accuracy: 0.8802 - val_loss: 7.1106 - val_accuracy: 0.3208\n",
            "Epoch 779/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.6088 - accuracy: 0.8447 - val_loss: 7.1120 - val_accuracy: 0.3019\n",
            "Epoch 780/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5932 - accuracy: 0.8702 - val_loss: 7.1354 - val_accuracy: 0.2642\n",
            "Epoch 781/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5741 - accuracy: 0.8452 - val_loss: 7.1260 - val_accuracy: 0.2830\n",
            "Epoch 782/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5723 - accuracy: 0.8502 - val_loss: 7.1300 - val_accuracy: 0.2830\n",
            "Epoch 783/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5376 - accuracy: 0.8885 - val_loss: 7.1473 - val_accuracy: 0.2642\n",
            "Epoch 784/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5937 - accuracy: 0.8746 - val_loss: 7.1333 - val_accuracy: 0.3019\n",
            "Epoch 785/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.5534 - accuracy: 0.8700 - val_loss: 7.1316 - val_accuracy: 0.3019\n",
            "Epoch 786/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5729 - accuracy: 0.8557 - val_loss: 7.1159 - val_accuracy: 0.2830\n",
            "Epoch 787/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5969 - accuracy: 0.8592 - val_loss: 7.1137 - val_accuracy: 0.2830\n",
            "Epoch 788/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5687 - accuracy: 0.8694 - val_loss: 7.1152 - val_accuracy: 0.2830\n",
            "Epoch 789/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.6029 - accuracy: 0.8488 - val_loss: 7.1243 - val_accuracy: 0.2830\n",
            "Epoch 790/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.5123 - accuracy: 0.9039 - val_loss: 7.1065 - val_accuracy: 0.2830\n",
            "Epoch 791/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5959 - accuracy: 0.8432 - val_loss: 7.0912 - val_accuracy: 0.3019\n",
            "Epoch 792/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5693 - accuracy: 0.8580 - val_loss: 7.0917 - val_accuracy: 0.3019\n",
            "Epoch 793/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.4886 - accuracy: 0.8899 - val_loss: 7.0842 - val_accuracy: 0.3019\n",
            "Epoch 794/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5627 - accuracy: 0.8398 - val_loss: 7.0897 - val_accuracy: 0.3019\n",
            "Epoch 795/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5662 - accuracy: 0.8961 - val_loss: 7.0788 - val_accuracy: 0.3019\n",
            "Epoch 796/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5670 - accuracy: 0.8749 - val_loss: 7.0756 - val_accuracy: 0.3019\n",
            "Epoch 797/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5491 - accuracy: 0.8740 - val_loss: 7.0745 - val_accuracy: 0.3019\n",
            "Epoch 798/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5554 - accuracy: 0.8508 - val_loss: 7.0713 - val_accuracy: 0.3019\n",
            "Epoch 799/2000\n",
            "5/5 [==============================] - 0s 76ms/step - loss: 5.5714 - accuracy: 0.8332 - val_loss: 7.0901 - val_accuracy: 0.2830\n",
            "Epoch 800/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.5674 - accuracy: 0.8491 - val_loss: 7.0865 - val_accuracy: 0.2830\n",
            "Epoch 801/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5996 - accuracy: 0.8625 - val_loss: 7.0721 - val_accuracy: 0.2830\n",
            "Epoch 802/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5035 - accuracy: 0.8964 - val_loss: 7.0751 - val_accuracy: 0.2830\n",
            "Epoch 803/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5522 - accuracy: 0.8811 - val_loss: 7.0789 - val_accuracy: 0.2830\n",
            "Epoch 804/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.5981 - accuracy: 0.8268 - val_loss: 7.0635 - val_accuracy: 0.2830\n",
            "Epoch 805/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.5256 - accuracy: 0.8772 - val_loss: 7.0613 - val_accuracy: 0.2830\n",
            "Epoch 806/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5019 - accuracy: 0.9013 - val_loss: 7.0491 - val_accuracy: 0.3019\n",
            "Epoch 807/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.5107 - accuracy: 0.8841 - val_loss: 7.0404 - val_accuracy: 0.3208\n",
            "Epoch 808/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.5455 - accuracy: 0.8639 - val_loss: 7.0435 - val_accuracy: 0.3208\n",
            "Epoch 809/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5880 - accuracy: 0.8593 - val_loss: 7.0362 - val_accuracy: 0.3396\n",
            "Epoch 810/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5154 - accuracy: 0.8818 - val_loss: 7.0339 - val_accuracy: 0.3396\n",
            "Epoch 811/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.5645 - accuracy: 0.8245 - val_loss: 7.0230 - val_accuracy: 0.3396\n",
            "Epoch 812/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4861 - accuracy: 0.9227 - val_loss: 7.0134 - val_accuracy: 0.3396\n",
            "Epoch 813/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.4971 - accuracy: 0.9098 - val_loss: 6.9997 - val_accuracy: 0.3396\n",
            "Epoch 814/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.4996 - accuracy: 0.8990 - val_loss: 6.9931 - val_accuracy: 0.3396\n",
            "Epoch 815/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.5658 - accuracy: 0.8609 - val_loss: 6.9878 - val_accuracy: 0.3396\n",
            "Epoch 816/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5336 - accuracy: 0.8987 - val_loss: 6.9923 - val_accuracy: 0.3396\n",
            "Epoch 817/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5592 - accuracy: 0.8426 - val_loss: 7.0153 - val_accuracy: 0.3396\n",
            "Epoch 818/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5171 - accuracy: 0.8657 - val_loss: 7.0026 - val_accuracy: 0.3396\n",
            "Epoch 819/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5494 - accuracy: 0.8736 - val_loss: 7.0168 - val_accuracy: 0.3208\n",
            "Epoch 820/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5302 - accuracy: 0.8777 - val_loss: 7.0168 - val_accuracy: 0.3208\n",
            "Epoch 821/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.5237 - accuracy: 0.9015 - val_loss: 7.0005 - val_accuracy: 0.3208\n",
            "Epoch 822/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.5737 - accuracy: 0.8718 - val_loss: 6.9958 - val_accuracy: 0.3396\n",
            "Epoch 823/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4962 - accuracy: 0.8831 - val_loss: 6.9992 - val_accuracy: 0.3396\n",
            "Epoch 824/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5498 - accuracy: 0.8697 - val_loss: 7.0054 - val_accuracy: 0.3396\n",
            "Epoch 825/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5658 - accuracy: 0.8529 - val_loss: 6.9968 - val_accuracy: 0.3396\n",
            "Epoch 826/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5313 - accuracy: 0.8645 - val_loss: 6.9966 - val_accuracy: 0.3396\n",
            "Epoch 827/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.5400 - accuracy: 0.8876 - val_loss: 6.9941 - val_accuracy: 0.3396\n",
            "Epoch 828/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5446 - accuracy: 0.8779 - val_loss: 6.9960 - val_accuracy: 0.3396\n",
            "Epoch 829/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5026 - accuracy: 0.8706 - val_loss: 6.9909 - val_accuracy: 0.3396\n",
            "Epoch 830/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5293 - accuracy: 0.8905 - val_loss: 6.9821 - val_accuracy: 0.3396\n",
            "Epoch 831/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.5087 - accuracy: 0.9089 - val_loss: 6.9792 - val_accuracy: 0.3396\n",
            "Epoch 832/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5692 - accuracy: 0.8592 - val_loss: 6.9706 - val_accuracy: 0.3585\n",
            "Epoch 833/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4978 - accuracy: 0.8799 - val_loss: 6.9846 - val_accuracy: 0.3585\n",
            "Epoch 834/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5117 - accuracy: 0.8965 - val_loss: 6.9789 - val_accuracy: 0.3208\n",
            "Epoch 835/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5226 - accuracy: 0.8820 - val_loss: 6.9763 - val_accuracy: 0.3208\n",
            "Epoch 836/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5635 - accuracy: 0.8428 - val_loss: 6.9778 - val_accuracy: 0.3208\n",
            "Epoch 837/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5480 - accuracy: 0.8989 - val_loss: 6.9682 - val_accuracy: 0.3208\n",
            "Epoch 838/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.5016 - accuracy: 0.9159 - val_loss: 6.9735 - val_accuracy: 0.3396\n",
            "Epoch 839/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5228 - accuracy: 0.8942 - val_loss: 6.9716 - val_accuracy: 0.3396\n",
            "Epoch 840/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5403 - accuracy: 0.8626 - val_loss: 6.9656 - val_accuracy: 0.3208\n",
            "Epoch 841/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5231 - accuracy: 0.8580 - val_loss: 6.9740 - val_accuracy: 0.3208\n",
            "Epoch 842/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5182 - accuracy: 0.8706 - val_loss: 6.9662 - val_accuracy: 0.3396\n",
            "Epoch 843/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.5384 - accuracy: 0.8722 - val_loss: 6.9595 - val_accuracy: 0.3396\n",
            "Epoch 844/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.5207 - accuracy: 0.8599 - val_loss: 6.9643 - val_accuracy: 0.3396\n",
            "Epoch 845/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5201 - accuracy: 0.8749 - val_loss: 6.9731 - val_accuracy: 0.3396\n",
            "Epoch 846/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5292 - accuracy: 0.8594 - val_loss: 6.9817 - val_accuracy: 0.3396\n",
            "Epoch 847/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5684 - accuracy: 0.8570 - val_loss: 6.9909 - val_accuracy: 0.3208\n",
            "Epoch 848/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4588 - accuracy: 0.9014 - val_loss: 6.9830 - val_accuracy: 0.3208\n",
            "Epoch 849/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5291 - accuracy: 0.8689 - val_loss: 6.9962 - val_accuracy: 0.3208\n",
            "Epoch 850/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5189 - accuracy: 0.9049 - val_loss: 6.9840 - val_accuracy: 0.3208\n",
            "Epoch 851/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5194 - accuracy: 0.8681 - val_loss: 7.0028 - val_accuracy: 0.3396\n",
            "Epoch 852/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4767 - accuracy: 0.8847 - val_loss: 6.9937 - val_accuracy: 0.3396\n",
            "Epoch 853/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5342 - accuracy: 0.8527 - val_loss: 6.9958 - val_accuracy: 0.3396\n",
            "Epoch 854/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.5104 - accuracy: 0.8961 - val_loss: 6.9967 - val_accuracy: 0.3396\n",
            "Epoch 855/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4981 - accuracy: 0.8794 - val_loss: 6.9911 - val_accuracy: 0.3396\n",
            "Epoch 856/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5393 - accuracy: 0.8521 - val_loss: 6.9836 - val_accuracy: 0.3585\n",
            "Epoch 857/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5405 - accuracy: 0.8836 - val_loss: 6.9771 - val_accuracy: 0.3585\n",
            "Epoch 858/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4828 - accuracy: 0.8873 - val_loss: 7.0086 - val_accuracy: 0.3396\n",
            "Epoch 859/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.4789 - accuracy: 0.8828 - val_loss: 6.9962 - val_accuracy: 0.3396\n",
            "Epoch 860/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.4753 - accuracy: 0.8939 - val_loss: 6.9831 - val_accuracy: 0.3208\n",
            "Epoch 861/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.5598 - accuracy: 0.8378 - val_loss: 6.9821 - val_accuracy: 0.3208\n",
            "Epoch 862/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4536 - accuracy: 0.8689 - val_loss: 6.9729 - val_accuracy: 0.3396\n",
            "Epoch 863/2000\n",
            "5/5 [==============================] - 0s 76ms/step - loss: 5.4421 - accuracy: 0.9075 - val_loss: 6.9628 - val_accuracy: 0.3396\n",
            "Epoch 864/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.5220 - accuracy: 0.8631 - val_loss: 6.9548 - val_accuracy: 0.3396\n",
            "Epoch 865/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.4665 - accuracy: 0.9040 - val_loss: 6.9522 - val_accuracy: 0.3396\n",
            "Epoch 866/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5421 - accuracy: 0.8647 - val_loss: 6.9458 - val_accuracy: 0.3396\n",
            "Epoch 867/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.4820 - accuracy: 0.9015 - val_loss: 6.9383 - val_accuracy: 0.3396\n",
            "Epoch 868/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4363 - accuracy: 0.8969 - val_loss: 6.9373 - val_accuracy: 0.3396\n",
            "Epoch 869/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5099 - accuracy: 0.8945 - val_loss: 6.9457 - val_accuracy: 0.3396\n",
            "Epoch 870/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.4839 - accuracy: 0.8739 - val_loss: 6.9435 - val_accuracy: 0.3396\n",
            "Epoch 871/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4456 - accuracy: 0.9102 - val_loss: 6.9551 - val_accuracy: 0.3396\n",
            "Epoch 872/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.5241 - accuracy: 0.8765 - val_loss: 6.9490 - val_accuracy: 0.3396\n",
            "Epoch 873/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4951 - accuracy: 0.8971 - val_loss: 6.9375 - val_accuracy: 0.3396\n",
            "Epoch 874/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4756 - accuracy: 0.8947 - val_loss: 6.9272 - val_accuracy: 0.3396\n",
            "Epoch 875/2000\n",
            "5/5 [==============================] - 0s 52ms/step - loss: 5.5239 - accuracy: 0.8835 - val_loss: 6.9336 - val_accuracy: 0.3396\n",
            "Epoch 876/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4919 - accuracy: 0.8698 - val_loss: 6.9377 - val_accuracy: 0.3208\n",
            "Epoch 877/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4993 - accuracy: 0.8854 - val_loss: 6.9328 - val_accuracy: 0.3208\n",
            "Epoch 878/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4649 - accuracy: 0.8831 - val_loss: 6.9365 - val_accuracy: 0.3208\n",
            "Epoch 879/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4979 - accuracy: 0.8735 - val_loss: 6.9390 - val_accuracy: 0.3396\n",
            "Epoch 880/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4610 - accuracy: 0.8968 - val_loss: 6.9354 - val_accuracy: 0.3396\n",
            "Epoch 881/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4095 - accuracy: 0.9433 - val_loss: 6.9394 - val_accuracy: 0.3396\n",
            "Epoch 882/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4951 - accuracy: 0.8989 - val_loss: 6.9562 - val_accuracy: 0.3208\n",
            "Epoch 883/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.5320 - accuracy: 0.8675 - val_loss: 6.9595 - val_accuracy: 0.3208\n",
            "Epoch 884/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4930 - accuracy: 0.8964 - val_loss: 6.9626 - val_accuracy: 0.3208\n",
            "Epoch 885/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4961 - accuracy: 0.8613 - val_loss: 6.9566 - val_accuracy: 0.3208\n",
            "Epoch 886/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.5171 - accuracy: 0.8663 - val_loss: 6.9614 - val_accuracy: 0.3208\n",
            "Epoch 887/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4211 - accuracy: 0.9265 - val_loss: 6.9692 - val_accuracy: 0.3208\n",
            "Epoch 888/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4942 - accuracy: 0.8483 - val_loss: 6.9695 - val_accuracy: 0.3208\n",
            "Epoch 889/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4889 - accuracy: 0.8858 - val_loss: 6.9524 - val_accuracy: 0.3208\n",
            "Epoch 890/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.4860 - accuracy: 0.8876 - val_loss: 6.9634 - val_accuracy: 0.3208\n",
            "Epoch 891/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4277 - accuracy: 0.8936 - val_loss: 6.9741 - val_accuracy: 0.3396\n",
            "Epoch 892/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4510 - accuracy: 0.8675 - val_loss: 6.9610 - val_accuracy: 0.3396\n",
            "Epoch 893/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4753 - accuracy: 0.8679 - val_loss: 6.9605 - val_accuracy: 0.3396\n",
            "Epoch 894/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4885 - accuracy: 0.8848 - val_loss: 6.9552 - val_accuracy: 0.3396\n",
            "Epoch 895/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4528 - accuracy: 0.8980 - val_loss: 6.9605 - val_accuracy: 0.3396\n",
            "Epoch 896/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.4985 - accuracy: 0.8801 - val_loss: 6.9611 - val_accuracy: 0.3396\n",
            "Epoch 897/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.4454 - accuracy: 0.8756 - val_loss: 6.9646 - val_accuracy: 0.3396\n",
            "Epoch 898/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4571 - accuracy: 0.8733 - val_loss: 6.9919 - val_accuracy: 0.3019\n",
            "Epoch 899/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4777 - accuracy: 0.8738 - val_loss: 6.9939 - val_accuracy: 0.3019\n",
            "Epoch 900/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4536 - accuracy: 0.8683 - val_loss: 6.9819 - val_accuracy: 0.3019\n",
            "Epoch 901/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4223 - accuracy: 0.8953 - val_loss: 6.9891 - val_accuracy: 0.3019\n",
            "Epoch 902/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4683 - accuracy: 0.9089 - val_loss: 6.9704 - val_accuracy: 0.3019\n",
            "Epoch 903/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4142 - accuracy: 0.8968 - val_loss: 6.9532 - val_accuracy: 0.3019\n",
            "Epoch 904/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4648 - accuracy: 0.8811 - val_loss: 6.9425 - val_accuracy: 0.3396\n",
            "Epoch 905/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.4911 - accuracy: 0.8794 - val_loss: 6.9293 - val_accuracy: 0.3396\n",
            "Epoch 906/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4775 - accuracy: 0.8983 - val_loss: 6.9336 - val_accuracy: 0.3585\n",
            "Epoch 907/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.4626 - accuracy: 0.8798 - val_loss: 6.9375 - val_accuracy: 0.3396\n",
            "Epoch 908/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4504 - accuracy: 0.8966 - val_loss: 6.9373 - val_accuracy: 0.3396\n",
            "Epoch 909/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4571 - accuracy: 0.9056 - val_loss: 6.9272 - val_accuracy: 0.3396\n",
            "Epoch 910/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3872 - accuracy: 0.9064 - val_loss: 6.9429 - val_accuracy: 0.3396\n",
            "Epoch 911/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4666 - accuracy: 0.8616 - val_loss: 6.9495 - val_accuracy: 0.3208\n",
            "Epoch 912/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4868 - accuracy: 0.8831 - val_loss: 6.9448 - val_accuracy: 0.3208\n",
            "Epoch 913/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4698 - accuracy: 0.8684 - val_loss: 6.9482 - val_accuracy: 0.3396\n",
            "Epoch 914/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4255 - accuracy: 0.8779 - val_loss: 6.9406 - val_accuracy: 0.3396\n",
            "Epoch 915/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.4320 - accuracy: 0.9112 - val_loss: 6.9314 - val_accuracy: 0.3396\n",
            "Epoch 916/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4524 - accuracy: 0.8828 - val_loss: 6.9257 - val_accuracy: 0.3396\n",
            "Epoch 917/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.4201 - accuracy: 0.8802 - val_loss: 6.9203 - val_accuracy: 0.3396\n",
            "Epoch 918/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.4440 - accuracy: 0.8961 - val_loss: 6.9193 - val_accuracy: 0.3396\n",
            "Epoch 919/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4362 - accuracy: 0.8801 - val_loss: 6.9252 - val_accuracy: 0.3396\n",
            "Epoch 920/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4199 - accuracy: 0.8945 - val_loss: 6.9150 - val_accuracy: 0.3396\n",
            "Epoch 921/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4569 - accuracy: 0.8902 - val_loss: 6.9159 - val_accuracy: 0.3396\n",
            "Epoch 922/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.3776 - accuracy: 0.8901 - val_loss: 6.9309 - val_accuracy: 0.3396\n",
            "Epoch 923/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4521 - accuracy: 0.8765 - val_loss: 6.9376 - val_accuracy: 0.3396\n",
            "Epoch 924/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.4290 - accuracy: 0.8977 - val_loss: 6.9313 - val_accuracy: 0.3396\n",
            "Epoch 925/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4582 - accuracy: 0.8259 - val_loss: 6.9380 - val_accuracy: 0.3396\n",
            "Epoch 926/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.4487 - accuracy: 0.8822 - val_loss: 6.9417 - val_accuracy: 0.3208\n",
            "Epoch 927/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4891 - accuracy: 0.8590 - val_loss: 6.9372 - val_accuracy: 0.3208\n",
            "Epoch 928/2000\n",
            "5/5 [==============================] - 0s 74ms/step - loss: 5.4359 - accuracy: 0.8849 - val_loss: 6.9352 - val_accuracy: 0.3208\n",
            "Epoch 929/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4169 - accuracy: 0.9089 - val_loss: 6.9282 - val_accuracy: 0.3396\n",
            "Epoch 930/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4000 - accuracy: 0.9035 - val_loss: 6.9180 - val_accuracy: 0.3396\n",
            "Epoch 931/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4796 - accuracy: 0.8878 - val_loss: 6.9162 - val_accuracy: 0.3396\n",
