TCN_Self_generated_Data_Training.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "TCN_Self_generated Data Training.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPm/Y68WZ6ETHqGYGSqdExr",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ashishpatel26/tcn-keras-Examples/blob/master/TCN_Self_generated_Data_Training.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mOoq7seKTk6Q",
        "colab_type": "text"
      },
      "source": [
        "### Load dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9bF_d2wmT4Rh",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 340
        },
        "outputId": "1380a5ab-69a5-40c6-81cb-a2f63c2b9bfa"
      },
      "source": [
        "!pip install keras-tcn"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting keras-tcn\n",
            "  Downloading https://files.pythonhosted.org/packages/a8/5b/31eed031c196dc192eddf346f053ec6a97aefa4b931164fd8665c92a9d7d/keras_tcn-3.1.1-py3-none-any.whl\n",
            "Collecting keras==2.3.1\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/ad/fd/6bfe87920d7f4fd475acd28500a42482b6b84479832bdc0fe9e589a60ceb/Keras-2.3.1-py2.py3-none-any.whl (377kB)\n",
            "\u001b[K     |████████████████████████████████| 378kB 3.8MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.18.1 in /usr/local/lib/python3.6/dist-packages (from keras-tcn) (1.18.5)\n",
            "Requirement already satisfied: six>=1.9.0 in /usr/local/lib/python3.6/dist-packages (from keras==2.3.1->keras-tcn) (1.15.0)\n",
            "Collecting keras-applications>=1.0.6\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/71/e3/19762fdfc62877ae9102edf6342d71b28fbfd9dea3d2f96a882ce099b03f/Keras_Applications-1.0.8-py3-none-any.whl (50kB)\n",
            "\u001b[K     |████████████████████████████████| 51kB 6.6MB/s \n",
            "\u001b[?25hRequirement already satisfied: pyyaml in /usr/local/lib/python3.6/dist-packages (from keras==2.3.1->keras-tcn) (3.13)\n",
            "Requirement already satisfied: scipy>=0.14 in /usr/local/lib/python3.6/dist-packages (from keras==2.3.1->keras-tcn) (1.4.1)\n",
            "Requirement already satisfied: h5py in /usr/local/lib/python3.6/dist-packages (from keras==2.3.1->keras-tcn) (2.10.0)\n",
            "Requirement already satisfied: keras-preprocessing>=1.0.5 in /usr/local/lib/python3.6/dist-packages (from keras==2.3.1->keras-tcn) (1.1.2)\n",
            "Installing collected packages: keras-applications, keras, keras-tcn\n",
            "  Found existing installation: Keras 2.4.3\n",
            "    Uninstalling Keras-2.4.3:\n",
            "      Successfully uninstalled Keras-2.4.3\n",
