scrnn_highsnr.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "COXiFhxo_S87",
    "outputId": "d92bdc36-26da-4407-aeed-b75abdacce97"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import keras\n",
    "from keras.layers import Input,Dense,Flatten,MaxPool1D,Activation,LeakyReLU,LSTM,BatchNormalization,Dropout, Conv1D,Lambda\n",
    "from keras.models import Model,load_model\n",
    "import pandas as pd\n",
    "import keras.backend as K\n",
    "import h5py\n",
    "import model_architectures as ma\n",
    "import utils"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 122
    },
    "colab_type": "code",
    "id": "SwyW6KaOBhwq",
    "outputId": "d8ba4da2-1c69-4d9b-f20b-1daa2747f275"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Go to this URL in a browser: https://accounts.google.com/o/oauth2/auth?client_id=947318989803-6bn6qk8qdgf4n4g3pfee6491hc0brc4i.apps.googleusercontent.com&redirect_uri=urn%3aietf%3awg%3aoauth%3a2.0%3aoob&response_type=code&scope=email%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdocs.test%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive.photos.readonly%20https%3a%2f%2fwww.googleapis.com%2fauth%2fpeopleapi.readonly\n",
      "\n",
      "Enter your authorization code:\n",
      "··········\n",
      "Mounted at /content/gdrive\n"
     ]
    }
   ],
   "source": [
    "from google.colab import drive\n",
    "drive.mount('/content/gdrive') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "4z9PhjQA_S8-"
   },
   "outputs": [],
   "source": [
    "models_dir = '/content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/'\n",
    "feat_file='/content/gdrive/My Drive/Project-EE5611/project_features/base_scrnn_highsnr_features.h5'\n",
    "fid=h5py.File(feat_file,'r');\n",
    "x_tr=fid['x_tr']\n",
    "y_tr=fid['y_tr'] \n",
    "snr_tr = np.array(fid['snr_tr'])\n",
    "x_ts=fid['x_ts']\n",
    "y_ts=fid['y_ts']\n",
    "snr_ts = np.array(fid['snr_ts'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 493
    },
    "colab_type": "code",
    "id": "Lrj_O39EKzrn",
    "outputId": "d89c254b-dca1-4292-bf54-8bb881cfbfa0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         (None, 128, 2)            0         \n",
      "_________________________________________________________________\n",
      "batch_normalization_1 (Batch (None, 128, 2)            8         \n",
      "_________________________________________________________________\n",
      "conv1d_1 (Conv1D)            (None, 124, 128)          1408      \n",
      "_________________________________________________________________\n",
      "max_pooling1d_1 (MaxPooling1 (None, 41, 128)           0         \n",
      "_________________________________________________________________\n",
      "conv1d_2 (Conv1D)            (None, 37, 128)           82048     \n",
      "_________________________________________________________________\n",
      "lstm_1 (LSTM)                (None, 37, 128)           131584    \n",
      "_________________________________________________________________\n",
      "lstm_2 (LSTM)                (None, 37, 128)           131584    \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 37, 128)           0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 4736)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 11)                52107     \n",
      "=================================================================\n",
      "Total params: 398,739\n",
      "Trainable params: 398,735\n",
      "Non-trainable params: 4\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "m1 = ma.base_scrnn()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "QNDUSZVLLvmA"
   },
   "outputs": [],
   "source": [
    "checkpoint = keras.callbacks.ModelCheckpoint(models_dir+\"model_epoch-{epoch:02d}_val_acc-{val_accuracy:.2f}.h5\", monitor='val_accuracy', verbose=2, save_best_only=False, save_weights_only=False, mode='auto', period=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "YQdYici6MzxW",
    "outputId": "d5d0c311-7136-453b-c89c-5bd2cbf10ba2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 99000 samples, validate on 22000 samples\n",
      "Epoch 1/75\n",