            "Epoch 932/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4051 - accuracy: 0.9151 - val_loss: 6.9224 - val_accuracy: 0.3208\n",
            "Epoch 933/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.4194 - accuracy: 0.8831 - val_loss: 6.9253 - val_accuracy: 0.3396\n",
            "Epoch 934/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.4156 - accuracy: 0.8985 - val_loss: 6.9196 - val_accuracy: 0.3208\n",
            "Epoch 935/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4534 - accuracy: 0.8812 - val_loss: 6.9200 - val_accuracy: 0.3208\n",
            "Epoch 936/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4232 - accuracy: 0.8578 - val_loss: 6.9049 - val_accuracy: 0.3208\n",
            "Epoch 937/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.4558 - accuracy: 0.8847 - val_loss: 6.8922 - val_accuracy: 0.3396\n",
            "Epoch 938/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.4225 - accuracy: 0.8978 - val_loss: 6.8815 - val_accuracy: 0.3396\n",
            "Epoch 939/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4001 - accuracy: 0.8818 - val_loss: 6.8717 - val_accuracy: 0.3396\n",
            "Epoch 940/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.4284 - accuracy: 0.9103 - val_loss: 6.8628 - val_accuracy: 0.3585\n",
            "Epoch 941/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.3774 - accuracy: 0.9050 - val_loss: 6.8662 - val_accuracy: 0.3585\n",
            "Epoch 942/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4250 - accuracy: 0.8745 - val_loss: 6.8601 - val_accuracy: 0.3774\n",
            "Epoch 943/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4323 - accuracy: 0.9000 - val_loss: 6.8661 - val_accuracy: 0.3396\n",
            "Epoch 944/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3874 - accuracy: 0.9038 - val_loss: 6.8545 - val_accuracy: 0.3585\n",
            "Epoch 945/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4460 - accuracy: 0.8744 - val_loss: 6.8467 - val_accuracy: 0.3774\n",
            "Epoch 946/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3941 - accuracy: 0.8968 - val_loss: 6.8383 - val_accuracy: 0.3585\n",
            "Epoch 947/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.3700 - accuracy: 0.9096 - val_loss: 6.8335 - val_accuracy: 0.3585\n",
            "Epoch 948/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4280 - accuracy: 0.8752 - val_loss: 6.8263 - val_accuracy: 0.3774\n",
            "Epoch 949/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4205 - accuracy: 0.9068 - val_loss: 6.8214 - val_accuracy: 0.3774\n",
            "Epoch 950/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3790 - accuracy: 0.8649 - val_loss: 6.8268 - val_accuracy: 0.3585\n",
            "Epoch 951/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.4201 - accuracy: 0.8943 - val_loss: 6.8218 - val_accuracy: 0.3585\n",
            "Epoch 952/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4207 - accuracy: 0.8871 - val_loss: 6.8178 - val_accuracy: 0.3585\n",
            "Epoch 953/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.4191 - accuracy: 0.8904 - val_loss: 6.8110 - val_accuracy: 0.3396\n",
            "Epoch 954/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.4457 - accuracy: 0.8654 - val_loss: 6.8097 - val_accuracy: 0.3585\n",
            "Epoch 955/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.4234 - accuracy: 0.8961 - val_loss: 6.8059 - val_accuracy: 0.3585\n",
            "Epoch 956/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4228 - accuracy: 0.9090 - val_loss: 6.8112 - val_accuracy: 0.3585\n",
            "Epoch 957/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3770 - accuracy: 0.9207 - val_loss: 6.8128 - val_accuracy: 0.3585\n",
            "Epoch 958/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4443 - accuracy: 0.9071 - val_loss: 6.8213 - val_accuracy: 0.3774\n",
            "Epoch 959/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.3869 - accuracy: 0.8956 - val_loss: 6.8234 - val_accuracy: 0.3774\n",
            "Epoch 960/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4393 - accuracy: 0.8844 - val_loss: 6.8188 - val_accuracy: 0.3774\n",
            "Epoch 961/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4151 - accuracy: 0.8924 - val_loss: 6.8240 - val_accuracy: 0.3585\n",
            "Epoch 962/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3656 - accuracy: 0.9138 - val_loss: 6.8180 - val_accuracy: 0.3585\n",
            "Epoch 963/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.3955 - accuracy: 0.8841 - val_loss: 6.8249 - val_accuracy: 0.3585\n",
            "Epoch 964/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.4119 - accuracy: 0.8836 - val_loss: 6.8329 - val_accuracy: 0.3585\n",
            "Epoch 965/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4097 - accuracy: 0.8792 - val_loss: 6.8404 - val_accuracy: 0.3585\n",
            "Epoch 966/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4007 - accuracy: 0.8752 - val_loss: 6.8514 - val_accuracy: 0.3585\n",
            "Epoch 967/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3955 - accuracy: 0.9163 - val_loss: 6.8689 - val_accuracy: 0.3585\n",
            "Epoch 968/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4254 - accuracy: 0.8696 - val_loss: 6.8584 - val_accuracy: 0.3585\n",
            "Epoch 969/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4192 - accuracy: 0.8902 - val_loss: 6.8517 - val_accuracy: 0.3585\n",
            "Epoch 970/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3890 - accuracy: 0.8734 - val_loss: 6.8473 - val_accuracy: 0.3585\n",
            "Epoch 971/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4358 - accuracy: 0.8815 - val_loss: 6.8480 - val_accuracy: 0.3585\n",
            "Epoch 972/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3458 - accuracy: 0.8938 - val_loss: 6.8571 - val_accuracy: 0.3585\n",
            "Epoch 973/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.4497 - accuracy: 0.8651 - val_loss: 6.8524 - val_accuracy: 0.3585\n",
            "Epoch 974/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3761 - accuracy: 0.8857 - val_loss: 6.8609 - val_accuracy: 0.3585\n",
            "Epoch 975/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.4245 - accuracy: 0.8765 - val_loss: 6.8677 - val_accuracy: 0.3585\n",
            "Epoch 976/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.3764 - accuracy: 0.8874 - val_loss: 6.8565 - val_accuracy: 0.3585\n",
            "Epoch 977/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3823 - accuracy: 0.8772 - val_loss: 6.8615 - val_accuracy: 0.3585\n",
            "Epoch 978/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.4578 - accuracy: 0.8632 - val_loss: 6.8670 - val_accuracy: 0.3585\n",
            "Epoch 979/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.4004 - accuracy: 0.9087 - val_loss: 6.8767 - val_accuracy: 0.3585\n",
            "Epoch 980/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3894 - accuracy: 0.9026 - val_loss: 6.8726 - val_accuracy: 0.3585\n",
            "Epoch 981/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3943 - accuracy: 0.8886 - val_loss: 6.8673 - val_accuracy: 0.3396\n",
            "Epoch 982/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3718 - accuracy: 0.9168 - val_loss: 6.8692 - val_accuracy: 0.3396\n",
            "Epoch 983/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4195 - accuracy: 0.9074 - val_loss: 6.8919 - val_accuracy: 0.3396\n",
            "Epoch 984/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3605 - accuracy: 0.9082 - val_loss: 6.8908 - val_accuracy: 0.3396\n",
            "Epoch 985/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.3983 - accuracy: 0.8501 - val_loss: 6.8943 - val_accuracy: 0.3208\n",
            "Epoch 986/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.3786 - accuracy: 0.8993 - val_loss: 6.8837 - val_accuracy: 0.3396\n",
            "Epoch 987/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3963 - accuracy: 0.9032 - val_loss: 6.8916 - val_accuracy: 0.3208\n",
            "Epoch 988/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4040 - accuracy: 0.9052 - val_loss: 6.9044 - val_accuracy: 0.3208\n",
            "Epoch 989/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3721 - accuracy: 0.8951 - val_loss: 6.9090 - val_accuracy: 0.3208\n",
            "Epoch 990/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3743 - accuracy: 0.8938 - val_loss: 6.9052 - val_accuracy: 0.3208\n",
            "Epoch 991/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4055 - accuracy: 0.8684 - val_loss: 6.9156 - val_accuracy: 0.3019\n",
            "Epoch 992/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3598 - accuracy: 0.8888 - val_loss: 6.9271 - val_accuracy: 0.3019\n",
            "Epoch 993/2000\n",
            "5/5 [==============================] - 0s 75ms/step - loss: 5.3755 - accuracy: 0.8663 - val_loss: 6.9313 - val_accuracy: 0.3019\n",
            "Epoch 994/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.3769 - accuracy: 0.8787 - val_loss: 6.9328 - val_accuracy: 0.3019\n",
            "Epoch 995/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3550 - accuracy: 0.9160 - val_loss: 6.9170 - val_accuracy: 0.3019\n",
            "Epoch 996/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.3834 - accuracy: 0.9076 - val_loss: 6.9300 - val_accuracy: 0.3019\n",
            "Epoch 997/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3785 - accuracy: 0.8852 - val_loss: 6.9163 - val_accuracy: 0.3019\n",
            "Epoch 998/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3959 - accuracy: 0.8715 - val_loss: 6.9324 - val_accuracy: 0.3019\n",
            "Epoch 999/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3687 - accuracy: 0.8789 - val_loss: 6.9135 - val_accuracy: 0.3019\n",
            "Epoch 1000/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4186 - accuracy: 0.8947 - val_loss: 6.8980 - val_accuracy: 0.3019\n",
            "Epoch 1001/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3629 - accuracy: 0.9044 - val_loss: 6.8837 - val_accuracy: 0.3396\n",
            "Epoch 1002/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3905 - accuracy: 0.8946 - val_loss: 6.8722 - val_accuracy: 0.3396\n",
            "Epoch 1003/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.3941 - accuracy: 0.8671 - val_loss: 6.9076 - val_accuracy: 0.3019\n",
            "Epoch 1004/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.4262 - accuracy: 0.8734 - val_loss: 6.8981 - val_accuracy: 0.3019\n",
            "Epoch 1005/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.3818 - accuracy: 0.8660 - val_loss: 6.9021 - val_accuracy: 0.3019\n",
            "Epoch 1006/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3873 - accuracy: 0.8848 - val_loss: 6.9027 - val_accuracy: 0.3208\n",
            "Epoch 1007/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.4166 - accuracy: 0.8725 - val_loss: 6.9030 - val_accuracy: 0.3208\n",
            "Epoch 1008/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3688 - accuracy: 0.8870 - val_loss: 6.8846 - val_accuracy: 0.3396\n",
            "Epoch 1009/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3816 - accuracy: 0.8715 - val_loss: 6.8673 - val_accuracy: 0.3396\n",
            "Epoch 1010/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3998 - accuracy: 0.8750 - val_loss: 6.8677 - val_accuracy: 0.3396\n",
            "Epoch 1011/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.3614 - accuracy: 0.8797 - val_loss: 6.8705 - val_accuracy: 0.3396\n",
            "Epoch 1012/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3696 - accuracy: 0.8927 - val_loss: 6.8864 - val_accuracy: 0.3396\n",
            "Epoch 1013/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.4109 - accuracy: 0.8688 - val_loss: 6.8919 - val_accuracy: 0.3208\n",
            "Epoch 1014/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3755 - accuracy: 0.8628 - val_loss: 6.8794 - val_accuracy: 0.3396\n",
            "Epoch 1015/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.4210 - accuracy: 0.8786 - val_loss: 6.8803 - val_accuracy: 0.3208\n",
            "Epoch 1016/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3861 - accuracy: 0.8742 - val_loss: 6.8959 - val_accuracy: 0.3208\n",
            "Epoch 1017/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.3342 - accuracy: 0.9089 - val_loss: 6.8900 - val_accuracy: 0.3208\n",
            "Epoch 1018/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3841 - accuracy: 0.8962 - val_loss: 6.8998 - val_accuracy: 0.3208\n",
            "Epoch 1019/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.4099 - accuracy: 0.8919 - val_loss: 6.8916 - val_accuracy: 0.3208\n",
            "Epoch 1020/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3919 - accuracy: 0.8709 - val_loss: 6.9002 - val_accuracy: 0.3208\n",
            "Epoch 1021/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3891 - accuracy: 0.8599 - val_loss: 6.9034 - val_accuracy: 0.3208\n",
            "Epoch 1022/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3702 - accuracy: 0.8851 - val_loss: 6.8875 - val_accuracy: 0.3208\n",
            "Epoch 1023/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3653 - accuracy: 0.9058 - val_loss: 6.8859 - val_accuracy: 0.3396\n",
            "Epoch 1024/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3621 - accuracy: 0.9032 - val_loss: 6.8958 - val_accuracy: 0.3208\n",
            "Epoch 1025/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3807 - accuracy: 0.8884 - val_loss: 6.8786 - val_accuracy: 0.3396\n",
            "Epoch 1026/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3732 - accuracy: 0.8968 - val_loss: 6.8936 - val_accuracy: 0.3208\n",
            "Epoch 1027/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3089 - accuracy: 0.9101 - val_loss: 6.8853 - val_accuracy: 0.3208\n",
            "Epoch 1028/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3606 - accuracy: 0.8810 - val_loss: 6.8948 - val_accuracy: 0.3208\n",
            "Epoch 1029/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3701 - accuracy: 0.8719 - val_loss: 6.8851 - val_accuracy: 0.3208\n",
            "Epoch 1030/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3697 - accuracy: 0.8976 - val_loss: 6.9113 - val_accuracy: 0.3396\n",
            "Epoch 1031/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3486 - accuracy: 0.8766 - val_loss: 6.9264 - val_accuracy: 0.3208\n",
            "Epoch 1032/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3483 - accuracy: 0.9031 - val_loss: 6.9164 - val_accuracy: 0.3208\n",
            "Epoch 1033/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.3820 - accuracy: 0.8728 - val_loss: 6.9113 - val_accuracy: 0.3208\n",
            "Epoch 1034/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3541 - accuracy: 0.8585 - val_loss: 6.8944 - val_accuracy: 0.3396\n",
            "Epoch 1035/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3776 - accuracy: 0.8805 - val_loss: 6.8814 - val_accuracy: 0.3396\n",
            "Epoch 1036/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.3213 - accuracy: 0.9060 - val_loss: 6.8797 - val_accuracy: 0.3396\n",
            "Epoch 1037/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3625 - accuracy: 0.8738 - val_loss: 6.8853 - val_accuracy: 0.3396\n",
            "Epoch 1038/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3350 - accuracy: 0.8962 - val_loss: 6.8793 - val_accuracy: 0.3396\n",
            "Epoch 1039/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3336 - accuracy: 0.9068 - val_loss: 6.8672 - val_accuracy: 0.3396\n",
            "Epoch 1040/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3740 - accuracy: 0.8525 - val_loss: 6.8818 - val_accuracy: 0.3396\n",
            "Epoch 1041/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3182 - accuracy: 0.9059 - val_loss: 6.8873 - val_accuracy: 0.3396\n",
            "Epoch 1042/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3674 - accuracy: 0.8883 - val_loss: 6.8828 - val_accuracy: 0.3396\n",
            "Epoch 1043/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3495 - accuracy: 0.8782 - val_loss: 6.8754 - val_accuracy: 0.3396\n",
            "Epoch 1044/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3546 - accuracy: 0.8882 - val_loss: 6.9015 - val_accuracy: 0.3208\n",
            "Epoch 1045/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3391 - accuracy: 0.8974 - val_loss: 6.8998 - val_accuracy: 0.3208\n",
            "Epoch 1046/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3283 - accuracy: 0.8981 - val_loss: 6.9246 - val_accuracy: 0.3019\n",
            "Epoch 1047/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.3209 - accuracy: 0.8953 - val_loss: 6.9063 - val_accuracy: 0.3208\n",
            "Epoch 1048/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3676 - accuracy: 0.8688 - val_loss: 6.8968 - val_accuracy: 0.3208\n",
            "Epoch 1049/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3371 - accuracy: 0.8915 - val_loss: 6.9050 - val_accuracy: 0.3208\n",
            "Epoch 1050/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3386 - accuracy: 0.8882 - val_loss: 6.9243 - val_accuracy: 0.3019\n",
            "Epoch 1051/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3547 - accuracy: 0.8796 - val_loss: 6.9115 - val_accuracy: 0.3019\n",
            "Epoch 1052/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.3132 - accuracy: 0.8822 - val_loss: 6.8959 - val_accuracy: 0.3019\n",
            "Epoch 1053/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3383 - accuracy: 0.8733 - val_loss: 6.8973 - val_accuracy: 0.3019\n",
            "Epoch 1054/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3643 - accuracy: 0.8696 - val_loss: 6.9068 - val_accuracy: 0.3019\n",
            "Epoch 1055/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3667 - accuracy: 0.8472 - val_loss: 6.8887 - val_accuracy: 0.3208\n",
            "Epoch 1056/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3328 - accuracy: 0.8762 - val_loss: 6.8744 - val_accuracy: 0.3208\n",
            "Epoch 1057/2000\n",
            "5/5 [==============================] - 0s 81ms/step - loss: 5.3081 - accuracy: 0.9101 - val_loss: 6.8738 - val_accuracy: 0.3396\n",
            "Epoch 1058/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.4018 - accuracy: 0.8759 - val_loss: 6.8565 - val_accuracy: 0.3396\n",
            "Epoch 1059/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3218 - accuracy: 0.9001 - val_loss: 6.8718 - val_accuracy: 0.3396\n",
            "Epoch 1060/2000\n",
            "5/5 [==============================] - 0s 47ms/step - loss: 5.2894 - accuracy: 0.8988 - val_loss: 6.8663 - val_accuracy: 0.3396\n",
            "Epoch 1061/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.3320 - accuracy: 0.9028 - val_loss: 6.8669 - val_accuracy: 0.3208\n",
            "Epoch 1062/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3475 - accuracy: 0.8954 - val_loss: 6.8811 - val_accuracy: 0.3208\n",
            "Epoch 1063/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3681 - accuracy: 0.9165 - val_loss: 6.8641 - val_accuracy: 0.3208\n",
            "Epoch 1064/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.3538 - accuracy: 0.9061 - val_loss: 6.8489 - val_accuracy: 0.3396\n",
            "Epoch 1065/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2753 - accuracy: 0.9301 - val_loss: 6.8528 - val_accuracy: 0.3396\n",
            "Epoch 1066/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.3528 - accuracy: 0.8754 - val_loss: 6.8400 - val_accuracy: 0.3585\n",
            "Epoch 1067/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.2857 - accuracy: 0.9055 - val_loss: 6.8235 - val_accuracy: 0.3585\n",
            "Epoch 1068/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3215 - accuracy: 0.9122 - val_loss: 6.8122 - val_accuracy: 0.3585\n",
            "Epoch 1069/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3338 - accuracy: 0.8954 - val_loss: 6.8069 - val_accuracy: 0.3585\n",
            "Epoch 1070/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3420 - accuracy: 0.8656 - val_loss: 6.7963 - val_accuracy: 0.3585\n",
            "Epoch 1071/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3687 - accuracy: 0.8704 - val_loss: 6.7999 - val_accuracy: 0.3585\n",
            "Epoch 1072/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3216 - accuracy: 0.8924 - val_loss: 6.7898 - val_accuracy: 0.3585\n",
            "Epoch 1073/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.3839 - accuracy: 0.8826 - val_loss: 6.7966 - val_accuracy: 0.3585\n",
            "Epoch 1074/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2668 - accuracy: 0.9130 - val_loss: 6.8055 - val_accuracy: 0.3585\n",
            "Epoch 1075/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3363 - accuracy: 0.9035 - val_loss: 6.7927 - val_accuracy: 0.3585\n",
            "Epoch 1076/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3147 - accuracy: 0.8935 - val_loss: 6.7893 - val_accuracy: 0.3585\n",
            "Epoch 1077/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3298 - accuracy: 0.9014 - val_loss: 6.7928 - val_accuracy: 0.3396\n",
            "Epoch 1078/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3400 - accuracy: 0.8898 - val_loss: 6.7902 - val_accuracy: 0.3396\n",
            "Epoch 1079/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.3259 - accuracy: 0.8934 - val_loss: 6.7958 - val_accuracy: 0.3396\n",
            "Epoch 1080/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3018 - accuracy: 0.9077 - val_loss: 6.7944 - val_accuracy: 0.3396\n",
            "Epoch 1081/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3560 - accuracy: 0.8826 - val_loss: 6.7908 - val_accuracy: 0.3396\n",
            "Epoch 1082/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2853 - accuracy: 0.9193 - val_loss: 6.7843 - val_accuracy: 0.3396\n",
            "Epoch 1083/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3507 - accuracy: 0.8898 - val_loss: 6.7754 - val_accuracy: 0.3396\n",
            "Epoch 1084/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3341 - accuracy: 0.8906 - val_loss: 6.7645 - val_accuracy: 0.3208\n",
            "Epoch 1085/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.3148 - accuracy: 0.8948 - val_loss: 6.7586 - val_accuracy: 0.3208\n",