            "Successfully installed keras-2.3.1 keras-applications-1.0.8 keras-tcn-3.1.1\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P-g__wZ8TPtx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "from tcn import compiled_tcn"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5XHqBBlbT-aA",
        "colab_type": "text"
      },
      "source": [
        "### Dataset Generation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iYyT0LwJTgoH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        },
        "outputId": "7ac16e2f-0f4c-4d44-923d-cc8a9b38452d"
      },
      "source": [
        "batch_size=1024\n",
        "sequence_length=32\n",
        "# input image dimensions\n",
        "pos_indices = np.random.choice(batch_size, size=int(batch_size // 2), replace=False)\n",
        "\n",
        "x_train = np.zeros(shape=(batch_size, sequence_length))\n",
        "print(f\"X_train shape: {x_train.shape}\")\n",
        "y_train = np.zeros(shape=(batch_size, 1))\n",
        "print(f\"y_train shape:{y_train.shape}\")\n",
        "x_train[pos_indices, 0] = 1.0\n",
        "y_train[pos_indices, 0] = 1.0\n",
        "\n",
        "# y_train = to_categorical(y_train, num_classes=2)\n",
        "\n",
        "print(\"After expanding the dimension shape is like this...!!!\")\n",
        "x_train = np.expand_dims(x_train, axis=2)\n",
        "y_train = y_train.copy()\n",
        "print(f\"X_train shape: {x_train.shape}\")\n",
        "print(f\"y_train shape:{y_train.shape}\")"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "X_train shape: (1024, 32)\n",
            "y_train shape:(1024, 1)\n",
            "After expanding the dimension shape is like this...!!!\n",
            "X_train shape: (1024, 32, 1)\n",
            "y_train shape:(1024, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nDTzraP1U4SS",
        "colab_type": "text"
      },
      "source": [
        "### Model Design"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Fe7b6MhpTu43",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 340
        },
        "outputId": "c3ee9850-dfef-4a69-891c-0f24830127ce"
      },
      "source": [
        "model = compiled_tcn(return_sequences=False,\n",
        "                         num_feat=1,\n",
        "                         num_classes=10,\n",
        "                         nb_filters=10,\n",
        "                         kernel_size=10,\n",
        "                         dilations=[1, 2, 4, 8, 16, 32],\n",
        "                         nb_stacks=6,\n",
        "                         max_len=x_train[0:1].shape[1],\n",
        "                         use_skip_connections=False)\n",