      "99000/99000 [==============================] - 42s 420us/step - loss: 0.9561 - accuracy: 0.5780 - val_loss: 2.1554 - val_accuracy: 0.4149\n",
      "\n",
      "Epoch 00001: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-01_val_acc-0.41.h5\n",
      "Epoch 2/75\n",
      "99000/99000 [==============================] - 27s 269us/step - loss: 0.4861 - accuracy: 0.7644 - val_loss: 2.3689 - val_accuracy: 0.4915\n",
      "\n",
      "Epoch 00002: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-02_val_acc-0.49.h5\n",
      "Epoch 3/75\n",
      "99000/99000 [==============================] - 28s 284us/step - loss: 0.4481 - accuracy: 0.7775 - val_loss: 2.4997 - val_accuracy: 0.4881\n",
      "\n",
      "Epoch 00003: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-03_val_acc-0.49.h5\n",
      "Epoch 4/75\n",
      "99000/99000 [==============================] - 28s 283us/step - loss: 0.3651 - accuracy: 0.8106 - val_loss: 2.7450 - val_accuracy: 0.4964\n",
      "\n",
      "Epoch 00004: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-04_val_acc-0.50.h5\n",
      "Epoch 5/75\n",
      "99000/99000 [==============================] - 28s 280us/step - loss: 0.3441 - accuracy: 0.8148 - val_loss: 2.5559 - val_accuracy: 0.5125\n",
      "\n",
      "Epoch 00005: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-05_val_acc-0.51.h5\n",
      "Epoch 6/75\n",
      "99000/99000 [==============================] - 27s 276us/step - loss: 0.3288 - accuracy: 0.8229 - val_loss: 2.7041 - val_accuracy: 0.5145\n",
      "\n",
      "Epoch 00006: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-06_val_acc-0.51.h5\n",
      "Epoch 7/75\n",
      "99000/99000 [==============================] - 28s 285us/step - loss: 0.3100 - accuracy: 0.8337 - val_loss: 2.5065 - val_accuracy: 0.5017\n",
      "\n",
      "Epoch 00007: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-07_val_acc-0.50.h5\n",
      "Epoch 8/75\n",
      "99000/99000 [==============================] - 27s 276us/step - loss: 0.2999 - accuracy: 0.8396 - val_loss: 2.4941 - val_accuracy: 0.5186\n",
      "\n",
      "Epoch 00008: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-08_val_acc-0.52.h5\n",
      "Epoch 9/75\n",
      "99000/99000 [==============================] - 28s 280us/step - loss: 0.2928 - accuracy: 0.8433 - val_loss: 2.5070 - val_accuracy: 0.5214\n",
      "\n",
      "Epoch 00009: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-09_val_acc-0.52.h5\n",
      "Epoch 10/75\n",
      "99000/99000 [==============================] - 27s 277us/step - loss: 0.2891 - accuracy: 0.8460 - val_loss: 2.4139 - val_accuracy: 0.5274\n",
      "\n",
      "Epoch 00010: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-10_val_acc-0.53.h5\n",
      "Epoch 11/75\n",
      "99000/99000 [==============================] - 27s 275us/step - loss: 0.2834 - accuracy: 0.8502 - val_loss: 2.2862 - val_accuracy: 0.5265\n",
      "\n",
      "Epoch 00011: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-11_val_acc-0.53.h5\n",
      "Epoch 12/75\n",
      "99000/99000 [==============================] - 28s 279us/step - loss: 0.2824 - accuracy: 0.8543 - val_loss: 2.4009 - val_accuracy: 0.5289\n",
      "\n",
      "Epoch 00012: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-12_val_acc-0.53.h5\n",
      "Epoch 13/75\n",
      "99000/99000 [==============================] - 28s 286us/step - loss: 0.2678 - accuracy: 0.8635 - val_loss: 2.4595 - val_accuracy: 0.5338\n",
      "\n",
      "Epoch 00013: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-13_val_acc-0.53.h5\n",
      "Epoch 14/75\n",
      "99000/99000 [==============================] - 27s 277us/step - loss: 0.2679 - accuracy: 0.8706 - val_loss: 2.5867 - val_accuracy: 0.5339\n",
      "\n",
      "Epoch 00014: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-14_val_acc-0.53.h5\n",
      "Epoch 15/75\n",
      "99000/99000 [==============================] - 28s 282us/step - loss: 0.2493 - accuracy: 0.8806 - val_loss: 2.6236 - val_accuracy: 0.5370\n",
      "\n",
      "Epoch 00015: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-15_val_acc-0.54.h5\n",
      "Epoch 16/75\n",