            "Epoch 1086/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2935 - accuracy: 0.9253 - val_loss: 6.7580 - val_accuracy: 0.3396\n",
            "Epoch 1087/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.3100 - accuracy: 0.8902 - val_loss: 6.7645 - val_accuracy: 0.3396\n",
            "Epoch 1088/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.3042 - accuracy: 0.9222 - val_loss: 6.7607 - val_accuracy: 0.3396\n",
            "Epoch 1089/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2387 - accuracy: 0.9069 - val_loss: 6.7566 - val_accuracy: 0.3396\n",
            "Epoch 1090/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.3181 - accuracy: 0.9276 - val_loss: 6.7486 - val_accuracy: 0.3585\n",
            "Epoch 1091/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3014 - accuracy: 0.9070 - val_loss: 6.7422 - val_accuracy: 0.3585\n",
            "Epoch 1092/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3030 - accuracy: 0.9103 - val_loss: 6.7374 - val_accuracy: 0.3585\n",
            "Epoch 1093/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.3639 - accuracy: 0.8730 - val_loss: 6.7502 - val_accuracy: 0.3585\n",
            "Epoch 1094/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3461 - accuracy: 0.9018 - val_loss: 6.7516 - val_accuracy: 0.3585\n",
            "Epoch 1095/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2910 - accuracy: 0.9128 - val_loss: 6.7510 - val_accuracy: 0.3585\n",
            "Epoch 1096/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3292 - accuracy: 0.8834 - val_loss: 6.7431 - val_accuracy: 0.3585\n",
            "Epoch 1097/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2348 - accuracy: 0.9160 - val_loss: 6.7370 - val_accuracy: 0.3585\n",
            "Epoch 1098/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.2912 - accuracy: 0.9027 - val_loss: 6.7461 - val_accuracy: 0.3585\n",
            "Epoch 1099/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3247 - accuracy: 0.9112 - val_loss: 6.7414 - val_accuracy: 0.3585\n",
            "Epoch 1100/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3353 - accuracy: 0.8853 - val_loss: 6.7351 - val_accuracy: 0.3585\n",
            "Epoch 1101/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2987 - accuracy: 0.9058 - val_loss: 6.7330 - val_accuracy: 0.3585\n",
            "Epoch 1102/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.2818 - accuracy: 0.9081 - val_loss: 6.7302 - val_accuracy: 0.3585\n",
            "Epoch 1103/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2982 - accuracy: 0.8808 - val_loss: 6.7327 - val_accuracy: 0.3585\n",
            "Epoch 1104/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3027 - accuracy: 0.8914 - val_loss: 6.7347 - val_accuracy: 0.3585\n",
            "Epoch 1105/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2875 - accuracy: 0.9178 - val_loss: 6.7384 - val_accuracy: 0.3585\n",
            "Epoch 1106/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2876 - accuracy: 0.9105 - val_loss: 6.7439 - val_accuracy: 0.3585\n",
            "Epoch 1107/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2957 - accuracy: 0.9004 - val_loss: 6.7387 - val_accuracy: 0.3585\n",
            "Epoch 1108/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2684 - accuracy: 0.9311 - val_loss: 6.7493 - val_accuracy: 0.3585\n",
            "Epoch 1109/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3627 - accuracy: 0.8733 - val_loss: 6.7407 - val_accuracy: 0.3585\n",
            "Epoch 1110/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3209 - accuracy: 0.8727 - val_loss: 6.7340 - val_accuracy: 0.3585\n",
            "Epoch 1111/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3015 - accuracy: 0.9095 - val_loss: 6.7292 - val_accuracy: 0.3585\n",
            "Epoch 1112/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.3583 - accuracy: 0.8953 - val_loss: 6.7374 - val_accuracy: 0.3585\n",
            "Epoch 1113/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2763 - accuracy: 0.9014 - val_loss: 6.7321 - val_accuracy: 0.3585\n",
            "Epoch 1114/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.3489 - accuracy: 0.8646 - val_loss: 6.7337 - val_accuracy: 0.3585\n",
            "Epoch 1115/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2978 - accuracy: 0.8835 - val_loss: 6.7352 - val_accuracy: 0.3585\n",
            "Epoch 1116/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3335 - accuracy: 0.8661 - val_loss: 6.7329 - val_accuracy: 0.3585\n",
            "Epoch 1117/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2641 - accuracy: 0.9194 - val_loss: 6.7299 - val_accuracy: 0.3585\n",
            "Epoch 1118/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3456 - accuracy: 0.8948 - val_loss: 6.7366 - val_accuracy: 0.3585\n",
            "Epoch 1119/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.2843 - accuracy: 0.8984 - val_loss: 6.7333 - val_accuracy: 0.3585\n",
            "Epoch 1120/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2887 - accuracy: 0.8799 - val_loss: 6.7472 - val_accuracy: 0.3585\n",
            "Epoch 1121/2000\n",
            "5/5 [==============================] - 0s 77ms/step - loss: 5.2870 - accuracy: 0.8751 - val_loss: 6.7439 - val_accuracy: 0.3585\n",
            "Epoch 1122/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2768 - accuracy: 0.8747 - val_loss: 6.7363 - val_accuracy: 0.3585\n",
            "Epoch 1123/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2636 - accuracy: 0.9045 - val_loss: 6.7414 - val_accuracy: 0.3585\n",
            "Epoch 1124/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.3035 - accuracy: 0.9094 - val_loss: 6.7519 - val_accuracy: 0.3585\n",
            "Epoch 1125/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2971 - accuracy: 0.8964 - val_loss: 6.7504 - val_accuracy: 0.3396\n",
            "Epoch 1126/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.3060 - accuracy: 0.8860 - val_loss: 6.7599 - val_accuracy: 0.3396\n",
            "Epoch 1127/2000\n",
            "5/5 [==============================] - 0s 51ms/step - loss: 5.3256 - accuracy: 0.8909 - val_loss: 6.7549 - val_accuracy: 0.3396\n",
            "Epoch 1128/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2970 - accuracy: 0.9078 - val_loss: 6.7468 - val_accuracy: 0.3396\n",
            "Epoch 1129/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2755 - accuracy: 0.8848 - val_loss: 6.7384 - val_accuracy: 0.3585\n",
            "Epoch 1130/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.2819 - accuracy: 0.8825 - val_loss: 6.7322 - val_accuracy: 0.3585\n",
            "Epoch 1131/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2724 - accuracy: 0.8837 - val_loss: 6.7256 - val_accuracy: 0.3585\n",
            "Epoch 1132/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2948 - accuracy: 0.8880 - val_loss: 6.7298 - val_accuracy: 0.3585\n",
            "Epoch 1133/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2846 - accuracy: 0.9163 - val_loss: 6.7397 - val_accuracy: 0.3585\n",
            "Epoch 1134/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2802 - accuracy: 0.8934 - val_loss: 6.7357 - val_accuracy: 0.3585\n",
            "Epoch 1135/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3247 - accuracy: 0.8736 - val_loss: 6.7336 - val_accuracy: 0.3585\n",
            "Epoch 1136/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.2846 - accuracy: 0.8784 - val_loss: 6.7434 - val_accuracy: 0.3585\n",
            "Epoch 1137/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2715 - accuracy: 0.9256 - val_loss: 6.7358 - val_accuracy: 0.3585\n",
            "Epoch 1138/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2742 - accuracy: 0.8956 - val_loss: 6.7565 - val_accuracy: 0.3585\n",
            "Epoch 1139/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.3569 - accuracy: 0.8661 - val_loss: 6.7653 - val_accuracy: 0.3396\n",
            "Epoch 1140/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.3081 - accuracy: 0.8944 - val_loss: 6.7790 - val_accuracy: 0.3396\n",
            "Epoch 1141/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2620 - accuracy: 0.9325 - val_loss: 6.7721 - val_accuracy: 0.3396\n",
            "Epoch 1142/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3098 - accuracy: 0.8855 - val_loss: 6.7790 - val_accuracy: 0.3396\n",
            "Epoch 1143/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.2884 - accuracy: 0.9024 - val_loss: 6.7716 - val_accuracy: 0.3396\n",
            "Epoch 1144/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.3401 - accuracy: 0.8597 - val_loss: 6.7599 - val_accuracy: 0.3396\n",
            "Epoch 1145/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2585 - accuracy: 0.8949 - val_loss: 6.7658 - val_accuracy: 0.3396\n",
            "Epoch 1146/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2309 - accuracy: 0.9229 - val_loss: 6.7583 - val_accuracy: 0.3396\n",
            "Epoch 1147/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2590 - accuracy: 0.9022 - val_loss: 6.7671 - val_accuracy: 0.3396\n",
            "Epoch 1148/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3132 - accuracy: 0.8920 - val_loss: 6.7783 - val_accuracy: 0.3585\n",
            "Epoch 1149/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2453 - accuracy: 0.9140 - val_loss: 6.7733 - val_accuracy: 0.3585\n",
            "Epoch 1150/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3055 - accuracy: 0.8919 - val_loss: 6.7781 - val_accuracy: 0.3585\n",
            "Epoch 1151/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.2578 - accuracy: 0.8913 - val_loss: 6.7811 - val_accuracy: 0.3396\n",
            "Epoch 1152/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2790 - accuracy: 0.8912 - val_loss: 6.7834 - val_accuracy: 0.3396\n",
            "Epoch 1153/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2617 - accuracy: 0.8949 - val_loss: 6.7705 - val_accuracy: 0.3396\n",
            "Epoch 1154/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.3160 - accuracy: 0.8914 - val_loss: 6.7763 - val_accuracy: 0.3396\n",
            "Epoch 1155/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2587 - accuracy: 0.9140 - val_loss: 6.7760 - val_accuracy: 0.3396\n",
            "Epoch 1156/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3026 - accuracy: 0.8849 - val_loss: 6.7830 - val_accuracy: 0.3396\n",
            "Epoch 1157/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.2704 - accuracy: 0.8918 - val_loss: 6.7921 - val_accuracy: 0.3396\n",
            "Epoch 1158/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.2292 - accuracy: 0.9219 - val_loss: 6.7986 - val_accuracy: 0.3396\n",
            "Epoch 1159/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2533 - accuracy: 0.9192 - val_loss: 6.8006 - val_accuracy: 0.3396\n",
            "Epoch 1160/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2541 - accuracy: 0.9139 - val_loss: 6.7938 - val_accuracy: 0.3396\n",
            "Epoch 1161/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.2621 - accuracy: 0.8916 - val_loss: 6.8073 - val_accuracy: 0.3396\n",
            "Epoch 1162/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2602 - accuracy: 0.8768 - val_loss: 6.7936 - val_accuracy: 0.3396\n",
            "Epoch 1163/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2729 - accuracy: 0.8959 - val_loss: 6.7792 - val_accuracy: 0.3585\n",
            "Epoch 1164/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2562 - accuracy: 0.8940 - val_loss: 6.8016 - val_accuracy: 0.3208\n",
            "Epoch 1165/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2670 - accuracy: 0.8994 - val_loss: 6.7958 - val_accuracy: 0.3208\n",
            "Epoch 1166/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2595 - accuracy: 0.8905 - val_loss: 6.7844 - val_accuracy: 0.3396\n",
            "Epoch 1167/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2279 - accuracy: 0.9155 - val_loss: 6.7733 - val_accuracy: 0.3396\n",
            "Epoch 1168/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2965 - accuracy: 0.8700 - val_loss: 6.7634 - val_accuracy: 0.3396\n",
            "Epoch 1169/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2330 - accuracy: 0.9218 - val_loss: 6.7578 - val_accuracy: 0.3396\n",
            "Epoch 1170/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2548 - accuracy: 0.9186 - val_loss: 6.7633 - val_accuracy: 0.3396\n",
            "Epoch 1171/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2422 - accuracy: 0.9144 - val_loss: 6.7563 - val_accuracy: 0.3396\n",
            "Epoch 1172/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2598 - accuracy: 0.9204 - val_loss: 6.7616 - val_accuracy: 0.3585\n",
            "Epoch 1173/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2420 - accuracy: 0.8991 - val_loss: 6.7656 - val_accuracy: 0.3585\n",
            "Epoch 1174/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2589 - accuracy: 0.9151 - val_loss: 6.7954 - val_accuracy: 0.3208\n",
            "Epoch 1175/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.2681 - accuracy: 0.8881 - val_loss: 6.7904 - val_accuracy: 0.3396\n",
            "Epoch 1176/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.3154 - accuracy: 0.9150 - val_loss: 6.7974 - val_accuracy: 0.3208\n",
            "Epoch 1177/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.2900 - accuracy: 0.8914 - val_loss: 6.7949 - val_accuracy: 0.3396\n",
            "Epoch 1178/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2368 - accuracy: 0.9111 - val_loss: 6.7985 - val_accuracy: 0.3396\n",
            "Epoch 1179/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2469 - accuracy: 0.9056 - val_loss: 6.7888 - val_accuracy: 0.3396\n",
            "Epoch 1180/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.2515 - accuracy: 0.8951 - val_loss: 6.7979 - val_accuracy: 0.3208\n",
            "Epoch 1181/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2211 - accuracy: 0.9224 - val_loss: 6.7939 - val_accuracy: 0.3208\n",
            "Epoch 1182/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2608 - accuracy: 0.8950 - val_loss: 6.7938 - val_accuracy: 0.3208\n",
            "Epoch 1183/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2305 - accuracy: 0.8977 - val_loss: 6.7947 - val_accuracy: 0.3208\n",
            "Epoch 1184/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2851 - accuracy: 0.9080 - val_loss: 6.8068 - val_accuracy: 0.3019\n",
            "Epoch 1185/2000\n",
            "5/5 [==============================] - 0s 48ms/step - loss: 5.2416 - accuracy: 0.8890 - val_loss: 6.8120 - val_accuracy: 0.3019\n",
            "Epoch 1186/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2594 - accuracy: 0.9065 - val_loss: 6.8030 - val_accuracy: 0.3019\n",
            "Epoch 1187/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2217 - accuracy: 0.9083 - val_loss: 6.7938 - val_accuracy: 0.3019\n",
            "Epoch 1188/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2767 - accuracy: 0.9058 - val_loss: 6.8055 - val_accuracy: 0.3019\n",
            "Epoch 1189/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2681 - accuracy: 0.9079 - val_loss: 6.7959 - val_accuracy: 0.3019\n",
            "Epoch 1190/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2518 - accuracy: 0.8824 - val_loss: 6.7871 - val_accuracy: 0.3019\n",
            "Epoch 1191/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2948 - accuracy: 0.8887 - val_loss: 6.8066 - val_accuracy: 0.3019\n",
            "Epoch 1192/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.2736 - accuracy: 0.8911 - val_loss: 6.8006 - val_accuracy: 0.3019\n",
            "Epoch 1193/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.2648 - accuracy: 0.9022 - val_loss: 6.8180 - val_accuracy: 0.3019\n",
            "Epoch 1194/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2558 - accuracy: 0.8745 - val_loss: 6.8149 - val_accuracy: 0.3019\n",
            "Epoch 1195/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.2933 - accuracy: 0.8687 - val_loss: 6.8129 - val_accuracy: 0.3019\n",
            "Epoch 1196/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2502 - accuracy: 0.9063 - val_loss: 6.7998 - val_accuracy: 0.3019\n",
            "Epoch 1197/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2437 - accuracy: 0.9196 - val_loss: 6.7884 - val_accuracy: 0.3019\n",
            "Epoch 1198/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2634 - accuracy: 0.8852 - val_loss: 6.7767 - val_accuracy: 0.3019\n",
            "Epoch 1199/2000\n",
            "5/5 [==============================] - 0s 79ms/step - loss: 5.2541 - accuracy: 0.9258 - val_loss: 6.7634 - val_accuracy: 0.3208\n",
            "Epoch 1200/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2490 - accuracy: 0.8894 - val_loss: 6.7791 - val_accuracy: 0.3208\n",
            "Epoch 1201/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.3041 - accuracy: 0.8853 - val_loss: 6.7862 - val_accuracy: 0.3208\n",
            "Epoch 1202/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.2907 - accuracy: 0.8780 - val_loss: 6.7910 - val_accuracy: 0.3208\n",
            "Epoch 1203/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2534 - accuracy: 0.8955 - val_loss: 6.7992 - val_accuracy: 0.3208\n",
            "Epoch 1204/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2688 - accuracy: 0.8841 - val_loss: 6.7979 - val_accuracy: 0.3208\n",
            "Epoch 1205/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2908 - accuracy: 0.8814 - val_loss: 6.7897 - val_accuracy: 0.3208\n",
            "Epoch 1206/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2367 - accuracy: 0.9013 - val_loss: 6.8053 - val_accuracy: 0.3208\n",
            "Epoch 1207/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2314 - accuracy: 0.9063 - val_loss: 6.8125 - val_accuracy: 0.3208\n",
            "Epoch 1208/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2723 - accuracy: 0.8739 - val_loss: 6.7979 - val_accuracy: 0.3208\n",
            "Epoch 1209/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2452 - accuracy: 0.9149 - val_loss: 6.7913 - val_accuracy: 0.3208\n",
            "Epoch 1210/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2324 - accuracy: 0.8994 - val_loss: 6.8012 - val_accuracy: 0.3208\n",
            "Epoch 1211/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2374 - accuracy: 0.8848 - val_loss: 6.8039 - val_accuracy: 0.3208\n",
            "Epoch 1212/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.2474 - accuracy: 0.8803 - val_loss: 6.8021 - val_accuracy: 0.3396\n",
            "Epoch 1213/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2260 - accuracy: 0.9179 - val_loss: 6.7888 - val_accuracy: 0.3396\n",
            "Epoch 1214/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2391 - accuracy: 0.8853 - val_loss: 6.7916 - val_accuracy: 0.3208\n",
            "Epoch 1215/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2029 - accuracy: 0.8930 - val_loss: 6.8005 - val_accuracy: 0.3208\n",
            "Epoch 1216/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2316 - accuracy: 0.9097 - val_loss: 6.7841 - val_accuracy: 0.3396\n",
            "Epoch 1217/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.2176 - accuracy: 0.9164 - val_loss: 6.7779 - val_accuracy: 0.3208\n",
            "Epoch 1218/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1755 - accuracy: 0.9061 - val_loss: 6.7603 - val_accuracy: 0.3396\n",
            "Epoch 1219/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2368 - accuracy: 0.8685 - val_loss: 6.7461 - val_accuracy: 0.3396\n",
            "Epoch 1220/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2192 - accuracy: 0.8910 - val_loss: 6.7575 - val_accuracy: 0.3396\n",
            "Epoch 1221/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.1959 - accuracy: 0.9192 - val_loss: 6.7464 - val_accuracy: 0.3396\n",
            "Epoch 1222/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2198 - accuracy: 0.8940 - val_loss: 6.7446 - val_accuracy: 0.3396\n",
            "Epoch 1223/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2269 - accuracy: 0.9394 - val_loss: 6.7587 - val_accuracy: 0.3396\n",
            "Epoch 1224/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1918 - accuracy: 0.9384 - val_loss: 6.7465 - val_accuracy: 0.3396\n",
            "Epoch 1225/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2202 - accuracy: 0.9308 - val_loss: 6.7407 - val_accuracy: 0.3396\n",
            "Epoch 1226/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2016 - accuracy: 0.9064 - val_loss: 6.7494 - val_accuracy: 0.3396\n",
            "Epoch 1227/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.2499 - accuracy: 0.8877 - val_loss: 6.7594 - val_accuracy: 0.3396\n",
            "Epoch 1228/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2026 - accuracy: 0.9254 - val_loss: 6.7619 - val_accuracy: 0.3396\n",
            "Epoch 1229/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2175 - accuracy: 0.9126 - val_loss: 6.7601 - val_accuracy: 0.3396\n",
            "Epoch 1230/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1897 - accuracy: 0.9414 - val_loss: 6.7518 - val_accuracy: 0.3396\n",
            "Epoch 1231/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2496 - accuracy: 0.8854 - val_loss: 6.7559 - val_accuracy: 0.3585\n",
            "Epoch 1232/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2269 - accuracy: 0.9071 - val_loss: 6.7399 - val_accuracy: 0.3774\n",
            "Epoch 1233/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2190 - accuracy: 0.9297 - val_loss: 6.7261 - val_accuracy: 0.3774\n",
            "Epoch 1234/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.2108 - accuracy: 0.8954 - val_loss: 6.7165 - val_accuracy: 0.3774\n",
            "Epoch 1235/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2337 - accuracy: 0.8936 - val_loss: 6.7288 - val_accuracy: 0.3774\n",
            "Epoch 1236/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2891 - accuracy: 0.8747 - val_loss: 6.7397 - val_accuracy: 0.3585\n",
            "Epoch 1237/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.2070 - accuracy: 0.8863 - val_loss: 6.7263 - val_accuracy: 0.3585\n",
            "Epoch 1238/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2354 - accuracy: 0.8793 - val_loss: 6.7318 - val_accuracy: 0.3585\n",