        "model.summary()"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "x.shape= (None, 10)\n",
            "model.x = (None, 32, 1)\n",
            "model.y = (None, 10)\n",
            "Model: \"functional_1\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "input_1 (InputLayer)         [(None, 32, 1)]           0         \n",
            "_________________________________________________________________\n",
            "tcn (TCN)                    (None, 10)                71840     \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 10)                110       \n",
            "_________________________________________________________________\n",
            "activation (Activation)      (None, 10)                0         \n",
            "=================================================================\n",
            "Total params: 71,950\n",
            "Trainable params: 71,950\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gigAUKI0VSUO",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 357
        },
        "outputId": "6a968760-3d2d-4871-d40e-ea967d84cdbb"
      },
      "source": [
        "history = model.fit(x_train, y_train, epochs=10)"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "32/32 [==============================] - 4s 116ms/step - loss: 4.7999 - accuracy: 0.8525\n",
            "Epoch 2/10\n",
            "32/32 [==============================] - 4s 121ms/step - loss: 0.0022 - accuracy: 0.9990\n",
            "Epoch 3/10\n",
            "32/32 [==============================] - 4s 119ms/step - loss: 6.4029e-05 - accuracy: 1.0000\n",
            "Epoch 4/10\n",
            "32/32 [==============================] - 4s 120ms/step - loss: 6.0811e-05 - accuracy: 1.0000\n",
            "Epoch 5/10\n",
            "32/32 [==============================] - 4s 118ms/step - loss: 1.1593e-04 - accuracy: 1.0000\n",
            "Epoch 6/10\n",
            "32/32 [==============================] - 4s 119ms/step - loss: 1.0923e-05 - accuracy: 1.0000\n",
            "Epoch 7/10\n",
            "32/32 [==============================] - 4s 122ms/step - loss: 7.6293e-06 - accuracy: 1.0000\n",
            "Epoch 8/10\n",
            "32/32 [==============================] - 4s 125ms/step - loss: 2.6019e-04 - accuracy: 1.0000\n",
            "Epoch 9/10\n",
            "32/32 [==============================] - 4s 120ms/step - loss: 4.6707e-05 - accuracy: 1.0000\n",
            "Epoch 10/10\n",
            "32/32 [==============================] - 4s 123ms/step - loss: 1.2664e-05 - accuracy: 1.0000\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_Hsxwp42VZug",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "7bf33fae-0ef5-48a8-8d4c-82a8eefd6a63"