      "99000/99000 [==============================] - 28s 281us/step - loss: 0.2364 - accuracy: 0.8880 - val_loss: 2.4489 - val_accuracy: 0.5394\n",
      "\n",
      "Epoch 00016: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-16_val_acc-0.54.h5\n",
      "Epoch 17/75\n",
      "99000/99000 [==============================] - 28s 282us/step - loss: 0.2235 - accuracy: 0.8952 - val_loss: 2.6424 - val_accuracy: 0.5488\n",
      "\n",
      "Epoch 00017: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-17_val_acc-0.55.h5\n",
      "Epoch 18/75\n",
      "99000/99000 [==============================] - 27s 277us/step - loss: 0.2177 - accuracy: 0.8991 - val_loss: 2.6057 - val_accuracy: 0.5469\n",
      "\n",
      "Epoch 00018: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-18_val_acc-0.55.h5\n",
      "Epoch 19/75\n",
      "99000/99000 [==============================] - 28s 284us/step - loss: 0.2138 - accuracy: 0.9025 - val_loss: 2.3927 - val_accuracy: 0.5536\n",
      "\n",
      "Epoch 00019: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-19_val_acc-0.55.h5\n",
      "Epoch 20/75\n",
      "99000/99000 [==============================] - 28s 287us/step - loss: 0.2039 - accuracy: 0.9064 - val_loss: 2.6436 - val_accuracy: 0.5512\n",
      "\n",
      "Epoch 00020: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-20_val_acc-0.55.h5\n",
      "Epoch 21/75\n",
      "99000/99000 [==============================] - 29s 289us/step - loss: 0.1932 - accuracy: 0.9110 - val_loss: 2.4578 - val_accuracy: 0.5514\n",
      "\n",
      "Epoch 00021: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-21_val_acc-0.55.h5\n",
      "Epoch 22/75\n",
      "99000/99000 [==============================] - 27s 275us/step - loss: 0.1897 - accuracy: 0.9136 - val_loss: 2.7972 - val_accuracy: 0.5525\n",
      "\n",
      "Epoch 00022: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-22_val_acc-0.55.h5\n",
      "Epoch 23/75\n",
      "99000/99000 [==============================] - 27s 272us/step - loss: 0.1849 - accuracy: 0.9158 - val_loss: 2.7342 - val_accuracy: 0.5523\n",
      "\n",
      "Epoch 00023: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-23_val_acc-0.55.h5\n",
      "Epoch 24/75\n",
      "99000/99000 [==============================] - 28s 285us/step - loss: 0.1771 - accuracy: 0.9178 - val_loss: 2.6597 - val_accuracy: 0.5563\n",
      "\n",
      "Epoch 00024: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-24_val_acc-0.56.h5\n",
      "Epoch 25/75\n",
      "99000/99000 [==============================] - 28s 285us/step - loss: 0.1750 - accuracy: 0.9196 - val_loss: 2.4276 - val_accuracy: 0.5563\n",
      "\n",
      "Epoch 00025: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-25_val_acc-0.56.h5\n",
      "Epoch 26/75\n",
      "99000/99000 [==============================] - 28s 282us/step - loss: 0.1693 - accuracy: 0.9229 - val_loss: 2.7612 - val_accuracy: 0.5458\n",
      "\n",
      "Epoch 00026: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-26_val_acc-0.55.h5\n",
      "Epoch 27/75\n",
      "99000/99000 [==============================] - 28s 286us/step - loss: 0.1664 - accuracy: 0.9245 - val_loss: 2.7631 - val_accuracy: 0.5488\n",
      "\n",
      "Epoch 00027: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-27_val_acc-0.55.h5\n",
      "Epoch 28/75\n",
      "99000/99000 [==============================] - 27s 274us/step - loss: 0.1693 - accuracy: 0.9227 - val_loss: 2.8181 - val_accuracy: 0.5485\n",
      "\n",
      "Epoch 00028: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-28_val_acc-0.55.h5\n",
      "Epoch 29/75\n",
      "99000/99000 [==============================] - 28s 279us/step - loss: 0.1599 - accuracy: 0.9255 - val_loss: 2.7434 - val_accuracy: 0.5588\n",
      "\n",
      "Epoch 00029: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-29_val_acc-0.56.h5\n",
      "Epoch 30/75\n",
      "99000/99000 [==============================] - 28s 279us/step - loss: 0.1558 - accuracy: 0.9272 - val_loss: 2.8644 - val_accuracy: 0.5525\n",
      "\n",
      "Epoch 00030: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-30_val_acc-0.55.h5\n",
      "Epoch 31/75\n",