            "Epoch 1239/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1994 - accuracy: 0.9057 - val_loss: 6.7215 - val_accuracy: 0.3585\n",
            "Epoch 1240/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2270 - accuracy: 0.9091 - val_loss: 6.7208 - val_accuracy: 0.3585\n",
            "Epoch 1241/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2372 - accuracy: 0.8983 - val_loss: 6.7335 - val_accuracy: 0.3585\n",
            "Epoch 1242/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2155 - accuracy: 0.8954 - val_loss: 6.7355 - val_accuracy: 0.3774\n",
            "Epoch 1243/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1858 - accuracy: 0.9004 - val_loss: 6.7351 - val_accuracy: 0.3774\n",
            "Epoch 1244/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1958 - accuracy: 0.8706 - val_loss: 6.7349 - val_accuracy: 0.3774\n",
            "Epoch 1245/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1811 - accuracy: 0.9378 - val_loss: 6.7545 - val_accuracy: 0.3585\n",
            "Epoch 1246/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2851 - accuracy: 0.8952 - val_loss: 6.7530 - val_accuracy: 0.3585\n",
            "Epoch 1247/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2479 - accuracy: 0.8991 - val_loss: 6.7559 - val_accuracy: 0.3585\n",
            "Epoch 1248/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1933 - accuracy: 0.9043 - val_loss: 6.7456 - val_accuracy: 0.3585\n",
            "Epoch 1249/2000\n",
            "5/5 [==============================] - 0s 54ms/step - loss: 5.2519 - accuracy: 0.9093 - val_loss: 6.7604 - val_accuracy: 0.3585\n",
            "Epoch 1250/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1727 - accuracy: 0.9166 - val_loss: 6.7448 - val_accuracy: 0.3585\n",
            "Epoch 1251/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1895 - accuracy: 0.8874 - val_loss: 6.7314 - val_accuracy: 0.3774\n",
            "Epoch 1252/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.2251 - accuracy: 0.9055 - val_loss: 6.7322 - val_accuracy: 0.3585\n",
            "Epoch 1253/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2173 - accuracy: 0.9087 - val_loss: 6.7330 - val_accuracy: 0.3585\n",
            "Epoch 1254/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1955 - accuracy: 0.9044 - val_loss: 6.7276 - val_accuracy: 0.3585\n",
            "Epoch 1255/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1854 - accuracy: 0.9113 - val_loss: 6.7476 - val_accuracy: 0.3585\n",
            "Epoch 1256/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2736 - accuracy: 0.8780 - val_loss: 6.7641 - val_accuracy: 0.3585\n",
            "Epoch 1257/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1926 - accuracy: 0.9171 - val_loss: 6.7535 - val_accuracy: 0.3396\n",
            "Epoch 1258/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.2265 - accuracy: 0.9273 - val_loss: 6.7586 - val_accuracy: 0.3396\n",
            "Epoch 1259/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.1705 - accuracy: 0.9133 - val_loss: 6.7461 - val_accuracy: 0.3396\n",
            "Epoch 1260/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.1519 - accuracy: 0.9390 - val_loss: 6.7337 - val_accuracy: 0.3396\n",
            "Epoch 1261/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2245 - accuracy: 0.9161 - val_loss: 6.7342 - val_accuracy: 0.3396\n",
            "Epoch 1262/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1979 - accuracy: 0.9279 - val_loss: 6.7363 - val_accuracy: 0.3396\n",
            "Epoch 1263/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1922 - accuracy: 0.9138 - val_loss: 6.7240 - val_accuracy: 0.3396\n",
            "Epoch 1264/2000\n",
            "5/5 [==============================] - 0s 47ms/step - loss: 5.2230 - accuracy: 0.8959 - val_loss: 6.7104 - val_accuracy: 0.3396\n",
            "Epoch 1265/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2088 - accuracy: 0.8842 - val_loss: 6.6944 - val_accuracy: 0.3774\n",
            "Epoch 1266/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2090 - accuracy: 0.9225 - val_loss: 6.7026 - val_accuracy: 0.3396\n",
            "Epoch 1267/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2050 - accuracy: 0.9361 - val_loss: 6.7050 - val_accuracy: 0.3396\n",
            "Epoch 1268/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2121 - accuracy: 0.9034 - val_loss: 6.6929 - val_accuracy: 0.3396\n",
            "Epoch 1269/2000\n",
            "5/5 [==============================] - 0s 49ms/step - loss: 5.2325 - accuracy: 0.9125 - val_loss: 6.6861 - val_accuracy: 0.3585\n",
            "Epoch 1270/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.1534 - accuracy: 0.9367 - val_loss: 6.6915 - val_accuracy: 0.3396\n",
            "Epoch 1271/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1780 - accuracy: 0.8983 - val_loss: 6.6789 - val_accuracy: 0.3585\n",
            "Epoch 1272/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.2216 - accuracy: 0.8942 - val_loss: 6.6914 - val_accuracy: 0.3396\n",
            "Epoch 1273/2000\n",
            "5/5 [==============================] - 0s 74ms/step - loss: 5.1939 - accuracy: 0.9173 - val_loss: 6.7081 - val_accuracy: 0.3396\n",
            "Epoch 1274/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2273 - accuracy: 0.9168 - val_loss: 6.7094 - val_accuracy: 0.3396\n",
            "Epoch 1275/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1882 - accuracy: 0.9129 - val_loss: 6.7144 - val_accuracy: 0.3396\n",
            "Epoch 1276/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.2159 - accuracy: 0.8795 - val_loss: 6.7115 - val_accuracy: 0.3396\n",
            "Epoch 1277/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.2045 - accuracy: 0.9101 - val_loss: 6.7035 - val_accuracy: 0.3396\n",
            "Epoch 1278/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1918 - accuracy: 0.9033 - val_loss: 6.6947 - val_accuracy: 0.3396\n",
            "Epoch 1279/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.2222 - accuracy: 0.9095 - val_loss: 6.6849 - val_accuracy: 0.3396\n",
            "Epoch 1280/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.1887 - accuracy: 0.9205 - val_loss: 6.6796 - val_accuracy: 0.3396\n",
            "Epoch 1281/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1615 - accuracy: 0.9195 - val_loss: 6.6684 - val_accuracy: 0.3585\n",
            "Epoch 1282/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1707 - accuracy: 0.8800 - val_loss: 6.6672 - val_accuracy: 0.3585\n",
            "Epoch 1283/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1867 - accuracy: 0.8941 - val_loss: 6.6577 - val_accuracy: 0.3585\n",
            "Epoch 1284/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1712 - accuracy: 0.8925 - val_loss: 6.6677 - val_accuracy: 0.3396\n",
            "Epoch 1285/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1404 - accuracy: 0.9310 - val_loss: 6.6662 - val_accuracy: 0.3208\n",
            "Epoch 1286/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1792 - accuracy: 0.9056 - val_loss: 6.6589 - val_accuracy: 0.3396\n",
            "Epoch 1287/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1875 - accuracy: 0.9088 - val_loss: 6.6684 - val_accuracy: 0.3396\n",
            "Epoch 1288/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1793 - accuracy: 0.9073 - val_loss: 6.6600 - val_accuracy: 0.3585\n",
            "Epoch 1289/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1661 - accuracy: 0.9163 - val_loss: 6.6490 - val_accuracy: 0.3585\n",
            "Epoch 1290/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2120 - accuracy: 0.9063 - val_loss: 6.6573 - val_accuracy: 0.3585\n",
            "Epoch 1291/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.1965 - accuracy: 0.9095 - val_loss: 6.6629 - val_accuracy: 0.3585\n",
            "Epoch 1292/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1900 - accuracy: 0.8999 - val_loss: 6.6692 - val_accuracy: 0.3585\n",
            "Epoch 1293/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1552 - accuracy: 0.9124 - val_loss: 6.6959 - val_accuracy: 0.3208\n",
            "Epoch 1294/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.2059 - accuracy: 0.8938 - val_loss: 6.6930 - val_accuracy: 0.3208\n",
            "Epoch 1295/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1706 - accuracy: 0.9175 - val_loss: 6.6844 - val_accuracy: 0.3396\n",
            "Epoch 1296/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.2353 - accuracy: 0.9009 - val_loss: 6.6918 - val_accuracy: 0.3208\n",
            "Epoch 1297/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1381 - accuracy: 0.9318 - val_loss: 6.6956 - val_accuracy: 0.3208\n",
            "Epoch 1298/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2323 - accuracy: 0.8895 - val_loss: 6.6846 - val_accuracy: 0.3585\n",
            "Epoch 1299/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1388 - accuracy: 0.9092 - val_loss: 6.6913 - val_accuracy: 0.3585\n",
            "Epoch 1300/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2524 - accuracy: 0.8664 - val_loss: 6.7027 - val_accuracy: 0.3396\n",
            "Epoch 1301/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.2035 - accuracy: 0.8904 - val_loss: 6.6889 - val_accuracy: 0.3585\n",
            "Epoch 1302/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1433 - accuracy: 0.9102 - val_loss: 6.6920 - val_accuracy: 0.3585\n",
            "Epoch 1303/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.2381 - accuracy: 0.9155 - val_loss: 6.7065 - val_accuracy: 0.3396\n",
            "Epoch 1304/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1793 - accuracy: 0.9097 - val_loss: 6.6925 - val_accuracy: 0.3774\n",
            "Epoch 1305/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1556 - accuracy: 0.8845 - val_loss: 6.7006 - val_accuracy: 0.3585\n",
            "Epoch 1306/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.1594 - accuracy: 0.9117 - val_loss: 6.6924 - val_accuracy: 0.3774\n",
            "Epoch 1307/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1561 - accuracy: 0.9263 - val_loss: 6.6874 - val_accuracy: 0.3774\n",
            "Epoch 1308/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.2227 - accuracy: 0.8671 - val_loss: 6.6809 - val_accuracy: 0.3774\n",
            "Epoch 1309/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1888 - accuracy: 0.8903 - val_loss: 6.6736 - val_accuracy: 0.3774\n",
            "Epoch 1310/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1962 - accuracy: 0.9030 - val_loss: 6.6619 - val_accuracy: 0.3774\n",
            "Epoch 1311/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.1839 - accuracy: 0.8711 - val_loss: 6.6646 - val_accuracy: 0.3774\n",
            "Epoch 1312/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1554 - accuracy: 0.9257 - val_loss: 6.6539 - val_accuracy: 0.3585\n",
            "Epoch 1313/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1997 - accuracy: 0.8945 - val_loss: 6.6535 - val_accuracy: 0.3585\n",
            "Epoch 1314/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.1737 - accuracy: 0.9084 - val_loss: 6.6420 - val_accuracy: 0.3585\n",
            "Epoch 1315/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.1735 - accuracy: 0.9265 - val_loss: 6.6475 - val_accuracy: 0.3585\n",
            "Epoch 1316/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1372 - accuracy: 0.9328 - val_loss: 6.6383 - val_accuracy: 0.3774\n",
            "Epoch 1317/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1269 - accuracy: 0.8920 - val_loss: 6.6454 - val_accuracy: 0.3774\n",
            "Epoch 1318/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1737 - accuracy: 0.9004 - val_loss: 6.6509 - val_accuracy: 0.3585\n",
            "Epoch 1319/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2183 - accuracy: 0.8691 - val_loss: 6.6598 - val_accuracy: 0.3585\n",
            "Epoch 1320/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1316 - accuracy: 0.9167 - val_loss: 6.6917 - val_accuracy: 0.3396\n",
            "Epoch 1321/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1810 - accuracy: 0.8793 - val_loss: 6.6807 - val_accuracy: 0.3396\n",
            "Epoch 1322/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1730 - accuracy: 0.9310 - val_loss: 6.6834 - val_accuracy: 0.3396\n",
            "Epoch 1323/2000\n",
            "5/5 [==============================] - 0s 47ms/step - loss: 5.1606 - accuracy: 0.9204 - val_loss: 6.7060 - val_accuracy: 0.3396\n",
            "Epoch 1324/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1718 - accuracy: 0.9108 - val_loss: 6.7117 - val_accuracy: 0.3396\n",
            "Epoch 1325/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.1888 - accuracy: 0.9130 - val_loss: 6.7064 - val_accuracy: 0.3396\n",
            "Epoch 1326/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1613 - accuracy: 0.8967 - val_loss: 6.7244 - val_accuracy: 0.3396\n",
            "Epoch 1327/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.1199 - accuracy: 0.9357 - val_loss: 6.7168 - val_accuracy: 0.3396\n",
            "Epoch 1328/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1503 - accuracy: 0.9079 - val_loss: 6.7045 - val_accuracy: 0.3396\n",
            "Epoch 1329/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.1655 - accuracy: 0.9205 - val_loss: 6.7087 - val_accuracy: 0.3396\n",
            "Epoch 1330/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2010 - accuracy: 0.8786 - val_loss: 6.6965 - val_accuracy: 0.3396\n",
            "Epoch 1331/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1996 - accuracy: 0.8894 - val_loss: 6.6920 - val_accuracy: 0.3396\n",
            "Epoch 1332/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1113 - accuracy: 0.9138 - val_loss: 6.6833 - val_accuracy: 0.3396\n",
            "Epoch 1333/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1634 - accuracy: 0.8961 - val_loss: 6.6679 - val_accuracy: 0.3585\n",
            "Epoch 1334/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1444 - accuracy: 0.9279 - val_loss: 6.6562 - val_accuracy: 0.3585\n",
            "Epoch 1335/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.2030 - accuracy: 0.9106 - val_loss: 6.6760 - val_accuracy: 0.3396\n",
            "Epoch 1336/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1564 - accuracy: 0.9075 - val_loss: 6.6960 - val_accuracy: 0.3396\n",
            "Epoch 1337/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.2200 - accuracy: 0.8858 - val_loss: 6.6875 - val_accuracy: 0.3396\n",
            "Epoch 1338/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1500 - accuracy: 0.8966 - val_loss: 6.6764 - val_accuracy: 0.3396\n",
            "Epoch 1339/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1849 - accuracy: 0.9240 - val_loss: 6.6710 - val_accuracy: 0.3396\n",
            "Epoch 1340/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.2070 - accuracy: 0.9011 - val_loss: 6.6800 - val_accuracy: 0.3396\n",
            "Epoch 1341/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1404 - accuracy: 0.9169 - val_loss: 6.6732 - val_accuracy: 0.3585\n",
            "Epoch 1342/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1699 - accuracy: 0.9043 - val_loss: 6.6719 - val_accuracy: 0.3585\n",
            "Epoch 1343/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1807 - accuracy: 0.9052 - val_loss: 6.6716 - val_accuracy: 0.3585\n",
            "Epoch 1344/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.1312 - accuracy: 0.9149 - val_loss: 6.6688 - val_accuracy: 0.3585\n",
            "Epoch 1345/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1570 - accuracy: 0.9255 - val_loss: 6.6749 - val_accuracy: 0.3396\n",
            "Epoch 1346/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1853 - accuracy: 0.8754 - val_loss: 6.6660 - val_accuracy: 0.3396\n",
            "Epoch 1347/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1185 - accuracy: 0.9365 - val_loss: 6.6771 - val_accuracy: 0.3208\n",
            "Epoch 1348/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1630 - accuracy: 0.9239 - val_loss: 6.6694 - val_accuracy: 0.3208\n",
            "Epoch 1349/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.1779 - accuracy: 0.8944 - val_loss: 6.6835 - val_accuracy: 0.3208\n",
            "Epoch 1350/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1595 - accuracy: 0.8718 - val_loss: 6.6994 - val_accuracy: 0.3208\n",
            "Epoch 1351/2000\n",
            "5/5 [==============================] - 0s 47ms/step - loss: 5.1835 - accuracy: 0.9139 - val_loss: 6.7325 - val_accuracy: 0.3208\n",
            "Epoch 1352/2000\n",
            "5/5 [==============================] - 0s 80ms/step - loss: 5.1744 - accuracy: 0.8883 - val_loss: 6.7302 - val_accuracy: 0.3208\n",
            "Epoch 1353/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.1226 - accuracy: 0.9337 - val_loss: 6.7131 - val_accuracy: 0.3208\n",
            "Epoch 1354/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1517 - accuracy: 0.9206 - val_loss: 6.7167 - val_accuracy: 0.3208\n",
            "Epoch 1355/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1564 - accuracy: 0.9228 - val_loss: 6.7099 - val_accuracy: 0.3208\n",
            "Epoch 1356/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1401 - accuracy: 0.9304 - val_loss: 6.7009 - val_accuracy: 0.3208\n",
            "Epoch 1357/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1355 - accuracy: 0.9122 - val_loss: 6.6939 - val_accuracy: 0.3208\n",
            "Epoch 1358/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1558 - accuracy: 0.9129 - val_loss: 6.7001 - val_accuracy: 0.3208\n",
            "Epoch 1359/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1269 - accuracy: 0.9144 - val_loss: 6.6815 - val_accuracy: 0.3585\n",
            "Epoch 1360/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1905 - accuracy: 0.8952 - val_loss: 6.6886 - val_accuracy: 0.3396\n",
            "Epoch 1361/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1551 - accuracy: 0.9226 - val_loss: 6.6808 - val_accuracy: 0.3396\n",
            "Epoch 1362/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.1840 - accuracy: 0.9102 - val_loss: 6.6852 - val_accuracy: 0.3396\n",
            "Epoch 1363/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1367 - accuracy: 0.9064 - val_loss: 6.6892 - val_accuracy: 0.3208\n",
            "Epoch 1364/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1643 - accuracy: 0.8924 - val_loss: 6.6869 - val_accuracy: 0.3208\n",
            "Epoch 1365/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1577 - accuracy: 0.8951 - val_loss: 6.6919 - val_accuracy: 0.3208\n",
            "Epoch 1366/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1677 - accuracy: 0.8808 - val_loss: 6.7150 - val_accuracy: 0.3208\n",
            "Epoch 1367/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1114 - accuracy: 0.9293 - val_loss: 6.6990 - val_accuracy: 0.3208\n",
            "Epoch 1368/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1431 - accuracy: 0.9042 - val_loss: 6.7075 - val_accuracy: 0.3208\n",
            "Epoch 1369/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1082 - accuracy: 0.9168 - val_loss: 6.6934 - val_accuracy: 0.3396\n",
            "Epoch 1370/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1402 - accuracy: 0.9132 - val_loss: 6.6793 - val_accuracy: 0.3585\n",
            "Epoch 1371/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1581 - accuracy: 0.9153 - val_loss: 6.6734 - val_accuracy: 0.3585\n",
            "Epoch 1372/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1468 - accuracy: 0.8953 - val_loss: 6.6676 - val_accuracy: 0.3585\n",
            "Epoch 1373/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1285 - accuracy: 0.9243 - val_loss: 6.6882 - val_accuracy: 0.3396\n",
            "Epoch 1374/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1002 - accuracy: 0.9326 - val_loss: 6.7071 - val_accuracy: 0.3585\n",
            "Epoch 1375/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0969 - accuracy: 0.9434 - val_loss: 6.7085 - val_accuracy: 0.3396\n",
            "Epoch 1376/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1616 - accuracy: 0.9028 - val_loss: 6.7206 - val_accuracy: 0.3208\n",
            "Epoch 1377/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1731 - accuracy: 0.9040 - val_loss: 6.7224 - val_accuracy: 0.3396\n",
            "Epoch 1378/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1248 - accuracy: 0.9143 - val_loss: 6.7360 - val_accuracy: 0.3396\n",
            "Epoch 1379/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.1141 - accuracy: 0.9288 - val_loss: 6.7269 - val_accuracy: 0.3396\n",
            "Epoch 1380/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1325 - accuracy: 0.9133 - val_loss: 6.7469 - val_accuracy: 0.3396\n",
            "Epoch 1381/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1689 - accuracy: 0.9114 - val_loss: 6.7225 - val_accuracy: 0.3396\n",
            "Epoch 1382/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.1699 - accuracy: 0.9053 - val_loss: 6.7077 - val_accuracy: 0.3396\n",
            "Epoch 1383/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1343 - accuracy: 0.8944 - val_loss: 6.6911 - val_accuracy: 0.3396\n",
            "Epoch 1384/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1574 - accuracy: 0.8842 - val_loss: 6.6716 - val_accuracy: 0.3396\n",
            "Epoch 1385/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1992 - accuracy: 0.8759 - val_loss: 6.6914 - val_accuracy: 0.3396\n",
            "Epoch 1386/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1335 - accuracy: 0.9143 - val_loss: 6.6845 - val_accuracy: 0.3396\n",
            "Epoch 1387/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1114 - accuracy: 0.8802 - val_loss: 6.7102 - val_accuracy: 0.3396\n",
            "Epoch 1388/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1138 - accuracy: 0.9248 - val_loss: 6.7419 - val_accuracy: 0.3396\n",
            "Epoch 1389/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1245 - accuracy: 0.9411 - val_loss: 6.7230 - val_accuracy: 0.3396\n",
            "Epoch 1390/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1548 - accuracy: 0.9106 - val_loss: 6.7421 - val_accuracy: 0.3396\n",