      },
      "source": [
        "result = model.evaluate(x_train, y_train)\n",
        "print(f\"Accuracy : {result[1] * 100:.2f} %\")\n",
        "print(f\"Loss : {result[0] * 100:.2f} %\")"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "32/32 [==============================] - 1s 41ms/step - loss: 1.1921e-07 - accuracy: 1.0000\n",
            "Accuracy : 100.00 %\n",
            "Loss : 0.00 %\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "x85yW-baVwhC",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 422
        },
        "outputId": "2b5dbc16-6d56-47dc-8f6e-b9e6fec67cbe"
      },
      "source": [
        "from tensorflow import keras\n",
        "keras.utils.plot_model(model, show_shapes=True)"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SvCoSymaV0mR",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "e628a1e0-b027-41e7-d221-28003242b8c7"
      },
      "source": [
        "from matplotlib import pyplot as plt\n",
        "%matplotlib inline\n",
        "plt.plot(history.history['accuracy'])\n",
        "plt.plot(history.history['loss'])\n",
        "plt.title('Model Performance')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['Accuracy', 'Loss'], loc='upper left')\n",
        "lt.show()"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OoSCWvkdWNAi",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.metrics import precision_score\n",
        "from sklearn.metrics import recall_score\n",
        "from sklearn.metrics import f1_score\n",
        "from sklearn.metrics import cohen_kappa_score\n",
        "from sklearn.metrics import roc_auc_score\n",
        "from sklearn.metrics import confusion_matrix, classification_report\n",
        "import seaborn as sns\n"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-re6OXH3Wz3o",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 469
        },
        "outputId": "5ed1f3ea-15ee-4ebc-acb9-833fb2cfa2ef"
      },
      "source": [
        "# predict probabilities for test set\n",
        "yhat_probs = np.argmax(model.predict(x_train, verbose=0), axis=1)\n",
        " \n",
        "# kappa\n",
        "kappa = cohen_kappa_score(y_train, yhat_probs)\n",
        "print('Cohens kappa: %f' % kappa)\n",
        "# ROC AUC\n",
        "auc = roc_auc_score(y_train, yhat_probs)\n",
        "print('ROC AUC: %f' % auc)\n",
        "\n",
        "print(classification_report(y_train, yhat_probs))\n",
        "# confusion matrix\n",
        "matrix = confusion_matrix(y_train, yhat_probs)\n",
        "sns.heatmap(matrix, annot=True)"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Cohens kappa: 1.000000\n",