      "99000/99000 [==============================] - 28s 281us/step - loss: 0.1538 - accuracy: 0.9286 - val_loss: 2.7502 - val_accuracy: 0.5543\n",
      "\n",
      "Epoch 00031: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-31_val_acc-0.55.h5\n",
      "Epoch 32/75\n",
      "99000/99000 [==============================] - 28s 282us/step - loss: 0.1521 - accuracy: 0.9283 - val_loss: 2.9131 - val_accuracy: 0.5515\n",
      "\n",
      "Epoch 00032: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-32_val_acc-0.55.h5\n",
      "Epoch 33/75\n",
      "99000/99000 [==============================] - 27s 278us/step - loss: 0.1522 - accuracy: 0.9283 - val_loss: 2.9826 - val_accuracy: 0.5544\n",
      "\n",
      "Epoch 00033: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-33_val_acc-0.55.h5\n",
      "Epoch 34/75\n",
      "99000/99000 [==============================] - 28s 278us/step - loss: 0.1519 - accuracy: 0.9288 - val_loss: 3.3460 - val_accuracy: 0.5488\n",
      "\n",
      "Epoch 00034: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-34_val_acc-0.55.h5\n",
      "Epoch 35/75\n",
      "99000/99000 [==============================] - 28s 286us/step - loss: 0.1464 - accuracy: 0.9306 - val_loss: 2.7826 - val_accuracy: 0.5585\n",
      "\n",
      "Epoch 00035: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-35_val_acc-0.56.h5\n",
      "Epoch 36/75\n",
      "99000/99000 [==============================] - 28s 286us/step - loss: 0.2099 - accuracy: 0.9037 - val_loss: 2.4714 - val_accuracy: 0.5474\n",
      "\n",
      "Epoch 00036: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-36_val_acc-0.55.h5\n",
      "Epoch 37/75\n",
      "99000/99000 [==============================] - 27s 276us/step - loss: 0.1591 - accuracy: 0.9257 - val_loss: 2.3762 - val_accuracy: 0.5640\n",
      "\n",
      "Epoch 00037: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-37_val_acc-0.56.h5\n",
      "Epoch 38/75\n",
      "99000/99000 [==============================] - 27s 277us/step - loss: 0.2755 - accuracy: 0.8705 - val_loss: 2.5028 - val_accuracy: 0.5270\n",
      "\n",
      "Epoch 00038: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-38_val_acc-0.53.h5\n",
      "Epoch 39/75\n",
      "99000/99000 [==============================] - 28s 282us/step - loss: 0.2209 - accuracy: 0.8975 - val_loss: 2.6956 - val_accuracy: 0.5560\n",
      "\n",
      "Epoch 00039: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-39_val_acc-0.56.h5\n",
      "Epoch 40/75\n",
      "99000/99000 [==============================] - 27s 271us/step - loss: 0.1713 - accuracy: 0.9198 - val_loss: 3.0421 - val_accuracy: 0.5591\n",
      "\n",
      "Epoch 00040: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-40_val_acc-0.56.h5\n",
      "Epoch 41/75\n",
      "99000/99000 [==============================] - 28s 280us/step - loss: 0.1534 - accuracy: 0.9266 - val_loss: 2.5761 - val_accuracy: 0.5650\n",
      "\n",
      "Epoch 00041: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-41_val_acc-0.56.h5\n",
      "Epoch 42/75\n",
      "99000/99000 [==============================] - 28s 280us/step - loss: 0.1467 - accuracy: 0.9295 - val_loss: 3.1886 - val_accuracy: 0.5508\n",
      "\n",
      "Epoch 00042: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-42_val_acc-0.55.h5\n",
      "Epoch 43/75\n",
      "99000/99000 [==============================] - 28s 287us/step - loss: 0.1625 - accuracy: 0.9239 - val_loss: 2.8777 - val_accuracy: 0.5588\n",
      "\n",
      "Epoch 00043: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-43_val_acc-0.56.h5\n",
      "Epoch 44/75\n",
      "99000/99000 [==============================] - 28s 283us/step - loss: 0.1499 - accuracy: 0.9274 - val_loss: 3.3513 - val_accuracy: 0.5612\n",
      "\n",
      "Epoch 00044: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-44_val_acc-0.56.h5\n",
      "Epoch 45/75\n",
      "99000/99000 [==============================] - 27s 271us/step - loss: 0.1386 - accuracy: 0.9318 - val_loss: 3.0995 - val_accuracy: 0.5621\n",
      "\n",
      "Epoch 00045: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-45_val_acc-0.56.h5\n",
      "Epoch 46/75\n",
      "99000/99000 [==============================] - 28s 280us/step - loss: 0.1385 - accuracy: 0.9320 - val_loss: 3.1527 - val_accuracy: 0.5655\n",