            "Epoch 1391/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1518 - accuracy: 0.8777 - val_loss: 6.7206 - val_accuracy: 0.3396\n",
            "Epoch 1392/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1228 - accuracy: 0.8988 - val_loss: 6.7015 - val_accuracy: 0.3396\n",
            "Epoch 1393/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1216 - accuracy: 0.8977 - val_loss: 6.7032 - val_accuracy: 0.3396\n",
            "Epoch 1394/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1308 - accuracy: 0.9060 - val_loss: 6.6888 - val_accuracy: 0.3396\n",
            "Epoch 1395/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1267 - accuracy: 0.9114 - val_loss: 6.6739 - val_accuracy: 0.3396\n",
            "Epoch 1396/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1562 - accuracy: 0.9034 - val_loss: 6.6992 - val_accuracy: 0.3396\n",
            "Epoch 1397/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1547 - accuracy: 0.8940 - val_loss: 6.6988 - val_accuracy: 0.3208\n",
            "Epoch 1398/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1672 - accuracy: 0.9108 - val_loss: 6.6808 - val_accuracy: 0.3396\n",
            "Epoch 1399/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1366 - accuracy: 0.8992 - val_loss: 6.6810 - val_accuracy: 0.3585\n",
            "Epoch 1400/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1056 - accuracy: 0.9115 - val_loss: 6.6950 - val_accuracy: 0.3396\n",
            "Epoch 1401/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1303 - accuracy: 0.9101 - val_loss: 6.6896 - val_accuracy: 0.3396\n",
            "Epoch 1402/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1263 - accuracy: 0.9076 - val_loss: 6.6723 - val_accuracy: 0.3585\n",
            "Epoch 1403/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1379 - accuracy: 0.9231 - val_loss: 6.6746 - val_accuracy: 0.3585\n",
            "Epoch 1404/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1448 - accuracy: 0.8931 - val_loss: 6.6747 - val_accuracy: 0.3585\n",
            "Epoch 1405/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1440 - accuracy: 0.9105 - val_loss: 6.6793 - val_accuracy: 0.3585\n",
            "Epoch 1406/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1586 - accuracy: 0.8741 - val_loss: 6.6775 - val_accuracy: 0.3585\n",
            "Epoch 1407/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1128 - accuracy: 0.9221 - val_loss: 6.6701 - val_accuracy: 0.3585\n",
            "Epoch 1408/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1283 - accuracy: 0.8908 - val_loss: 6.6513 - val_accuracy: 0.3396\n",
            "Epoch 1409/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1204 - accuracy: 0.8926 - val_loss: 6.6521 - val_accuracy: 0.3396\n",
            "Epoch 1410/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.1432 - accuracy: 0.8836 - val_loss: 6.6368 - val_accuracy: 0.3396\n",
            "Epoch 1411/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0954 - accuracy: 0.9486 - val_loss: 6.6480 - val_accuracy: 0.3396\n",
            "Epoch 1412/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1058 - accuracy: 0.9182 - val_loss: 6.6512 - val_accuracy: 0.3396\n",
            "Epoch 1413/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0764 - accuracy: 0.9117 - val_loss: 6.6611 - val_accuracy: 0.3396\n",
            "Epoch 1414/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1028 - accuracy: 0.9323 - val_loss: 6.6682 - val_accuracy: 0.3396\n",
            "Epoch 1415/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1111 - accuracy: 0.9018 - val_loss: 6.6594 - val_accuracy: 0.3396\n",
            "Epoch 1416/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0883 - accuracy: 0.8878 - val_loss: 6.6663 - val_accuracy: 0.3585\n",
            "Epoch 1417/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.1601 - accuracy: 0.8839 - val_loss: 6.6571 - val_accuracy: 0.3585\n",
            "Epoch 1418/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1408 - accuracy: 0.9108 - val_loss: 6.6707 - val_accuracy: 0.3585\n",
            "Epoch 1419/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0599 - accuracy: 0.9516 - val_loss: 6.6676 - val_accuracy: 0.3585\n",
            "Epoch 1420/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1679 - accuracy: 0.8999 - val_loss: 6.6624 - val_accuracy: 0.3396\n",
            "Epoch 1421/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1081 - accuracy: 0.9092 - val_loss: 6.6641 - val_accuracy: 0.3396\n",
            "Epoch 1422/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1204 - accuracy: 0.9233 - val_loss: 6.6585 - val_accuracy: 0.3585\n",
            "Epoch 1423/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1305 - accuracy: 0.8995 - val_loss: 6.6543 - val_accuracy: 0.3585\n",
            "Epoch 1424/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0682 - accuracy: 0.9258 - val_loss: 6.6380 - val_accuracy: 0.3585\n",
            "Epoch 1425/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0749 - accuracy: 0.9115 - val_loss: 6.6378 - val_accuracy: 0.3585\n",
            "Epoch 1426/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.1394 - accuracy: 0.8794 - val_loss: 6.6462 - val_accuracy: 0.3396\n",
            "Epoch 1427/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.1096 - accuracy: 0.9148 - val_loss: 6.6475 - val_accuracy: 0.3396\n",
            "Epoch 1428/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1298 - accuracy: 0.9115 - val_loss: 6.6355 - val_accuracy: 0.3396\n",
            "Epoch 1429/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1194 - accuracy: 0.9307 - val_loss: 6.6293 - val_accuracy: 0.3396\n",
            "Epoch 1430/2000\n",
            "5/5 [==============================] - 0s 76ms/step - loss: 5.1073 - accuracy: 0.9057 - val_loss: 6.6237 - val_accuracy: 0.3585\n",
            "Epoch 1431/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0915 - accuracy: 0.9228 - val_loss: 6.6359 - val_accuracy: 0.3396\n",
            "Epoch 1432/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0760 - accuracy: 0.8774 - val_loss: 6.6231 - val_accuracy: 0.3396\n",
            "Epoch 1433/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0971 - accuracy: 0.8972 - val_loss: 6.6143 - val_accuracy: 0.3585\n",
            "Epoch 1434/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.0677 - accuracy: 0.9133 - val_loss: 6.6019 - val_accuracy: 0.3585\n",
            "Epoch 1435/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0948 - accuracy: 0.8984 - val_loss: 6.5942 - val_accuracy: 0.3585\n",
            "Epoch 1436/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1089 - accuracy: 0.8997 - val_loss: 6.5923 - val_accuracy: 0.3585\n",
            "Epoch 1437/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1280 - accuracy: 0.9148 - val_loss: 6.5867 - val_accuracy: 0.3585\n",
            "Epoch 1438/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0881 - accuracy: 0.9307 - val_loss: 6.5841 - val_accuracy: 0.3585\n",
            "Epoch 1439/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0911 - accuracy: 0.9114 - val_loss: 6.5713 - val_accuracy: 0.3585\n",
            "Epoch 1440/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.0669 - accuracy: 0.9257 - val_loss: 6.5603 - val_accuracy: 0.3585\n",
            "Epoch 1441/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1557 - accuracy: 0.9113 - val_loss: 6.5897 - val_accuracy: 0.3774\n",
            "Epoch 1442/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0933 - accuracy: 0.8992 - val_loss: 6.5988 - val_accuracy: 0.3774\n",
            "Epoch 1443/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0622 - accuracy: 0.9336 - val_loss: 6.6035 - val_accuracy: 0.3774\n",
            "Epoch 1444/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1512 - accuracy: 0.8989 - val_loss: 6.5942 - val_accuracy: 0.3774\n",
            "Epoch 1445/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0864 - accuracy: 0.8949 - val_loss: 6.5913 - val_accuracy: 0.3585\n",
            "Epoch 1446/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1277 - accuracy: 0.9065 - val_loss: 6.6059 - val_accuracy: 0.3585\n",
            "Epoch 1447/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1079 - accuracy: 0.9128 - val_loss: 6.5941 - val_accuracy: 0.3585\n",
            "Epoch 1448/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1932 - accuracy: 0.8535 - val_loss: 6.5945 - val_accuracy: 0.3585\n",
            "Epoch 1449/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1292 - accuracy: 0.9014 - val_loss: 6.6064 - val_accuracy: 0.3585\n",
            "Epoch 1450/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0993 - accuracy: 0.9175 - val_loss: 6.6203 - val_accuracy: 0.3585\n",
            "Epoch 1451/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.0904 - accuracy: 0.9176 - val_loss: 6.6146 - val_accuracy: 0.3585\n",
            "Epoch 1452/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0746 - accuracy: 0.9298 - val_loss: 6.6031 - val_accuracy: 0.3585\n",
            "Epoch 1453/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1110 - accuracy: 0.9189 - val_loss: 6.5940 - val_accuracy: 0.3585\n",
            "Epoch 1454/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1276 - accuracy: 0.8938 - val_loss: 6.6106 - val_accuracy: 0.3774\n",
            "Epoch 1455/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0907 - accuracy: 0.9086 - val_loss: 6.5983 - val_accuracy: 0.3774\n",
            "Epoch 1456/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.0581 - accuracy: 0.9171 - val_loss: 6.5930 - val_accuracy: 0.3962\n",
            "Epoch 1457/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.0721 - accuracy: 0.9069 - val_loss: 6.6032 - val_accuracy: 0.3774\n",
            "Epoch 1458/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1084 - accuracy: 0.9092 - val_loss: 6.6200 - val_accuracy: 0.3585\n",
            "Epoch 1459/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1043 - accuracy: 0.8811 - val_loss: 6.6095 - val_accuracy: 0.3585\n",
            "Epoch 1460/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.0233 - accuracy: 0.9532 - val_loss: 6.5982 - val_accuracy: 0.3962\n",
            "Epoch 1461/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0940 - accuracy: 0.9014 - val_loss: 6.6032 - val_accuracy: 0.3585\n",
            "Epoch 1462/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1066 - accuracy: 0.9026 - val_loss: 6.5963 - val_accuracy: 0.3774\n",
            "Epoch 1463/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0739 - accuracy: 0.9174 - val_loss: 6.5825 - val_accuracy: 0.3774\n",
            "Epoch 1464/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1322 - accuracy: 0.9040 - val_loss: 6.5725 - val_accuracy: 0.3962\n",
            "Epoch 1465/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0880 - accuracy: 0.9153 - val_loss: 6.5836 - val_accuracy: 0.3585\n",
            "Epoch 1466/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0751 - accuracy: 0.9396 - val_loss: 6.5807 - val_accuracy: 0.3585\n",
            "Epoch 1467/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1166 - accuracy: 0.9048 - val_loss: 6.5688 - val_accuracy: 0.3774\n",
            "Epoch 1468/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0868 - accuracy: 0.9404 - val_loss: 6.5724 - val_accuracy: 0.3774\n",
            "Epoch 1469/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0837 - accuracy: 0.9073 - val_loss: 6.5682 - val_accuracy: 0.3774\n",
            "Epoch 1470/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0883 - accuracy: 0.9259 - val_loss: 6.5612 - val_accuracy: 0.3774\n",
            "Epoch 1471/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1010 - accuracy: 0.9041 - val_loss: 6.5792 - val_accuracy: 0.3774\n",
            "Epoch 1472/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1431 - accuracy: 0.8904 - val_loss: 6.5743 - val_accuracy: 0.3774\n",
            "Epoch 1473/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1003 - accuracy: 0.9075 - val_loss: 6.5677 - val_accuracy: 0.3585\n",
            "Epoch 1474/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1226 - accuracy: 0.8873 - val_loss: 6.5580 - val_accuracy: 0.3585\n",
            "Epoch 1475/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0897 - accuracy: 0.9064 - val_loss: 6.5633 - val_accuracy: 0.3585\n",
            "Epoch 1476/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1369 - accuracy: 0.8846 - val_loss: 6.5759 - val_accuracy: 0.3585\n",
            "Epoch 1477/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0848 - accuracy: 0.9067 - val_loss: 6.5834 - val_accuracy: 0.3396\n",
            "Epoch 1478/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1008 - accuracy: 0.9174 - val_loss: 6.5898 - val_accuracy: 0.3208\n",
            "Epoch 1479/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0855 - accuracy: 0.9159 - val_loss: 6.5757 - val_accuracy: 0.3585\n",
            "Epoch 1480/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0962 - accuracy: 0.9198 - val_loss: 6.5882 - val_accuracy: 0.3396\n",
            "Epoch 1481/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.0824 - accuracy: 0.9205 - val_loss: 6.5808 - val_accuracy: 0.3396\n",
            "Epoch 1482/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.0598 - accuracy: 0.9122 - val_loss: 6.5754 - val_accuracy: 0.3585\n",
            "Epoch 1483/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1137 - accuracy: 0.9108 - val_loss: 6.5869 - val_accuracy: 0.3396\n",
            "Epoch 1484/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1056 - accuracy: 0.8967 - val_loss: 6.5775 - val_accuracy: 0.3585\n",
            "Epoch 1485/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0818 - accuracy: 0.9214 - val_loss: 6.5731 - val_accuracy: 0.3585\n",
            "Epoch 1486/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0950 - accuracy: 0.8855 - val_loss: 6.5672 - val_accuracy: 0.3585\n",
            "Epoch 1487/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.0979 - accuracy: 0.9251 - val_loss: 6.5747 - val_accuracy: 0.3585\n",
            "Epoch 1488/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1300 - accuracy: 0.9127 - val_loss: 6.5793 - val_accuracy: 0.3585\n",
            "Epoch 1489/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.0662 - accuracy: 0.9273 - val_loss: 6.5708 - val_accuracy: 0.3585\n",
            "Epoch 1490/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0974 - accuracy: 0.9082 - val_loss: 6.5752 - val_accuracy: 0.3585\n",
            "Epoch 1491/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.1071 - accuracy: 0.8982 - val_loss: 6.6068 - val_accuracy: 0.3585\n",
            "Epoch 1492/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0607 - accuracy: 0.9172 - val_loss: 6.6018 - val_accuracy: 0.3585\n",
            "Epoch 1493/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1436 - accuracy: 0.8761 - val_loss: 6.5945 - val_accuracy: 0.3585\n",
            "Epoch 1494/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0765 - accuracy: 0.9162 - val_loss: 6.5859 - val_accuracy: 0.3396\n",
            "Epoch 1495/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1071 - accuracy: 0.9055 - val_loss: 6.5904 - val_accuracy: 0.3396\n",
            "Epoch 1496/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1041 - accuracy: 0.9230 - val_loss: 6.5784 - val_accuracy: 0.3396\n",
            "Epoch 1497/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1159 - accuracy: 0.9077 - val_loss: 6.5890 - val_accuracy: 0.3396\n",
            "Epoch 1498/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0782 - accuracy: 0.9207 - val_loss: 6.5787 - val_accuracy: 0.3396\n",
            "Epoch 1499/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0813 - accuracy: 0.9333 - val_loss: 6.5826 - val_accuracy: 0.3396\n",
            "Epoch 1500/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.0784 - accuracy: 0.9201 - val_loss: 6.6009 - val_accuracy: 0.3396\n",
            "Epoch 1501/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.1249 - accuracy: 0.9081 - val_loss: 6.5936 - val_accuracy: 0.3585\n",
            "Epoch 1502/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.0868 - accuracy: 0.8813 - val_loss: 6.6219 - val_accuracy: 0.3396\n",
            "Epoch 1503/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0777 - accuracy: 0.9276 - val_loss: 6.6201 - val_accuracy: 0.3396\n",
            "Epoch 1504/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.0944 - accuracy: 0.9271 - val_loss: 6.6416 - val_accuracy: 0.3396\n",
            "Epoch 1505/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.1092 - accuracy: 0.9170 - val_loss: 6.6373 - val_accuracy: 0.3396\n",
            "Epoch 1506/2000\n",
            "5/5 [==============================] - 0s 76ms/step - loss: 5.0655 - accuracy: 0.9193 - val_loss: 6.6451 - val_accuracy: 0.3396\n",
            "Epoch 1507/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0527 - accuracy: 0.9158 - val_loss: 6.6323 - val_accuracy: 0.3585\n",
            "Epoch 1508/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0787 - accuracy: 0.9397 - val_loss: 6.6192 - val_accuracy: 0.3585\n",
            "Epoch 1509/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1086 - accuracy: 0.9070 - val_loss: 6.6126 - val_accuracy: 0.3585\n",
            "Epoch 1510/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0755 - accuracy: 0.9074 - val_loss: 6.6121 - val_accuracy: 0.3585\n",
            "Epoch 1511/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0674 - accuracy: 0.9172 - val_loss: 6.6161 - val_accuracy: 0.3585\n",
            "Epoch 1512/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1078 - accuracy: 0.9079 - val_loss: 6.6447 - val_accuracy: 0.3396\n",
            "Epoch 1513/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0478 - accuracy: 0.9141 - val_loss: 6.6338 - val_accuracy: 0.3585\n",
            "Epoch 1514/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0913 - accuracy: 0.9272 - val_loss: 6.6374 - val_accuracy: 0.3585\n",
            "Epoch 1515/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0941 - accuracy: 0.9057 - val_loss: 6.6261 - val_accuracy: 0.3585\n",
            "Epoch 1516/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0771 - accuracy: 0.8955 - val_loss: 6.6304 - val_accuracy: 0.3585\n",
            "Epoch 1517/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0674 - accuracy: 0.9185 - val_loss: 6.6466 - val_accuracy: 0.3585\n",
            "Epoch 1518/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0604 - accuracy: 0.9464 - val_loss: 6.6529 - val_accuracy: 0.3585\n",
            "Epoch 1519/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0766 - accuracy: 0.9012 - val_loss: 6.6446 - val_accuracy: 0.3585\n",
            "Epoch 1520/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0579 - accuracy: 0.9320 - val_loss: 6.6379 - val_accuracy: 0.3585\n",
            "Epoch 1521/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.1292 - accuracy: 0.8558 - val_loss: 6.6282 - val_accuracy: 0.3585\n",
            "Epoch 1522/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.1150 - accuracy: 0.9040 - val_loss: 6.6150 - val_accuracy: 0.3585\n",
            "Epoch 1523/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0194 - accuracy: 0.9321 - val_loss: 6.6190 - val_accuracy: 0.3585\n",
            "Epoch 1524/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0167 - accuracy: 0.9367 - val_loss: 6.6231 - val_accuracy: 0.3585\n",
            "Epoch 1525/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0352 - accuracy: 0.9209 - val_loss: 6.6254 - val_accuracy: 0.3585\n",
            "Epoch 1526/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.0661 - accuracy: 0.9274 - val_loss: 6.6111 - val_accuracy: 0.3585\n",
            "Epoch 1527/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0520 - accuracy: 0.9267 - val_loss: 6.5976 - val_accuracy: 0.3585\n",
            "Epoch 1528/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0813 - accuracy: 0.9207 - val_loss: 6.5803 - val_accuracy: 0.3585\n",
            "Epoch 1529/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0938 - accuracy: 0.8797 - val_loss: 6.5692 - val_accuracy: 0.3774\n",
            "Epoch 1530/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0495 - accuracy: 0.9254 - val_loss: 6.5591 - val_accuracy: 0.3774\n",
            "Epoch 1531/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.0935 - accuracy: 0.9153 - val_loss: 6.5521 - val_accuracy: 0.3774\n",
            "Epoch 1532/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0454 - accuracy: 0.9485 - val_loss: 6.5468 - val_accuracy: 0.3962\n",
            "Epoch 1533/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0770 - accuracy: 0.9359 - val_loss: 6.5519 - val_accuracy: 0.4151\n",
            "Epoch 1534/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0571 - accuracy: 0.9208 - val_loss: 6.5417 - val_accuracy: 0.4151\n",
            "Epoch 1535/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0688 - accuracy: 0.9101 - val_loss: 6.5540 - val_accuracy: 0.4151\n",
            "Epoch 1536/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0669 - accuracy: 0.9206 - val_loss: 6.5567 - val_accuracy: 0.4151\n",
            "Epoch 1537/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0789 - accuracy: 0.9093 - val_loss: 6.5550 - val_accuracy: 0.3962\n",
            "Epoch 1538/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.1196 - accuracy: 0.8890 - val_loss: 6.5482 - val_accuracy: 0.3962\n",
            "Epoch 1539/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0711 - accuracy: 0.8976 - val_loss: 6.5467 - val_accuracy: 0.3962\n",
            "Epoch 1540/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0372 - accuracy: 0.9328 - val_loss: 6.5430 - val_accuracy: 0.3962\n",
            "Epoch 1541/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.0704 - accuracy: 0.9319 - val_loss: 6.5388 - val_accuracy: 0.3962\n",
            "Epoch 1542/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0305 - accuracy: 0.9451 - val_loss: 6.5475 - val_accuracy: 0.3962\n",
            "Epoch 1543/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0740 - accuracy: 0.9040 - val_loss: 6.5587 - val_accuracy: 0.3774\n",
            "Epoch 1544/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0653 - accuracy: 0.9467 - val_loss: 6.5816 - val_accuracy: 0.3396\n",