            "ROC AUC: 1.000000\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "         0.0       1.00      1.00      1.00       512\n",
            "         1.0       1.00      1.00      1.00       512\n",
            "\n",
            "    accuracy                           1.00      1024\n",
            "   macro avg       1.00      1.00      1.00      1024\n",
            "weighted avg       1.00      1.00      1.00      1024\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7efc771fe278>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAD4CAYAAADSIzzWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAVqUlEQVR4nO3de5RdVX3A8e9v8mgVSCaEvJOS2CAVlwpIMW2hC0XDSwkUG8EqWZAyVVJIqLXlpRYQRZGgVIFmlUBQA0SRBmLEYEIKLJ4RqUiiMkQwM3lBCM9IkpnZ/SMHvMBk5k5mMjv38P1k7TXn7PPad5H1y4/f2ftOpJSQJPW+utwDkKS3KgOwJGViAJakTAzAkpSJAViSMum7sx+w9ZmVTrPQm7xt5KG5h6BdUMuW5ujuPboSc/rt9Y5uP687zIAlKRMDsKRyaWutvnUiIp6MiEcj4pGIWFb07RkRd0TE48XPQUV/RMQVEdEYEb+MiAM7u78BWFK5tLZU36rzwZTS/imlg4r9s4HFKaV9gMXFPsBRwD5FawCu6uzGBmBJpZJSW9VtB00C5hTbc4DjKvqvT9vcD9RHxIiObmQAllQubW1Vt4hoiIhlFa3hDXdLwKKI+HnFsWEppTXF9lpgWLE9ClhVcW1T0bddO30WhCT1qi5ktimlWcCsDk45JKXUHBFDgTsi4tdvuD5FxA7P9DIASyqXKl6uVSul1Fz8XB8RtwAHA+siYkRKaU1RYlhfnN4MjKm4fHTRt12WICSVS2qrvnUgInaLiD1e3QYmAr8CbgWmFKdNAeYX27cCJxezISYAz1eUKtplBiypVFL1sxs6Mwy4JSJgW6ycm1K6PSIeAuZFxFTgKWBycf5C4GigEdgEnNLZAwzAksqlbYdnN7xOSmkl8L52+jcAh7fTn4BpXXmGAVhSuez49LJeZwCWVC49+BJuZzMASyoXM2BJyqTnXsLtdAZgSeXSQy/heoMBWFKppGQNWJLysAYsSZlYgpCkTMyAJSmT1q25R1A1A7CkcrEEIUmZWIKQpEzMgCUpEwOwJOWRfAknSZlYA5akTCxBSFImZsCSlIkZsCRlYgYsSZm0+IXskpSHGbAkZWINWJIyMQOWpEzMgCUpEzNgScrEWRCSlElKuUdQNQOwpHKxBixJmRiAJSkTX8JJUiatrblHUDUDsKRysQQhSZnUUACuyz0ASepRqa36VoWI6BMRv4iIBcX+uIh4ICIaI+KmiOhf9P9Jsd9YHB/b2b0NwJJKJbWlqluVpgMrKva/BlyeUhoPbASmFv1TgY1F/+XFeR0yAEsql7a26lsnImI0cAzw38V+AB8CflicMgc4rtieVOxTHD+8OH+7DMCSyqW1teoWEQ0RsayiNbzhbt8E/g14NVoPBp5LKb263rkJGFVsjwJWARTHny/O3y5fwkkqly68hEspzQJmtXcsIj4KrE8p/TwiDuuZwb2eAVhSufTcLIi/AY6NiKOBPwUGAN8C6iOib5Hljgaai/ObgTFAU0T0BQYCGzp6gCWIDkw8YQrHf/qznDBlGpNPPfNNx1c+tYp/aDiLAw77GNfO/WE7d+i6LVu28LkvfJWjJp/KSafNoHnNOgDuffBhJp96Bsd/+rNMPvUMHvj5Iz3yPOV1xMTDeOxXd/Hr5ffwb5+flns45ZBS9a3D26RzUkqjU0pjgROBJSmlfwDuBD5enDYFmF9s31rsUxxfklLHDzED7sTs/7yEQfUD2z02cMAenH3WZ1hy131dvm/zmnWcd/FlXPftr7+u/0cLFjFgj935ybzZLPzZUmZeOZvLLjqHQfUD+PbX/oOhQwbz+Mon+aezzmfJ/O/t0GfSrqGuro4rvnUxRx59Ek1Na7j/voXctmARK1Y8nntotW3nzwP+d+DGiPgy8AvgmqL/GuC7EdEIPMu2oN2hTgNwRPwF297uvVpobgZuTSmt2P5Vbw2DB9UzeFA9d9370JuO3fbTJXz/B/PZurWF9757X87/3DT69OnT6T2X3H0fp0/9FAATDzuUr8y8ipQS73rn+NfOGT9ub17ZvJktW7bQv3//nvtA6lUH/+UBPPHEk/zud78HYN68+Rz7sSMMwN1V/fSyqqWUlgJLi+2VwMHtnPMK8PdduW+HJYiI+HfgRiCAB4sWwA0RcXZXHlSLIoKGs85j8qln8IP5C6u+7oknf8/ti/+X7159GTfP+Q51dXUsWHRnVdeuf3oDw4fuBUDfvn3Yfbe389zzL7zunDuW3sN++443+Na4kaOGs6pp9Wv7Tc1rGDlyeMYRlUQXZkHk1lkGPBV4d0ppa2VnRMwEHgMuae+iYipHA8CVl32Zfzz5pB4Yau+7/qpvMGzIXmzY+BynzTiXcXuP4aD939PpdQ8se4Tlv27kxKnTAdi8eTN7DqoH4MxzLqR59Tq2tmxlzbqnOWHKtrrfpyZP4vhjJnZ678aVTzHzytnMuvzibnwyqbxSDS1F7iwAtwEjgafe0D+CP86Le5PKqR1bn1lZO19P/wbDhmzLRAcPqufwv/1rHl3+m6oCcEqJY4/6MGd99pQ3Hbviq18Etl8DHjpkMGvXP8PwoUNoaWnlpZc3UT9wAABr1z/N9HMv4itf+Ff+bPTI7n48Zba6eS1jKv47jh41gtWr12YcUUnshBLEztLZLIgZwOKI+ElEzCra7cBiti3PK61Nf3iFl1/e9Nr2vQ8+zD7vGFvVtRMO2p87lt7Dho3PAfD8Cy+yeu26qq794CETmL/wZwAsWno3H3j/+4gIXnjxJU7//JeY8ZlTOPC97+76B9Iu56FljzB+/DjGjh1Dv379mDx5ErctWJR7WLWvh78LYmfqMANOKd0eEe9kW8G58iXcQyml/AWUnWjDsxuZfu5FALS2tHL0xMM4ZMJB3HTLjwH4xPHH8MyGZ/nE1DN56eVN1NXV8b15/8P87/8Xfz5ub8447WQaZpxHW2qjX9++nPcvpzNy+LBOn/t3Hz2Ccy66lKMmn8rAAXtw6QXbSu033Hwbq5pWc/W1c7n62rkAzPrmxQwuShuqPa2trUyfcT4LfzyXPnV1XDfnJpYv/23uYdW+GsqAo5Npat1WyyUI7TxvG3lo7iFoF9SypbnD706oxstfPLHqmLPbhTd2+3nd4TxgSeWyC5QWqmUAllQuNVSCMABLKpUyTUOTpNpiBixJmRiAJSmTXWCJcbUMwJJKpQu/6y07A7CkcjEAS1ImzoKQpEzMgCUpEwOwJOWRWi1BSFIeZsCSlIfT0CQpFwOwJGVSOyVgA7CkckkttROBDcCSyqV24q8BWFK5+BJOknIxA5akPMyAJSkXM2BJyiO15B5B9QzAkkqlhn4rvQFYUskYgCUpDzNgScqklgJwXe4BSFJPSq1RdetIRPxpRDwYEf8XEY9FxAVF/7iIeCAiGiPipojoX/T/SbHfWBwf29lYDcCSSiW1Vd86sRn4UErpfcD+wJERMQH4GnB5Smk8sBGYWpw/FdhY9F9enNchA7CkUkltUXXr8D7bvFTs9itaAj4E/LDonwMcV2xPKvYpjh8eER0+xAAsqVS6kgFHRENELKtoDZX3iog+EfEIsB64A3gCeC6l12YbNwGjiu1RwCqA4vjzwOCOxupLOEmlklLHme3rz02zgFkdHG8F9o+IeuAW4C+6PcAKZsCSSqUHa8B/vGdKzwF3An8F1EfEq8nraKC52G4GxgAUxwcCGzq6rwFYUqm0tUbVrSMRMaTIfImItwEfAVawLRB/vDhtCjC/2L612Kc4viSl1OE3A1mCkFQqnb1c64IRwJyI6MO2ZHVeSmlBRCwHboyILwO/AK4pzr8G+G5ENALPAid29gADsKRS6akAnFL6JXBAO/0rgYPb6X8F+PuuPMMALKlUOv6f/l2LAVhSqfRgCWKnMwBLKpWuTEPLzQAsqVRaO5ndsCsxAEsqFTNgScrEGrAkZeIsCEnKxAxYkjJpbaudb1gwAEsqFUsQkpRJm7MgJCkPp6FJUiaWICq8beShO/sRqkF/WH137iGopCxBSFImzoKQpExqqAJhAJZULpYgJCkTZ0FIUiZd+GXH2RmAJZVKwgxYkrJosQQhSXmYAUtSJtaAJSkTM2BJysQMWJIyaTUDlqQ8aug3EhmAJZVLmxmwJOXhl/FIUia+hJOkTNrCEoQkZdGaewBdYACWVCrOgpCkTGppFkTt/PIkSapC6kLrSESMiYg7I2J5RDwWEdOL/j0j4o6IeLz4Oajoj4i4IiIaI+KXEXFgZ2M1AEsqlbaovnWiBfhcSmk/YAIwLSL2A84GFqeU9gEWF/sARwH7FK0BuKqzBxiAJZVKWxdaR1JKa1JKDxfbLwIrgFHAJGBOcdoc4LhiexJwfdrmfqA+IkZ09AwDsKRSaY3qW0Q0RMSyitbQ3j0jYixwAPAAMCyltKY4tBYYVmyPAlZVXNZU9G2XL+EklUpXFmKklGYBszo6JyJ2B24GZqSUXoiKecYppRQRO7z4zgxYUqn0VAkCICL6sS34fj+l9KOie92rpYXi5/qivxkYU3H56KJvuwzAkkolRfWtI7Et1b0GWJFSmllx6FZgSrE9BZhf0X9yMRtiAvB8RamiXZYgJJVKD34XxN8AnwYejYhHir5zgUuAeRExFXgKmFwcWwgcDTQCm4BTOnuAAVhSqfTUUuSU0j2w3VUdh7dzfgKmdeUZBmBJpeJSZEnKxK+jlKRMDMCSlIm/EUOSMrEGLEmZ+IXskpRJWw0VIQzAkkrFl3CSlEnt5L8GYEklYwYsSZm07Pi3Q/Y6A7CkUqmd8GsAllQyliAkKROnoUlSJrUTfg3AkkrGEoQkZdJaQzmwAVhSqZgBS1ImyQxYkvIwA9abHDHxMGbOvJA+dXXMvvYGvn7pd3IPSTto4glT2O3tb6euro4+ffowb/YVrzu+8qlVfOHimSz/bSNnNkzhlE9+vNvP3LJlC+dcdBnLf/M49QMH8I0Lz2HUiGHc++DDfPPqa9m6tYV+/fryuWlT+cD79+/282qZ09D0OnV1dVzxrYs58uiTaGpaw/33LeS2BYtYseLx3EPTDpr9n5cwqH5gu8cGDtiDs8/6DEvuuq/L921es47zLr6M67799df1/2jBIgbssTs/mTebhT9byswrZ3PZRecwqH4A3/7afzB0yGAeX/kk/3TW+SyZ/70d+kxlUTvhF+pyD+Ct4OC/PIAnnniS3/3u92zdupV58+Zz7MeOyD0s7SSDB9XznnftS9++b85vbvvpEk78x+mcMGUaF3z9Clpbq/v68CV338ekoz8MwMTDDuWBnz9CSol3vXM8Q4cMBmD8uL15ZfNmtmzZ0nMfpga1kKpuuRmAe8HIUcNZ1bT6tf2m5jWMHDk844jUHRFBw1nnMfnUM/jB/IVVX/fEk7/n9sX/y3evvoyb53yHuro6Fiy6s6pr1z+9geFD9wKgb98+7L7b23nu+Rded84dS+9hv33H079//+o/TAmlLvzJbYdLEBFxSkrp2u0cawAaAKLPQOrqdtvRx0i7nOuv+gbDhuzFho3PcdqMcxm39xgO2v89nV73wLJHWP7rRk6cOh2AzZs3s+egegDOPOdCmlevY2vLVtase5oTpkwD4FOTJ3H8MRM7vXfjyqeYeeVsZl1+cTc+WTm8VV7CXQC0G4BTSrOAWQB9+4/K/89MZqub1zJm9MjX9kePGsHq1WszjkjdMWzItkx08KB6Dv/bv+bR5b+pKgCnlDj2qA9z1mdPedOxK776RWD7NeChQwazdv0zDB86hJaWVl56eRP1AwcAsHb900w/9yK+8oV/5c8q/p69Ve0KmW21OixBRMQvt9MeBYb10hhr3kPLHmH8+HGMHTuGfv36MXnyJG5bsCj3sLQDNv3hFV5+edNr2/c++DD7vGNsVddOOGh/7lh6Dxs2PgfA8y+8yOq166q69oOHTGD+wp8BsGjp3Xzg/e8jInjhxZc4/fNfYsZnTuHA97676x+ohNq60HLrLAMeBhwBbHxDfwD37pQRlVBrayvTZ5zPwh/PpU9dHdfNuYnly3+be1jaARue3cj0cy8CoLWllaMnHsYhEw7iplt+DMAnjj+GZzY8yyemnslLL2+irq6O7837H+Z//7/483F7c8ZpJ9Mw4zzaUhv9+vblvH85nZHDO89l/u6jR3DORZdy1ORTGThgDy694GwAbrj5NlY1rebqa+dy9bVzAZj1zYsZXJQ23opaU+1kwJE6GGxEXANcm1K6p51jc1NKn+zsAZYg1J4/rL479xC0C+q31zuiu/f45N7HVx1z5j51S7ef1x0dZsAppakdHOs0+EpSb6ulGrALMSSVyq5Q262WAVhSqbgUWZIyqaUShCvhJJVKa0pVt85ExOyIWB8Rv6ro2zMi7oiIx4ufg4r+iIgrIqKxmK57YGf3NwBLKpU2UtWtCtcBR76h72xgcUppH2BxsQ9wFLBP0RqAqzq7uQFYUqn05EKMlNJdwLNv6J4EzCm25wDHVfRfn7a5H6iPiBEd3d8ALKlUuvJlPBHREBHLKlpDFY8YllJaU2yv5Y+rgkcBqyrOayr6tsuXcJJKpSuzICq/t2ZHpJRSROzwWz8DsKRS6Wh1bw9ZFxEjUkprihLD+qK/GRhTcd7oom+7LEFIKpVWUtVtB90KTCm2pwDzK/pPLmZDTACeryhVtMsMWFKp9ORCjIi4ATgM2CsimoAvAZcA8yJiKvAUMLk4fSFwNNAIbALe/L2jb2AAllQqPVmCSCmdtJ1Dh7dzbgKmdeX+BmBJpeJSZEnKpJaWIhuAJZVKLX0huwFYUqlYgpCkTAzAkpRJLyzE6DEGYEmlYgYsSZk4C0KSMmlNtfNb4QzAkkrFGrAkZWINWJIysQYsSZm0WYKQpDzMgCUpE2dBSFImliAkKRNLEJKUiRmwJGViBixJmbSm1txDqJoBWFKpuBRZkjJxKbIkZWIGLEmZOAtCkjJxFoQkZeJSZEnKxBqwJGViDViSMjEDlqRMnAcsSZmYAUtSJs6CkKRMfAknSZnUUgmiLvcAJKknpS786UxEHBkRv4mIxog4u6fHagYsqVR6KgOOiD7Ad4CPAE3AQxFxa0ppeY88AAOwpJLpwRrwwUBjSmklQETcCEwCaicAt2xpjp39jFoREQ0ppVm5x6Fdi38velZXYk5ENAANFV2zKv5bjAJWVRxrAj7Q/RH+kTXg3tXQ+Sl6C/LvRSYppVkppYMqWq/+Q2gAlqT2NQNjKvZHF309xgAsSe17CNgnIsZFRH/gRODWnnyAL+F6l3U+tce/F7uglFJLRPwz8FOgDzA7pfRYTz4jamnSsiSViSUIScrEACxJmRiAe8nOXtKo2hMRsyNifUT8KvdYlIcBuBdULGk8CtgPOCki9ss7Ku0CrgOOzD0I5WMA7h2vLWlMKW0BXl3SqLewlNJdwLO5x6F8DMC9o70ljaMyjUXSLsIALEmZGIB7x05f0iip9hiAe8dOX9IoqfYYgHtBSqkFeHVJ4wpgXk8vaVTtiYgbgPuAfSOiKSKm5h6TepdLkSUpEzNgScrEACxJmRiAJSkTA7AkZWIAlqRMDMCSlIkBWJIy+X+VPWaBg+8pdQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}