            "Epoch 1545/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0876 - accuracy: 0.8990 - val_loss: 6.6018 - val_accuracy: 0.3396\n",
            "Epoch 1546/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0282 - accuracy: 0.9171 - val_loss: 6.6329 - val_accuracy: 0.3585\n",
            "Epoch 1547/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0268 - accuracy: 0.9105 - val_loss: 6.6224 - val_accuracy: 0.3774\n",
            "Epoch 1548/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0838 - accuracy: 0.9024 - val_loss: 6.6079 - val_accuracy: 0.3774\n",
            "Epoch 1549/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0828 - accuracy: 0.8835 - val_loss: 6.6247 - val_accuracy: 0.3774\n",
            "Epoch 1550/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0937 - accuracy: 0.8858 - val_loss: 6.6346 - val_accuracy: 0.3774\n",
            "Epoch 1551/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0703 - accuracy: 0.9141 - val_loss: 6.6498 - val_accuracy: 0.3585\n",
            "Epoch 1552/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0742 - accuracy: 0.9145 - val_loss: 6.6257 - val_accuracy: 0.3774\n",
            "Epoch 1553/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0403 - accuracy: 0.9390 - val_loss: 6.6217 - val_accuracy: 0.3774\n",
            "Epoch 1554/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0310 - accuracy: 0.8970 - val_loss: 6.6187 - val_accuracy: 0.3585\n",
            "Epoch 1555/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0471 - accuracy: 0.9370 - val_loss: 6.6023 - val_accuracy: 0.3585\n",
            "Epoch 1556/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0410 - accuracy: 0.9189 - val_loss: 6.5888 - val_accuracy: 0.3585\n",
            "Epoch 1557/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0692 - accuracy: 0.9201 - val_loss: 6.5785 - val_accuracy: 0.3585\n",
            "Epoch 1558/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0450 - accuracy: 0.9220 - val_loss: 6.5742 - val_accuracy: 0.3774\n",
            "Epoch 1559/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0978 - accuracy: 0.8958 - val_loss: 6.5761 - val_accuracy: 0.3774\n",
            "Epoch 1560/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0344 - accuracy: 0.9255 - val_loss: 6.5696 - val_accuracy: 0.3774\n",
            "Epoch 1561/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0254 - accuracy: 0.9168 - val_loss: 6.5584 - val_accuracy: 0.3962\n",
            "Epoch 1562/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0765 - accuracy: 0.8988 - val_loss: 6.5669 - val_accuracy: 0.3962\n",
            "Epoch 1563/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0473 - accuracy: 0.9022 - val_loss: 6.5599 - val_accuracy: 0.3962\n",
            "Epoch 1564/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0647 - accuracy: 0.9356 - val_loss: 6.5707 - val_accuracy: 0.3774\n",
            "Epoch 1565/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0644 - accuracy: 0.9092 - val_loss: 6.5614 - val_accuracy: 0.3774\n",
            "Epoch 1566/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0682 - accuracy: 0.8911 - val_loss: 6.5586 - val_accuracy: 0.3585\n",
            "Epoch 1567/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9912 - accuracy: 0.9294 - val_loss: 6.5725 - val_accuracy: 0.3585\n",
            "Epoch 1568/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0001 - accuracy: 0.9417 - val_loss: 6.5599 - val_accuracy: 0.3774\n",
            "Epoch 1569/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0551 - accuracy: 0.9141 - val_loss: 6.5481 - val_accuracy: 0.3962\n",
            "Epoch 1570/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.1131 - accuracy: 0.8803 - val_loss: 6.5430 - val_accuracy: 0.3962\n",
            "Epoch 1571/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0560 - accuracy: 0.9159 - val_loss: 6.5292 - val_accuracy: 0.3962\n",
            "Epoch 1572/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.0295 - accuracy: 0.9235 - val_loss: 6.5262 - val_accuracy: 0.4151\n",
            "Epoch 1573/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0930 - accuracy: 0.9288 - val_loss: 6.5431 - val_accuracy: 0.4151\n",
            "Epoch 1574/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0385 - accuracy: 0.9207 - val_loss: 6.5501 - val_accuracy: 0.3962\n",
            "Epoch 1575/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.0438 - accuracy: 0.8906 - val_loss: 6.5419 - val_accuracy: 0.3962\n",
            "Epoch 1576/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0154 - accuracy: 0.9181 - val_loss: 6.5484 - val_accuracy: 0.3962\n",
            "Epoch 1577/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0605 - accuracy: 0.9353 - val_loss: 6.5471 - val_accuracy: 0.4151\n",
            "Epoch 1578/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0601 - accuracy: 0.8988 - val_loss: 6.5488 - val_accuracy: 0.4151\n",
            "Epoch 1579/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0040 - accuracy: 0.9486 - val_loss: 6.5684 - val_accuracy: 0.3774\n",
            "Epoch 1580/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0711 - accuracy: 0.8974 - val_loss: 6.5711 - val_accuracy: 0.3774\n",
            "Epoch 1581/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.0162 - accuracy: 0.9144 - val_loss: 6.5746 - val_accuracy: 0.3774\n",
            "Epoch 1582/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0415 - accuracy: 0.9125 - val_loss: 6.5616 - val_accuracy: 0.3962\n",
            "Epoch 1583/2000\n",
            "5/5 [==============================] - 0s 74ms/step - loss: 5.0409 - accuracy: 0.9190 - val_loss: 6.5713 - val_accuracy: 0.3962\n",
            "Epoch 1584/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.0252 - accuracy: 0.9359 - val_loss: 6.5644 - val_accuracy: 0.3962\n",
            "Epoch 1585/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0472 - accuracy: 0.8942 - val_loss: 6.5541 - val_accuracy: 0.3962\n",
            "Epoch 1586/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0586 - accuracy: 0.8938 - val_loss: 6.5435 - val_accuracy: 0.3962\n",
            "Epoch 1587/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0348 - accuracy: 0.9545 - val_loss: 6.5465 - val_accuracy: 0.3962\n",
            "Epoch 1588/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0293 - accuracy: 0.9412 - val_loss: 6.5511 - val_accuracy: 0.3774\n",
            "Epoch 1589/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0205 - accuracy: 0.9207 - val_loss: 6.5468 - val_accuracy: 0.3962\n",
            "Epoch 1590/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0597 - accuracy: 0.9087 - val_loss: 6.5515 - val_accuracy: 0.3774\n",
            "Epoch 1591/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 5.0589 - accuracy: 0.9031 - val_loss: 6.5409 - val_accuracy: 0.3774\n",
            "Epoch 1592/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0107 - accuracy: 0.9273 - val_loss: 6.5483 - val_accuracy: 0.3962\n",
            "Epoch 1593/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0795 - accuracy: 0.9103 - val_loss: 6.5522 - val_accuracy: 0.3585\n",
            "Epoch 1594/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9806 - accuracy: 0.9503 - val_loss: 6.5577 - val_accuracy: 0.3585\n",
            "Epoch 1595/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0180 - accuracy: 0.9410 - val_loss: 6.5424 - val_accuracy: 0.3585\n",
            "Epoch 1596/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0539 - accuracy: 0.9138 - val_loss: 6.5429 - val_accuracy: 0.3585\n",
            "Epoch 1597/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9860 - accuracy: 0.9356 - val_loss: 6.5447 - val_accuracy: 0.3585\n",
            "Epoch 1598/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0636 - accuracy: 0.8863 - val_loss: 6.5357 - val_accuracy: 0.3774\n",
            "Epoch 1599/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9756 - accuracy: 0.9242 - val_loss: 6.5288 - val_accuracy: 0.3962\n",
            "Epoch 1600/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0403 - accuracy: 0.9432 - val_loss: 6.5252 - val_accuracy: 0.3962\n",
            "Epoch 1601/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9876 - accuracy: 0.9133 - val_loss: 6.5163 - val_accuracy: 0.3962\n",
            "Epoch 1602/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 4.9793 - accuracy: 0.9458 - val_loss: 6.5050 - val_accuracy: 0.3962\n",
            "Epoch 1603/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0546 - accuracy: 0.9103 - val_loss: 6.5123 - val_accuracy: 0.3962\n",
            "Epoch 1604/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0583 - accuracy: 0.9093 - val_loss: 6.5241 - val_accuracy: 0.3962\n",
            "Epoch 1605/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0358 - accuracy: 0.9381 - val_loss: 6.5368 - val_accuracy: 0.3774\n",
            "Epoch 1606/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0268 - accuracy: 0.9063 - val_loss: 6.5297 - val_accuracy: 0.3962\n",
            "Epoch 1607/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0413 - accuracy: 0.9240 - val_loss: 6.5305 - val_accuracy: 0.3585\n",
            "Epoch 1608/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0514 - accuracy: 0.8968 - val_loss: 6.5185 - val_accuracy: 0.3774\n",
            "Epoch 1609/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0367 - accuracy: 0.9176 - val_loss: 6.5098 - val_accuracy: 0.3774\n",
            "Epoch 1610/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0398 - accuracy: 0.9352 - val_loss: 6.5074 - val_accuracy: 0.3962\n",
            "Epoch 1611/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0093 - accuracy: 0.9318 - val_loss: 6.5169 - val_accuracy: 0.3962\n",
            "Epoch 1612/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0595 - accuracy: 0.8994 - val_loss: 6.5077 - val_accuracy: 0.3962\n",
            "Epoch 1613/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0330 - accuracy: 0.9168 - val_loss: 6.5213 - val_accuracy: 0.3962\n",
            "Epoch 1614/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0036 - accuracy: 0.9444 - val_loss: 6.5303 - val_accuracy: 0.3774\n",
            "Epoch 1615/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0504 - accuracy: 0.9098 - val_loss: 6.5366 - val_accuracy: 0.3774\n",
            "Epoch 1616/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0656 - accuracy: 0.9038 - val_loss: 6.5461 - val_accuracy: 0.3585\n",
            "Epoch 1617/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0169 - accuracy: 0.9145 - val_loss: 6.5331 - val_accuracy: 0.3962\n",
            "Epoch 1618/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.0034 - accuracy: 0.9447 - val_loss: 6.5355 - val_accuracy: 0.3962\n",
            "Epoch 1619/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0054 - accuracy: 0.9179 - val_loss: 6.5431 - val_accuracy: 0.3585\n",
            "Epoch 1620/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0271 - accuracy: 0.9064 - val_loss: 6.5288 - val_accuracy: 0.3962\n",
            "Epoch 1621/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0232 - accuracy: 0.9441 - val_loss: 6.5352 - val_accuracy: 0.3585\n",
            "Epoch 1622/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0173 - accuracy: 0.9209 - val_loss: 6.5266 - val_accuracy: 0.3962\n",
            "Epoch 1623/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0476 - accuracy: 0.9080 - val_loss: 6.5204 - val_accuracy: 0.3962\n",
            "Epoch 1624/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0625 - accuracy: 0.8894 - val_loss: 6.5212 - val_accuracy: 0.3774\n",
            "Epoch 1625/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0195 - accuracy: 0.9420 - val_loss: 6.5454 - val_accuracy: 0.3585\n",
            "Epoch 1626/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0394 - accuracy: 0.8986 - val_loss: 6.5341 - val_accuracy: 0.3585\n",
            "Epoch 1627/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0379 - accuracy: 0.9438 - val_loss: 6.5210 - val_accuracy: 0.3962\n",
            "Epoch 1628/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9858 - accuracy: 0.9448 - val_loss: 6.5262 - val_accuracy: 0.3774\n",
            "Epoch 1629/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0001 - accuracy: 0.9328 - val_loss: 6.5352 - val_accuracy: 0.3585\n",
            "Epoch 1630/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0175 - accuracy: 0.9283 - val_loss: 6.5301 - val_accuracy: 0.3585\n",
            "Epoch 1631/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0876 - accuracy: 0.8850 - val_loss: 6.5356 - val_accuracy: 0.3396\n",
            "Epoch 1632/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0323 - accuracy: 0.9092 - val_loss: 6.5237 - val_accuracy: 0.3585\n",
            "Epoch 1633/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0317 - accuracy: 0.9204 - val_loss: 6.5115 - val_accuracy: 0.3774\n",
            "Epoch 1634/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0021 - accuracy: 0.9107 - val_loss: 6.4989 - val_accuracy: 0.3774\n",
            "Epoch 1635/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0539 - accuracy: 0.8911 - val_loss: 6.4921 - val_accuracy: 0.3962\n",
            "Epoch 1636/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0061 - accuracy: 0.9172 - val_loss: 6.4964 - val_accuracy: 0.3962\n",
            "Epoch 1637/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9947 - accuracy: 0.9365 - val_loss: 6.5026 - val_accuracy: 0.3585\n",
            "Epoch 1638/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0259 - accuracy: 0.9130 - val_loss: 6.5079 - val_accuracy: 0.3585\n",
            "Epoch 1639/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0077 - accuracy: 0.9336 - val_loss: 6.5054 - val_accuracy: 0.3585\n",
            "Epoch 1640/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0159 - accuracy: 0.9288 - val_loss: 6.5008 - val_accuracy: 0.3585\n",
            "Epoch 1641/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9821 - accuracy: 0.9174 - val_loss: 6.5024 - val_accuracy: 0.3774\n",
            "Epoch 1642/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9847 - accuracy: 0.9336 - val_loss: 6.4930 - val_accuracy: 0.3962\n",
            "Epoch 1643/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9988 - accuracy: 0.9214 - val_loss: 6.4992 - val_accuracy: 0.3774\n",
            "Epoch 1644/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0242 - accuracy: 0.9249 - val_loss: 6.5312 - val_accuracy: 0.3396\n",
            "Epoch 1645/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9907 - accuracy: 0.9245 - val_loss: 6.5260 - val_accuracy: 0.3208\n",
            "Epoch 1646/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9863 - accuracy: 0.9454 - val_loss: 6.5143 - val_accuracy: 0.3396\n",
            "Epoch 1647/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0058 - accuracy: 0.9053 - val_loss: 6.5030 - val_accuracy: 0.3585\n",
            "Epoch 1648/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0348 - accuracy: 0.9147 - val_loss: 6.5022 - val_accuracy: 0.3585\n",
            "Epoch 1649/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.0405 - accuracy: 0.9402 - val_loss: 6.5223 - val_accuracy: 0.3396\n",
            "Epoch 1650/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9976 - accuracy: 0.9230 - val_loss: 6.5242 - val_accuracy: 0.3396\n",
            "Epoch 1651/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0343 - accuracy: 0.9137 - val_loss: 6.5361 - val_accuracy: 0.3396\n",
            "Epoch 1652/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9646 - accuracy: 0.9528 - val_loss: 6.5380 - val_accuracy: 0.3396\n",
            "Epoch 1653/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0248 - accuracy: 0.9030 - val_loss: 6.5273 - val_accuracy: 0.3396\n",
            "Epoch 1654/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0010 - accuracy: 0.9312 - val_loss: 6.5403 - val_accuracy: 0.3396\n",
            "Epoch 1655/2000\n",
            "5/5 [==============================] - 0s 75ms/step - loss: 5.0258 - accuracy: 0.8987 - val_loss: 6.5287 - val_accuracy: 0.3774\n",
            "Epoch 1656/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0040 - accuracy: 0.9127 - val_loss: 6.5149 - val_accuracy: 0.3774\n",
            "Epoch 1657/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0018 - accuracy: 0.9431 - val_loss: 6.5267 - val_accuracy: 0.3774\n",
            "Epoch 1658/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0036 - accuracy: 0.9295 - val_loss: 6.5219 - val_accuracy: 0.3774\n",
            "Epoch 1659/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0022 - accuracy: 0.9215 - val_loss: 6.5074 - val_accuracy: 0.3774\n",
            "Epoch 1660/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9614 - accuracy: 0.9275 - val_loss: 6.5094 - val_accuracy: 0.3962\n",
            "Epoch 1661/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0178 - accuracy: 0.9061 - val_loss: 6.5161 - val_accuracy: 0.3774\n",
            "Epoch 1662/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9744 - accuracy: 0.9323 - val_loss: 6.5202 - val_accuracy: 0.3774\n",
            "Epoch 1663/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9967 - accuracy: 0.9244 - val_loss: 6.5106 - val_accuracy: 0.4151\n",
            "Epoch 1664/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0308 - accuracy: 0.9103 - val_loss: 6.5024 - val_accuracy: 0.3962\n",
            "Epoch 1665/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9724 - accuracy: 0.9463 - val_loss: 6.4947 - val_accuracy: 0.4151\n",
            "Epoch 1666/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9988 - accuracy: 0.9201 - val_loss: 6.4909 - val_accuracy: 0.4151\n",
            "Epoch 1667/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 5.0180 - accuracy: 0.9127 - val_loss: 6.4798 - val_accuracy: 0.4151\n",
            "Epoch 1668/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 5.0030 - accuracy: 0.9230 - val_loss: 6.4679 - val_accuracy: 0.4151\n",
            "Epoch 1669/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0424 - accuracy: 0.8898 - val_loss: 6.4664 - val_accuracy: 0.4151\n",
            "Epoch 1670/2000\n",
            "5/5 [==============================] - 0s 48ms/step - loss: 5.0049 - accuracy: 0.9203 - val_loss: 6.4599 - val_accuracy: 0.4151\n",
            "Epoch 1671/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9813 - accuracy: 0.9333 - val_loss: 6.4539 - val_accuracy: 0.3962\n",
            "Epoch 1672/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0273 - accuracy: 0.9070 - val_loss: 6.4590 - val_accuracy: 0.3962\n",
            "Epoch 1673/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9915 - accuracy: 0.9102 - val_loss: 6.4635 - val_accuracy: 0.3962\n",
            "Epoch 1674/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0025 - accuracy: 0.9138 - val_loss: 6.4599 - val_accuracy: 0.3962\n",
            "Epoch 1675/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0190 - accuracy: 0.9324 - val_loss: 6.4608 - val_accuracy: 0.3962\n",
            "Epoch 1676/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9923 - accuracy: 0.9213 - val_loss: 6.4570 - val_accuracy: 0.3962\n",
            "Epoch 1677/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0020 - accuracy: 0.9106 - val_loss: 6.4772 - val_accuracy: 0.3962\n",
            "Epoch 1678/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0233 - accuracy: 0.8971 - val_loss: 6.4745 - val_accuracy: 0.3962\n",
            "Epoch 1679/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9807 - accuracy: 0.9229 - val_loss: 6.4755 - val_accuracy: 0.3962\n",
            "Epoch 1680/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0221 - accuracy: 0.8931 - val_loss: 6.4874 - val_accuracy: 0.3962\n",
            "Epoch 1681/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9809 - accuracy: 0.9244 - val_loss: 6.4893 - val_accuracy: 0.3962\n",
            "Epoch 1682/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0361 - accuracy: 0.9241 - val_loss: 6.4826 - val_accuracy: 0.3962\n",
            "Epoch 1683/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9922 - accuracy: 0.9215 - val_loss: 6.5066 - val_accuracy: 0.3774\n",
            "Epoch 1684/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0100 - accuracy: 0.9260 - val_loss: 6.5019 - val_accuracy: 0.3962\n",
            "Epoch 1685/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9745 - accuracy: 0.9193 - val_loss: 6.4940 - val_accuracy: 0.3962\n",
            "Epoch 1686/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9846 - accuracy: 0.9437 - val_loss: 6.4869 - val_accuracy: 0.3962\n",
            "Epoch 1687/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9759 - accuracy: 0.9371 - val_loss: 6.4955 - val_accuracy: 0.3962\n",
            "Epoch 1688/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.0119 - accuracy: 0.8988 - val_loss: 6.4887 - val_accuracy: 0.3962\n",
            "Epoch 1689/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9618 - accuracy: 0.9324 - val_loss: 6.4835 - val_accuracy: 0.3962\n",
            "Epoch 1690/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0387 - accuracy: 0.9238 - val_loss: 6.4927 - val_accuracy: 0.3962\n",
            "Epoch 1691/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 4.9953 - accuracy: 0.9232 - val_loss: 6.5154 - val_accuracy: 0.3585\n",
            "Epoch 1692/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9852 - accuracy: 0.9149 - val_loss: 6.5047 - val_accuracy: 0.3774\n",
            "Epoch 1693/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9445 - accuracy: 0.9457 - val_loss: 6.4956 - val_accuracy: 0.3774\n",
            "Epoch 1694/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9795 - accuracy: 0.9547 - val_loss: 6.5010 - val_accuracy: 0.3774\n",
            "Epoch 1695/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9993 - accuracy: 0.9216 - val_loss: 6.4952 - val_accuracy: 0.3774\n",
            "Epoch 1696/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9928 - accuracy: 0.9194 - val_loss: 6.5006 - val_accuracy: 0.3585\n",
            "Epoch 1697/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9538 - accuracy: 0.9419 - val_loss: 6.5051 - val_accuracy: 0.3585\n",
            "Epoch 1698/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9908 - accuracy: 0.9509 - val_loss: 6.5107 - val_accuracy: 0.3585\n",
            "Epoch 1699/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9729 - accuracy: 0.9241 - val_loss: 6.4960 - val_accuracy: 0.3585\n",
            "Epoch 1700/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9685 - accuracy: 0.9426 - val_loss: 6.5032 - val_accuracy: 0.3585\n",
            "Epoch 1701/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9925 - accuracy: 0.9367 - val_loss: 6.5181 - val_accuracy: 0.3585\n",
            "Epoch 1702/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9964 - accuracy: 0.8954 - val_loss: 6.5017 - val_accuracy: 0.3585\n",
            "Epoch 1703/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9830 - accuracy: 0.9080 - val_loss: 6.4942 - val_accuracy: 0.3585\n",
            "Epoch 1704/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9580 - accuracy: 0.9592 - val_loss: 6.4958 - val_accuracy: 0.3585\n",
            "Epoch 1705/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9811 - accuracy: 0.9109 - val_loss: 6.4964 - val_accuracy: 0.3585\n",
            "Epoch 1706/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0265 - accuracy: 0.9084 - val_loss: 6.4910 - val_accuracy: 0.3585\n",
            "Epoch 1707/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9665 - accuracy: 0.9176 - val_loss: 6.4849 - val_accuracy: 0.3774\n",
            "Epoch 1708/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0137 - accuracy: 0.9125 - val_loss: 6.4833 - val_accuracy: 0.3774\n",
            "Epoch 1709/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0311 - accuracy: 0.9211 - val_loss: 6.4915 - val_accuracy: 0.3585\n",
            "Epoch 1710/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9894 - accuracy: 0.9222 - val_loss: 6.5248 - val_accuracy: 0.3585\n",
            "Epoch 1711/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0285 - accuracy: 0.9011 - val_loss: 6.5248 - val_accuracy: 0.3585\n",
            "Epoch 1712/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9634 - accuracy: 0.9370 - val_loss: 6.5408 - val_accuracy: 0.3585\n",
            "Epoch 1713/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 4.9520 - accuracy: 0.9311 - val_loss: 6.5365 - val_accuracy: 0.3585\n",
            "Epoch 1714/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0030 - accuracy: 0.9364 - val_loss: 6.5208 - val_accuracy: 0.3585\n",
            "Epoch 1715/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 5.0413 - accuracy: 0.8682 - val_loss: 6.5214 - val_accuracy: 0.3585\n",
            "Epoch 1716/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9859 - accuracy: 0.9301 - val_loss: 6.5101 - val_accuracy: 0.3585\n",
            "Epoch 1717/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9872 - accuracy: 0.9112 - val_loss: 6.5203 - val_accuracy: 0.3585\n",
            "Epoch 1718/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9805 - accuracy: 0.9219 - val_loss: 6.5048 - val_accuracy: 0.3585\n",
            "Epoch 1719/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9859 - accuracy: 0.9279 - val_loss: 6.5021 - val_accuracy: 0.3585\n",
            "Epoch 1720/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9945 - accuracy: 0.9302 - val_loss: 6.4919 - val_accuracy: 0.3396\n",
            "Epoch 1721/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0206 - accuracy: 0.9004 - val_loss: 6.4948 - val_accuracy: 0.3396\n",
            "Epoch 1722/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0118 - accuracy: 0.9092 - val_loss: 6.5271 - val_accuracy: 0.3396\n",
            "Epoch 1723/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9829 - accuracy: 0.9146 - val_loss: 6.5184 - val_accuracy: 0.3396\n",
            "Epoch 1724/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9704 - accuracy: 0.9175 - val_loss: 6.5236 - val_accuracy: 0.3396\n",
            "Epoch 1725/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0129 - accuracy: 0.9058 - val_loss: 6.5079 - val_accuracy: 0.3396\n",
            "Epoch 1726/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 4.9655 - accuracy: 0.9402 - val_loss: 6.4992 - val_accuracy: 0.3396\n",
            "Epoch 1727/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9693 - accuracy: 0.9323 - val_loss: 6.5032 - val_accuracy: 0.3396\n",
            "Epoch 1728/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9478 - accuracy: 0.9287 - val_loss: 6.4895 - val_accuracy: 0.3396\n",
            "Epoch 1729/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9830 - accuracy: 0.9060 - val_loss: 6.5171 - val_accuracy: 0.3396\n",
            "Epoch 1730/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9781 - accuracy: 0.9249 - val_loss: 6.5274 - val_accuracy: 0.3396\n",
            "Epoch 1731/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9642 - accuracy: 0.9140 - val_loss: 6.5522 - val_accuracy: 0.3585\n",
            "Epoch 1732/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9909 - accuracy: 0.9229 - val_loss: 6.5432 - val_accuracy: 0.3585\n",
            "Epoch 1733/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9653 - accuracy: 0.9178 - val_loss: 6.5328 - val_accuracy: 0.3585\n",
            "Epoch 1734/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9758 - accuracy: 0.9088 - val_loss: 6.5219 - val_accuracy: 0.3585\n",
            "Epoch 1735/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9803 - accuracy: 0.9405 - val_loss: 6.5240 - val_accuracy: 0.3585\n",
            "Epoch 1736/2000\n",
            "5/5 [==============================] - 0s 75ms/step - loss: 4.9962 - accuracy: 0.9219 - val_loss: 6.5275 - val_accuracy: 0.3396\n",
            "Epoch 1737/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9518 - accuracy: 0.9269 - val_loss: 6.5108 - val_accuracy: 0.3962\n",
            "Epoch 1738/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9545 - accuracy: 0.9180 - val_loss: 6.5011 - val_accuracy: 0.3962\n",
            "Epoch 1739/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9947 - accuracy: 0.9201 - val_loss: 6.4868 - val_accuracy: 0.3962\n",
            "Epoch 1740/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0318 - accuracy: 0.9261 - val_loss: 6.4961 - val_accuracy: 0.3774\n",
            "Epoch 1741/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9932 - accuracy: 0.9080 - val_loss: 6.4843 - val_accuracy: 0.3774\n",
            "Epoch 1742/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9709 - accuracy: 0.9345 - val_loss: 6.4889 - val_accuracy: 0.3774\n",
            "Epoch 1743/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0059 - accuracy: 0.9144 - val_loss: 6.4952 - val_accuracy: 0.3774\n",
            "Epoch 1744/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0262 - accuracy: 0.8858 - val_loss: 6.5039 - val_accuracy: 0.3585\n",
            "Epoch 1745/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9785 - accuracy: 0.9284 - val_loss: 6.4900 - val_accuracy: 0.3585\n",
            "Epoch 1746/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9244 - accuracy: 0.9351 - val_loss: 6.4958 - val_accuracy: 0.3585\n",
            "Epoch 1747/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9727 - accuracy: 0.9053 - val_loss: 6.4998 - val_accuracy: 0.3585\n",
            "Epoch 1748/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9891 - accuracy: 0.9183 - val_loss: 6.5179 - val_accuracy: 0.3585\n",
            "Epoch 1749/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0304 - accuracy: 0.9020 - val_loss: 6.5230 - val_accuracy: 0.3396\n",
            "Epoch 1750/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9694 - accuracy: 0.9403 - val_loss: 6.5221 - val_accuracy: 0.3396\n",
            "Epoch 1751/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9522 - accuracy: 0.9289 - val_loss: 6.5186 - val_accuracy: 0.3396\n",
            "Epoch 1752/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9644 - accuracy: 0.9091 - val_loss: 6.5047 - val_accuracy: 0.3396\n",
            "Epoch 1753/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9979 - accuracy: 0.9103 - val_loss: 6.5286 - val_accuracy: 0.3396\n",
            "Epoch 1754/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9823 - accuracy: 0.9027 - val_loss: 6.5204 - val_accuracy: 0.3396\n",
            "Epoch 1755/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9416 - accuracy: 0.9371 - val_loss: 6.5122 - val_accuracy: 0.3585\n",
            "Epoch 1756/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0017 - accuracy: 0.9093 - val_loss: 6.5094 - val_accuracy: 0.3585\n",
            "Epoch 1757/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.8958 - accuracy: 0.9405 - val_loss: 6.5038 - val_accuracy: 0.3585\n",
            "Epoch 1758/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9339 - accuracy: 0.9305 - val_loss: 6.4925 - val_accuracy: 0.3585\n",
            "Epoch 1759/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0103 - accuracy: 0.9146 - val_loss: 6.4980 - val_accuracy: 0.3585\n",
            "Epoch 1760/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9867 - accuracy: 0.9111 - val_loss: 6.4904 - val_accuracy: 0.3585\n",
            "Epoch 1761/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9552 - accuracy: 0.9190 - val_loss: 6.4953 - val_accuracy: 0.3585\n",
            "Epoch 1762/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9457 - accuracy: 0.9362 - val_loss: 6.4953 - val_accuracy: 0.3585\n",
            "Epoch 1763/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9482 - accuracy: 0.9255 - val_loss: 6.5136 - val_accuracy: 0.3585\n",
            "Epoch 1764/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9512 - accuracy: 0.9345 - val_loss: 6.5046 - val_accuracy: 0.3585\n",
            "Epoch 1765/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 5.0051 - accuracy: 0.8915 - val_loss: 6.5241 - val_accuracy: 0.3774\n",
            "Epoch 1766/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9875 - accuracy: 0.9170 - val_loss: 6.5225 - val_accuracy: 0.3774\n",
            "Epoch 1767/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9590 - accuracy: 0.9389 - val_loss: 6.5037 - val_accuracy: 0.3585\n",
            "Epoch 1768/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9656 - accuracy: 0.9094 - val_loss: 6.4903 - val_accuracy: 0.3585\n",
            "Epoch 1769/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9877 - accuracy: 0.9110 - val_loss: 6.4751 - val_accuracy: 0.3774\n",
            "Epoch 1770/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9905 - accuracy: 0.9065 - val_loss: 6.4686 - val_accuracy: 0.3774\n",
            "Epoch 1771/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9299 - accuracy: 0.9323 - val_loss: 6.4741 - val_accuracy: 0.3774\n",
            "Epoch 1772/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9604 - accuracy: 0.9261 - val_loss: 6.4877 - val_accuracy: 0.3585\n",
            "Epoch 1773/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9847 - accuracy: 0.9049 - val_loss: 6.5167 - val_accuracy: 0.3585\n",
            "Epoch 1774/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9903 - accuracy: 0.9062 - val_loss: 6.5224 - val_accuracy: 0.3585\n",
            "Epoch 1775/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9468 - accuracy: 0.9564 - val_loss: 6.5112 - val_accuracy: 0.3585\n",
            "Epoch 1776/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9997 - accuracy: 0.9058 - val_loss: 6.5174 - val_accuracy: 0.3585\n",
            "Epoch 1777/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9652 - accuracy: 0.9060 - val_loss: 6.5043 - val_accuracy: 0.3585\n",
            "Epoch 1778/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9552 - accuracy: 0.9281 - val_loss: 6.5046 - val_accuracy: 0.3585\n",
            "Epoch 1779/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9800 - accuracy: 0.9210 - val_loss: 6.5109 - val_accuracy: 0.3585\n",
            "Epoch 1780/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9656 - accuracy: 0.9083 - val_loss: 6.4986 - val_accuracy: 0.3585\n",
            "Epoch 1781/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 5.0196 - accuracy: 0.9123 - val_loss: 6.4925 - val_accuracy: 0.3585\n",
            "Epoch 1782/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9165 - accuracy: 0.9324 - val_loss: 6.4979 - val_accuracy: 0.3585\n",
            "Epoch 1783/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 5.0027 - accuracy: 0.9232 - val_loss: 6.4956 - val_accuracy: 0.3585\n",
            "Epoch 1784/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0312 - accuracy: 0.8842 - val_loss: 6.4976 - val_accuracy: 0.3585\n",
            "Epoch 1785/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9272 - accuracy: 0.9540 - val_loss: 6.5072 - val_accuracy: 0.3585\n",
            "Epoch 1786/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9745 - accuracy: 0.9190 - val_loss: 6.4954 - val_accuracy: 0.3585\n",
            "Epoch 1787/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0076 - accuracy: 0.8838 - val_loss: 6.5112 - val_accuracy: 0.3585\n",
            "Epoch 1788/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9391 - accuracy: 0.9253 - val_loss: 6.5283 - val_accuracy: 0.3396\n",
            "Epoch 1789/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9391 - accuracy: 0.9256 - val_loss: 6.5121 - val_accuracy: 0.3585\n",
            "Epoch 1790/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9495 - accuracy: 0.9401 - val_loss: 6.5192 - val_accuracy: 0.3585\n",
            "Epoch 1791/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9829 - accuracy: 0.9117 - val_loss: 6.5618 - val_accuracy: 0.3396\n",
            "Epoch 1792/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9571 - accuracy: 0.9045 - val_loss: 6.5569 - val_accuracy: 0.3396\n",
            "Epoch 1793/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9728 - accuracy: 0.9243 - val_loss: 6.5504 - val_accuracy: 0.3396\n",
            "Epoch 1794/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9398 - accuracy: 0.9294 - val_loss: 6.5666 - val_accuracy: 0.3396\n",
            "Epoch 1795/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 5.0029 - accuracy: 0.9041 - val_loss: 6.5960 - val_accuracy: 0.3208\n",
            "Epoch 1796/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9960 - accuracy: 0.8890 - val_loss: 6.5833 - val_accuracy: 0.3585\n",
            "Epoch 1797/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9680 - accuracy: 0.9231 - val_loss: 6.5698 - val_accuracy: 0.3585\n",
            "Epoch 1798/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9992 - accuracy: 0.9253 - val_loss: 6.5492 - val_accuracy: 0.3585\n",
            "Epoch 1799/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9811 - accuracy: 0.9405 - val_loss: 6.5332 - val_accuracy: 0.3396\n",
            "Epoch 1800/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9292 - accuracy: 0.9352 - val_loss: 6.5181 - val_accuracy: 0.3396\n",
            "Epoch 1801/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9696 - accuracy: 0.9294 - val_loss: 6.5067 - val_accuracy: 0.3396\n",
            "Epoch 1802/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9430 - accuracy: 0.9426 - val_loss: 6.5190 - val_accuracy: 0.3396\n",
            "Epoch 1803/2000\n",
            "5/5 [==============================] - 0s 47ms/step - loss: 4.9696 - accuracy: 0.9117 - val_loss: 6.5125 - val_accuracy: 0.3585\n",
            "Epoch 1804/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9669 - accuracy: 0.9134 - val_loss: 6.5311 - val_accuracy: 0.3585\n",
            "Epoch 1805/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9994 - accuracy: 0.8853 - val_loss: 6.5379 - val_accuracy: 0.3585\n",
            "Epoch 1806/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9444 - accuracy: 0.9262 - val_loss: 6.5271 - val_accuracy: 0.3585\n",
            "Epoch 1807/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9065 - accuracy: 0.9287 - val_loss: 6.5335 - val_accuracy: 0.3585\n",
            "Epoch 1808/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9809 - accuracy: 0.9252 - val_loss: 6.5307 - val_accuracy: 0.3585\n",
            "Epoch 1809/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9506 - accuracy: 0.9077 - val_loss: 6.5116 - val_accuracy: 0.3585\n",
            "Epoch 1810/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9537 - accuracy: 0.9165 - val_loss: 6.4948 - val_accuracy: 0.3585\n",
            "Epoch 1811/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9204 - accuracy: 0.9460 - val_loss: 6.4987 - val_accuracy: 0.3585\n",
            "Epoch 1812/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9733 - accuracy: 0.9013 - val_loss: 6.5038 - val_accuracy: 0.3585\n",
            "Epoch 1813/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9424 - accuracy: 0.9229 - val_loss: 6.5062 - val_accuracy: 0.3585\n",
            "Epoch 1814/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9403 - accuracy: 0.9143 - val_loss: 6.4969 - val_accuracy: 0.3585\n",
            "Epoch 1815/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9626 - accuracy: 0.9169 - val_loss: 6.4891 - val_accuracy: 0.3774\n",
            "Epoch 1816/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9388 - accuracy: 0.9324 - val_loss: 6.4957 - val_accuracy: 0.3774\n",
            "Epoch 1817/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9452 - accuracy: 0.9263 - val_loss: 6.4961 - val_accuracy: 0.3774\n",
            "Epoch 1818/2000\n",
            "5/5 [==============================] - 0s 77ms/step - loss: 4.9296 - accuracy: 0.9181 - val_loss: 6.4802 - val_accuracy: 0.3774\n",
            "Epoch 1819/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9421 - accuracy: 0.9334 - val_loss: 6.4703 - val_accuracy: 0.3962\n",
            "Epoch 1820/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9701 - accuracy: 0.9093 - val_loss: 6.4666 - val_accuracy: 0.3962\n",
            "Epoch 1821/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 5.0055 - accuracy: 0.8952 - val_loss: 6.4701 - val_accuracy: 0.3962\n",
            "Epoch 1822/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9610 - accuracy: 0.9256 - val_loss: 6.4670 - val_accuracy: 0.3962\n",
            "Epoch 1823/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8962 - accuracy: 0.9648 - val_loss: 6.5013 - val_accuracy: 0.3585\n",
            "Epoch 1824/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9621 - accuracy: 0.9041 - val_loss: 6.4853 - val_accuracy: 0.3774\n",
            "Epoch 1825/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9045 - accuracy: 0.9487 - val_loss: 6.4888 - val_accuracy: 0.3774\n",
            "Epoch 1826/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9242 - accuracy: 0.9504 - val_loss: 6.4950 - val_accuracy: 0.3585\n",
            "Epoch 1827/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9654 - accuracy: 0.9314 - val_loss: 6.4782 - val_accuracy: 0.3774\n",
            "Epoch 1828/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9565 - accuracy: 0.9349 - val_loss: 6.4825 - val_accuracy: 0.3585\n",
            "Epoch 1829/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9352 - accuracy: 0.9460 - val_loss: 6.4898 - val_accuracy: 0.3585\n",
            "Epoch 1830/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9555 - accuracy: 0.9249 - val_loss: 6.4791 - val_accuracy: 0.3396\n",
            "Epoch 1831/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9712 - accuracy: 0.9144 - val_loss: 6.4798 - val_accuracy: 0.3396\n",
            "Epoch 1832/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9529 - accuracy: 0.9311 - val_loss: 6.4636 - val_accuracy: 0.3396\n",
            "Epoch 1833/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8700 - accuracy: 0.9355 - val_loss: 6.4526 - val_accuracy: 0.3396\n",
            "Epoch 1834/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8984 - accuracy: 0.9423 - val_loss: 6.4535 - val_accuracy: 0.3396\n",
            "Epoch 1835/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9122 - accuracy: 0.9499 - val_loss: 6.4756 - val_accuracy: 0.3585\n",
            "Epoch 1836/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9494 - accuracy: 0.9241 - val_loss: 6.4770 - val_accuracy: 0.3774\n",
            "Epoch 1837/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9135 - accuracy: 0.9220 - val_loss: 6.4829 - val_accuracy: 0.3774\n",
            "Epoch 1838/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9496 - accuracy: 0.9199 - val_loss: 6.5058 - val_accuracy: 0.3585\n",
            "Epoch 1839/2000\n",
            "5/5 [==============================] - 0s 48ms/step - loss: 4.9327 - accuracy: 0.9303 - val_loss: 6.5077 - val_accuracy: 0.3585\n",
            "Epoch 1840/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9468 - accuracy: 0.9265 - val_loss: 6.4902 - val_accuracy: 0.3774\n",
            "Epoch 1841/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9501 - accuracy: 0.9317 - val_loss: 6.4799 - val_accuracy: 0.3774\n",
            "Epoch 1842/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8981 - accuracy: 0.9218 - val_loss: 6.4640 - val_accuracy: 0.3774\n",
            "Epoch 1843/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9610 - accuracy: 0.9205 - val_loss: 6.4598 - val_accuracy: 0.3774\n",
            "Epoch 1844/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 4.9273 - accuracy: 0.9240 - val_loss: 6.4834 - val_accuracy: 0.3774\n",
            "Epoch 1845/2000\n",
            "5/5 [==============================] - 0s 51ms/step - loss: 4.9520 - accuracy: 0.9199 - val_loss: 6.4727 - val_accuracy: 0.3774\n",
            "Epoch 1846/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 4.9798 - accuracy: 0.8942 - val_loss: 6.5036 - val_accuracy: 0.3585\n",
            "Epoch 1847/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8779 - accuracy: 0.9483 - val_loss: 6.4889 - val_accuracy: 0.3774\n",
            "Epoch 1848/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9264 - accuracy: 0.9610 - val_loss: 6.4796 - val_accuracy: 0.3774\n",
            "Epoch 1849/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.8903 - accuracy: 0.9308 - val_loss: 6.4702 - val_accuracy: 0.3774\n",
            "Epoch 1850/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9390 - accuracy: 0.9116 - val_loss: 6.4565 - val_accuracy: 0.3962\n",
            "Epoch 1851/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9797 - accuracy: 0.9151 - val_loss: 6.4566 - val_accuracy: 0.3962\n",
            "Epoch 1852/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9353 - accuracy: 0.9201 - val_loss: 6.4677 - val_accuracy: 0.3774\n",
            "Epoch 1853/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9533 - accuracy: 0.9095 - val_loss: 6.4748 - val_accuracy: 0.3774\n",
            "Epoch 1854/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8990 - accuracy: 0.9294 - val_loss: 6.4633 - val_accuracy: 0.3962\n",
            "Epoch 1855/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9544 - accuracy: 0.9159 - val_loss: 6.4631 - val_accuracy: 0.3962\n",
            "Epoch 1856/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9692 - accuracy: 0.9270 - val_loss: 6.4634 - val_accuracy: 0.3962\n",
            "Epoch 1857/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9184 - accuracy: 0.9377 - val_loss: 6.4525 - val_accuracy: 0.3962\n",
            "Epoch 1858/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9790 - accuracy: 0.9068 - val_loss: 6.4402 - val_accuracy: 0.3962\n",
            "Epoch 1859/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9258 - accuracy: 0.9510 - val_loss: 6.4367 - val_accuracy: 0.3962\n",
            "Epoch 1860/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 4.9011 - accuracy: 0.9381 - val_loss: 6.4274 - val_accuracy: 0.3962\n",
            "Epoch 1861/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9475 - accuracy: 0.9335 - val_loss: 6.4237 - val_accuracy: 0.3962\n",
            "Epoch 1862/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9257 - accuracy: 0.9173 - val_loss: 6.4329 - val_accuracy: 0.3962\n",
            "Epoch 1863/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9507 - accuracy: 0.9184 - val_loss: 6.4636 - val_accuracy: 0.3962\n",
            "Epoch 1864/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8969 - accuracy: 0.9333 - val_loss: 6.4661 - val_accuracy: 0.3962\n",
            "Epoch 1865/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9481 - accuracy: 0.9180 - val_loss: 6.4695 - val_accuracy: 0.3774\n",
            "Epoch 1866/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9107 - accuracy: 0.9488 - val_loss: 6.4549 - val_accuracy: 0.3962\n",
            "Epoch 1867/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9486 - accuracy: 0.9296 - val_loss: 6.4399 - val_accuracy: 0.3962\n",
            "Epoch 1868/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9153 - accuracy: 0.8942 - val_loss: 6.4259 - val_accuracy: 0.3962\n",
            "Epoch 1869/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9556 - accuracy: 0.9151 - val_loss: 6.4141 - val_accuracy: 0.4151\n",
            "Epoch 1870/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9740 - accuracy: 0.9148 - val_loss: 6.4095 - val_accuracy: 0.4151\n",
            "Epoch 1871/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 4.9635 - accuracy: 0.8939 - val_loss: 6.4034 - val_accuracy: 0.4151\n",
            "Epoch 1872/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8848 - accuracy: 0.9319 - val_loss: 6.4060 - val_accuracy: 0.4151\n",
            "Epoch 1873/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9432 - accuracy: 0.9302 - val_loss: 6.3970 - val_accuracy: 0.4340\n",
            "Epoch 1874/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 4.9011 - accuracy: 0.9344 - val_loss: 6.4169 - val_accuracy: 0.4151\n",
            "Epoch 1875/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9280 - accuracy: 0.9442 - val_loss: 6.4305 - val_accuracy: 0.3962\n",
            "Epoch 1876/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9140 - accuracy: 0.9272 - val_loss: 6.4209 - val_accuracy: 0.3962\n",
            "Epoch 1877/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 4.9435 - accuracy: 0.9298 - val_loss: 6.4123 - val_accuracy: 0.4151\n",
            "Epoch 1878/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9079 - accuracy: 0.9527 - val_loss: 6.4281 - val_accuracy: 0.3962\n",
            "Epoch 1879/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9558 - accuracy: 0.9130 - val_loss: 6.4190 - val_accuracy: 0.3962\n",
            "Epoch 1880/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9592 - accuracy: 0.8992 - val_loss: 6.4372 - val_accuracy: 0.3962\n",
            "Epoch 1881/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9379 - accuracy: 0.9285 - val_loss: 6.4604 - val_accuracy: 0.3774\n",
            "Epoch 1882/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9137 - accuracy: 0.9312 - val_loss: 6.4516 - val_accuracy: 0.3962\n",
            "Epoch 1883/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9285 - accuracy: 0.9342 - val_loss: 6.4363 - val_accuracy: 0.3962\n",
            "Epoch 1884/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9018 - accuracy: 0.9561 - val_loss: 6.4236 - val_accuracy: 0.3962\n",
            "Epoch 1885/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8911 - accuracy: 0.9477 - val_loss: 6.4132 - val_accuracy: 0.4151\n",
            "Epoch 1886/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9415 - accuracy: 0.9232 - val_loss: 6.4178 - val_accuracy: 0.4151\n",
            "Epoch 1887/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9270 - accuracy: 0.9281 - val_loss: 6.4338 - val_accuracy: 0.3962\n",
            "Epoch 1888/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 4.8921 - accuracy: 0.9265 - val_loss: 6.4181 - val_accuracy: 0.4151\n",
            "Epoch 1889/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.8953 - accuracy: 0.9257 - val_loss: 6.4041 - val_accuracy: 0.4151\n",
            "Epoch 1890/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.8985 - accuracy: 0.9379 - val_loss: 6.3968 - val_accuracy: 0.4151\n",
            "Epoch 1891/2000\n",
            "5/5 [==============================] - 0s 95ms/step - loss: 4.9633 - accuracy: 0.9143 - val_loss: 6.3874 - val_accuracy: 0.3962\n",
            "Epoch 1892/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.8761 - accuracy: 0.9448 - val_loss: 6.4150 - val_accuracy: 0.4151\n",
            "Epoch 1893/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 4.8963 - accuracy: 0.9574 - val_loss: 6.4040 - val_accuracy: 0.4151\n",
            "Epoch 1894/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9072 - accuracy: 0.9368 - val_loss: 6.4063 - val_accuracy: 0.4151\n",
            "Epoch 1895/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9417 - accuracy: 0.9022 - val_loss: 6.4122 - val_accuracy: 0.4151\n",
            "Epoch 1896/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9226 - accuracy: 0.9357 - val_loss: 6.4174 - val_accuracy: 0.4151\n",
            "Epoch 1897/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9040 - accuracy: 0.9328 - val_loss: 6.4179 - val_accuracy: 0.4151\n",
            "Epoch 1898/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9268 - accuracy: 0.9417 - val_loss: 6.4111 - val_accuracy: 0.4151\n",
            "Epoch 1899/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8965 - accuracy: 0.9407 - val_loss: 6.4109 - val_accuracy: 0.4151\n",
            "Epoch 1900/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9622 - accuracy: 0.8920 - val_loss: 6.4034 - val_accuracy: 0.4151\n",
            "Epoch 1901/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9260 - accuracy: 0.9354 - val_loss: 6.3956 - val_accuracy: 0.4151\n",
            "Epoch 1902/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 4.9555 - accuracy: 0.9179 - val_loss: 6.3847 - val_accuracy: 0.4151\n",
            "Epoch 1903/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9110 - accuracy: 0.9470 - val_loss: 6.3778 - val_accuracy: 0.4151\n",
            "Epoch 1904/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9205 - accuracy: 0.9427 - val_loss: 6.3980 - val_accuracy: 0.4151\n",
            "Epoch 1905/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.8891 - accuracy: 0.9548 - val_loss: 6.3939 - val_accuracy: 0.4151\n",
            "Epoch 1906/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9233 - accuracy: 0.9195 - val_loss: 6.3990 - val_accuracy: 0.4151\n",
            "Epoch 1907/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8873 - accuracy: 0.9394 - val_loss: 6.3988 - val_accuracy: 0.4151\n",
            "Epoch 1908/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9395 - accuracy: 0.9093 - val_loss: 6.4001 - val_accuracy: 0.4151\n",
            "Epoch 1909/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9187 - accuracy: 0.8977 - val_loss: 6.3965 - val_accuracy: 0.4151\n",
            "Epoch 1910/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9280 - accuracy: 0.9531 - val_loss: 6.4031 - val_accuracy: 0.4151\n",
            "Epoch 1911/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9226 - accuracy: 0.9168 - val_loss: 6.4085 - val_accuracy: 0.4151\n",
            "Epoch 1912/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9578 - accuracy: 0.8945 - val_loss: 6.4137 - val_accuracy: 0.4151\n",
            "Epoch 1913/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8901 - accuracy: 0.9352 - val_loss: 6.4193 - val_accuracy: 0.4151\n",
            "Epoch 1914/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9062 - accuracy: 0.9223 - val_loss: 6.4347 - val_accuracy: 0.3774\n",
            "Epoch 1915/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9031 - accuracy: 0.9212 - val_loss: 6.4259 - val_accuracy: 0.4151\n",
            "Epoch 1916/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9061 - accuracy: 0.9215 - val_loss: 6.4359 - val_accuracy: 0.3774\n",
            "Epoch 1917/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9512 - accuracy: 0.9071 - val_loss: 6.4447 - val_accuracy: 0.3774\n",
            "Epoch 1918/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9118 - accuracy: 0.9359 - val_loss: 6.4391 - val_accuracy: 0.3774\n",
            "Epoch 1919/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.8929 - accuracy: 0.9440 - val_loss: 6.4302 - val_accuracy: 0.4151\n",
            "Epoch 1920/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9107 - accuracy: 0.9376 - val_loss: 6.4460 - val_accuracy: 0.3774\n",
            "Epoch 1921/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9391 - accuracy: 0.9132 - val_loss: 6.4441 - val_accuracy: 0.3774\n",
            "Epoch 1922/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9676 - accuracy: 0.8912 - val_loss: 6.4595 - val_accuracy: 0.3774\n",
            "Epoch 1923/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8974 - accuracy: 0.9487 - val_loss: 6.4458 - val_accuracy: 0.3774\n",
            "Epoch 1924/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9668 - accuracy: 0.8736 - val_loss: 6.4369 - val_accuracy: 0.3962\n",
            "Epoch 1925/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9013 - accuracy: 0.9422 - val_loss: 6.4432 - val_accuracy: 0.3774\n",
            "Epoch 1926/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8872 - accuracy: 0.9391 - val_loss: 6.4793 - val_accuracy: 0.3585\n",
            "Epoch 1927/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8659 - accuracy: 0.9514 - val_loss: 6.4825 - val_accuracy: 0.3585\n",
            "Epoch 1928/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9142 - accuracy: 0.9231 - val_loss: 6.4908 - val_accuracy: 0.3585\n",
            "Epoch 1929/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9104 - accuracy: 0.9256 - val_loss: 6.4723 - val_accuracy: 0.3585\n",
            "Epoch 1930/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8803 - accuracy: 0.9502 - val_loss: 6.4749 - val_accuracy: 0.3585\n",
            "Epoch 1931/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8796 - accuracy: 0.9251 - val_loss: 6.4624 - val_accuracy: 0.3774\n",
            "Epoch 1932/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9186 - accuracy: 0.9291 - val_loss: 6.4460 - val_accuracy: 0.3962\n",
            "Epoch 1933/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9866 - accuracy: 0.9070 - val_loss: 6.4552 - val_accuracy: 0.3962\n",
            "Epoch 1934/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.8787 - accuracy: 0.9338 - val_loss: 6.4631 - val_accuracy: 0.3774\n",
            "Epoch 1935/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.8932 - accuracy: 0.9293 - val_loss: 6.4473 - val_accuracy: 0.3962\n",
            "Epoch 1936/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9282 - accuracy: 0.9054 - val_loss: 6.4363 - val_accuracy: 0.3962\n",
            "Epoch 1937/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8926 - accuracy: 0.9112 - val_loss: 6.4456 - val_accuracy: 0.3962\n",
            "Epoch 1938/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8963 - accuracy: 0.9362 - val_loss: 6.4312 - val_accuracy: 0.3962\n",
            "Epoch 1939/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9389 - accuracy: 0.8952 - val_loss: 6.4176 - val_accuracy: 0.3962\n",
            "Epoch 1940/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 4.9121 - accuracy: 0.9435 - val_loss: 6.4182 - val_accuracy: 0.3962\n",
            "Epoch 1941/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9308 - accuracy: 0.9246 - val_loss: 6.4315 - val_accuracy: 0.3962\n",
            "Epoch 1942/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9271 - accuracy: 0.9316 - val_loss: 6.4337 - val_accuracy: 0.3962\n",
            "Epoch 1943/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9742 - accuracy: 0.9230 - val_loss: 6.4248 - val_accuracy: 0.3962\n",
            "Epoch 1944/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9295 - accuracy: 0.8921 - val_loss: 6.4286 - val_accuracy: 0.3962\n",
            "Epoch 1945/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8488 - accuracy: 0.9415 - val_loss: 6.4347 - val_accuracy: 0.3962\n",
            "Epoch 1946/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8816 - accuracy: 0.9243 - val_loss: 6.4237 - val_accuracy: 0.3962\n",
            "Epoch 1947/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8418 - accuracy: 0.9500 - val_loss: 6.4102 - val_accuracy: 0.3962\n",
            "Epoch 1948/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9195 - accuracy: 0.9213 - val_loss: 6.4008 - val_accuracy: 0.4151\n",
            "Epoch 1949/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9026 - accuracy: 0.8982 - val_loss: 6.3893 - val_accuracy: 0.4151\n",
            "Epoch 1950/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9486 - accuracy: 0.9135 - val_loss: 6.4012 - val_accuracy: 0.3962\n",
            "Epoch 1951/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9235 - accuracy: 0.9417 - val_loss: 6.3897 - val_accuracy: 0.4151\n",
            "Epoch 1952/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9205 - accuracy: 0.9131 - val_loss: 6.3946 - val_accuracy: 0.4151\n",
            "Epoch 1953/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9387 - accuracy: 0.9297 - val_loss: 6.4019 - val_accuracy: 0.3962\n",
            "Epoch 1954/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8855 - accuracy: 0.9349 - val_loss: 6.4151 - val_accuracy: 0.3962\n",
            "Epoch 1955/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9308 - accuracy: 0.9105 - val_loss: 6.4054 - val_accuracy: 0.3962\n",
            "Epoch 1956/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8945 - accuracy: 0.9030 - val_loss: 6.3941 - val_accuracy: 0.3962\n",
            "Epoch 1957/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9288 - accuracy: 0.9017 - val_loss: 6.3868 - val_accuracy: 0.4151\n",
            "Epoch 1958/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8827 - accuracy: 0.9334 - val_loss: 6.3813 - val_accuracy: 0.4151\n",
            "Epoch 1959/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 4.9300 - accuracy: 0.9165 - val_loss: 6.3928 - val_accuracy: 0.3962\n",
            "Epoch 1960/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.8669 - accuracy: 0.9213 - val_loss: 6.3878 - val_accuracy: 0.3774\n",
            "Epoch 1961/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 4.8674 - accuracy: 0.9262 - val_loss: 6.3835 - val_accuracy: 0.3962\n",
            "Epoch 1962/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9169 - accuracy: 0.9236 - val_loss: 6.3922 - val_accuracy: 0.3774\n",
            "Epoch 1963/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8968 - accuracy: 0.9242 - val_loss: 6.4122 - val_accuracy: 0.3962\n",
            "Epoch 1964/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 4.9201 - accuracy: 0.9222 - val_loss: 6.3997 - val_accuracy: 0.3962\n",
            "Epoch 1965/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9005 - accuracy: 0.9418 - val_loss: 6.3874 - val_accuracy: 0.3962\n",
            "Epoch 1966/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8854 - accuracy: 0.9437 - val_loss: 6.3757 - val_accuracy: 0.3774\n",
            "Epoch 1967/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8973 - accuracy: 0.9037 - val_loss: 6.3708 - val_accuracy: 0.3774\n",
            "Epoch 1968/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8801 - accuracy: 0.9420 - val_loss: 6.3961 - val_accuracy: 0.3962\n",
            "Epoch 1969/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8923 - accuracy: 0.9175 - val_loss: 6.3871 - val_accuracy: 0.3962\n",
            "Epoch 1970/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.8548 - accuracy: 0.9200 - val_loss: 6.4107 - val_accuracy: 0.3774\n",
            "Epoch 1971/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.9336 - accuracy: 0.8959 - val_loss: 6.4046 - val_accuracy: 0.3774\n",
            "Epoch 1972/2000\n",
            "5/5 [==============================] - 0s 76ms/step - loss: 4.9020 - accuracy: 0.9476 - val_loss: 6.4091 - val_accuracy: 0.3774\n",
            "Epoch 1973/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8586 - accuracy: 0.9504 - val_loss: 6.3970 - val_accuracy: 0.3962\n",
            "Epoch 1974/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.8721 - accuracy: 0.9549 - val_loss: 6.3878 - val_accuracy: 0.3962\n",
            "Epoch 1975/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8616 - accuracy: 0.9283 - val_loss: 6.3812 - val_accuracy: 0.3962\n",
            "Epoch 1976/2000\n",
            "5/5 [==============================] - 0s 38ms/step - loss: 4.8639 - accuracy: 0.9491 - val_loss: 6.3723 - val_accuracy: 0.4151\n",
            "Epoch 1977/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.8851 - accuracy: 0.9262 - val_loss: 6.3761 - val_accuracy: 0.4151\n",
            "Epoch 1978/2000\n",
            "5/5 [==============================] - 0s 46ms/step - loss: 4.8587 - accuracy: 0.9394 - val_loss: 6.3675 - val_accuracy: 0.4151\n",
            "Epoch 1979/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9081 - accuracy: 0.9312 - val_loss: 6.3694 - val_accuracy: 0.4151\n",
            "Epoch 1980/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8742 - accuracy: 0.9344 - val_loss: 6.3723 - val_accuracy: 0.3962\n",
            "Epoch 1981/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9279 - accuracy: 0.9301 - val_loss: 6.3628 - val_accuracy: 0.3962\n",
            "Epoch 1982/2000\n",
            "5/5 [==============================] - 0s 43ms/step - loss: 4.9080 - accuracy: 0.9217 - val_loss: 6.3713 - val_accuracy: 0.3962\n",
            "Epoch 1983/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8644 - accuracy: 0.9556 - val_loss: 6.3605 - val_accuracy: 0.3962\n",
            "Epoch 1984/2000\n",
            "5/5 [==============================] - 0s 44ms/step - loss: 4.9301 - accuracy: 0.9236 - val_loss: 6.3563 - val_accuracy: 0.3962\n",
            "Epoch 1985/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8783 - accuracy: 0.9258 - val_loss: 6.3614 - val_accuracy: 0.3774\n",
            "Epoch 1986/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9110 - accuracy: 0.8995 - val_loss: 6.3541 - val_accuracy: 0.3962\n",
            "Epoch 1987/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8953 - accuracy: 0.9230 - val_loss: 6.3525 - val_accuracy: 0.3962\n",
            "Epoch 1988/2000\n",
            "5/5 [==============================] - 0s 48ms/step - loss: 4.8841 - accuracy: 0.9291 - val_loss: 6.3458 - val_accuracy: 0.3962\n",
            "Epoch 1989/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.8586 - accuracy: 0.9040 - val_loss: 6.3524 - val_accuracy: 0.3774\n",
            "Epoch 1990/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.8603 - accuracy: 0.9358 - val_loss: 6.3566 - val_accuracy: 0.3774\n",
            "Epoch 1991/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9324 - accuracy: 0.9217 - val_loss: 6.3621 - val_accuracy: 0.3774\n",
            "Epoch 1992/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.8918 - accuracy: 0.9074 - val_loss: 6.3618 - val_accuracy: 0.3774\n",
            "Epoch 1993/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.9371 - accuracy: 0.9167 - val_loss: 6.3661 - val_accuracy: 0.3774\n",
            "Epoch 1994/2000\n",
            "5/5 [==============================] - 0s 40ms/step - loss: 4.9056 - accuracy: 0.9186 - val_loss: 6.3719 - val_accuracy: 0.3774\n",
            "Epoch 1995/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.8511 - accuracy: 0.9526 - val_loss: 6.3683 - val_accuracy: 0.3774\n",
            "Epoch 1996/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.9358 - accuracy: 0.9178 - val_loss: 6.3550 - val_accuracy: 0.3774\n",
            "Epoch 1997/2000\n",
            "5/5 [==============================] - 0s 39ms/step - loss: 4.8561 - accuracy: 0.9409 - val_loss: 6.3465 - val_accuracy: 0.3962\n",
            "Epoch 1998/2000\n",
            "5/5 [==============================] - 0s 45ms/step - loss: 4.8859 - accuracy: 0.9316 - val_loss: 6.3375 - val_accuracy: 0.4151\n",
            "Epoch 1999/2000\n",
            "5/5 [==============================] - 0s 41ms/step - loss: 4.9636 - accuracy: 0.9137 - val_loss: 6.3540 - val_accuracy: 0.4151\n",
            "Epoch 2000/2000\n",
            "5/5 [==============================] - 0s 42ms/step - loss: 4.8550 - accuracy: 0.9578 - val_loss: 6.3784 - val_accuracy: 0.3962\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "j0DoMLlR8Cas"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}