      "\n",
      "Epoch 00046: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-46_val_acc-0.57.h5\n",
      "Epoch 47/75\n",
      "99000/99000 [==============================] - 28s 287us/step - loss: 0.1507 - accuracy: 0.9282 - val_loss: 3.6935 - val_accuracy: 0.5521\n",
      "\n",
      "Epoch 00047: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-47_val_acc-0.55.h5\n",
      "Epoch 48/75\n",
      "99000/99000 [==============================] - 28s 282us/step - loss: 0.1337 - accuracy: 0.9332 - val_loss: 3.3008 - val_accuracy: 0.5630\n",
      "\n",
      "Epoch 00048: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-48_val_acc-0.56.h5\n",
      "Epoch 49/75\n",
      "99000/99000 [==============================] - 27s 276us/step - loss: 0.2068 - accuracy: 0.9069 - val_loss: 2.7392 - val_accuracy: 0.5334\n",
      "\n",
      "Epoch 00049: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-49_val_acc-0.53.h5\n",
      "Epoch 50/75\n",
      "99000/99000 [==============================] - 27s 272us/step - loss: 0.2107 - accuracy: 0.9064 - val_loss: 2.4556 - val_accuracy: 0.5543\n",
      "\n",
      "Epoch 00050: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-50_val_acc-0.55.h5\n",
      "Epoch 51/75\n",
      "99000/99000 [==============================] - 28s 279us/step - loss: 0.1566 - accuracy: 0.9271 - val_loss: 2.6697 - val_accuracy: 0.5603\n",
      "\n",
      "Epoch 00051: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-51_val_acc-0.56.h5\n",
      "Epoch 52/75\n",
      "99000/99000 [==============================] - 27s 273us/step - loss: 0.1422 - accuracy: 0.9326 - val_loss: 2.8404 - val_accuracy: 0.5580\n",
      "\n",
      "Epoch 00052: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-52_val_acc-0.56.h5\n",
      "Epoch 53/75\n",
      "99000/99000 [==============================] - 27s 275us/step - loss: 0.1491 - accuracy: 0.9297 - val_loss: 2.7307 - val_accuracy: 0.5603\n",
      "\n",
      "Epoch 00053: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-53_val_acc-0.56.h5\n",
      "Epoch 54/75\n",
      "99000/99000 [==============================] - 27s 275us/step - loss: 0.1444 - accuracy: 0.9307 - val_loss: 3.3076 - val_accuracy: 0.5627\n",
      "\n",
      "Epoch 00054: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-54_val_acc-0.56.h5\n",
      "Epoch 55/75\n",
      "99000/99000 [==============================] - 28s 288us/step - loss: 0.1379 - accuracy: 0.9330 - val_loss: 3.0029 - val_accuracy: 0.5613\n",
      "\n",
      "Epoch 00055: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-55_val_acc-0.56.h5\n",
      "Epoch 56/75\n",
      "99000/99000 [==============================] - 28s 283us/step - loss: 0.1369 - accuracy: 0.9322 - val_loss: 3.8528 - val_accuracy: 0.5648\n",
      "\n",
      "Epoch 00056: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-56_val_acc-0.56.h5\n",
      "Epoch 57/75\n",
      "99000/99000 [==============================] - 27s 273us/step - loss: 0.1439 - accuracy: 0.9317 - val_loss: 2.7431 - val_accuracy: 0.5475\n",
      "\n",
      "Epoch 00057: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-57_val_acc-0.55.h5\n",
      "Epoch 58/75\n",
      "99000/99000 [==============================] - 28s 278us/step - loss: 0.1401 - accuracy: 0.9326 - val_loss: 3.2050 - val_accuracy: 0.5603\n",
      "\n",
      "Epoch 00058: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-58_val_acc-0.56.h5\n",
      "Epoch 59/75\n",
      "99000/99000 [==============================] - 28s 284us/step - loss: 0.1637 - accuracy: 0.9242 - val_loss: 2.5855 - val_accuracy: 0.5640\n",
      "\n",
      "Epoch 00059: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-59_val_acc-0.56.h5\n",
      "Epoch 60/75\n",
      "99000/99000 [==============================] - 27s 273us/step - loss: 0.4883 - accuracy: 0.7825 - val_loss: 2.5418 - val_accuracy: 0.4961\n",
      "\n",
      "Epoch 00060: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-60_val_acc-0.50.h5\n",
      "Epoch 61/75\n",
      "99000/99000 [==============================] - 28s 284us/step - loss: 0.3535 - accuracy: 0.8241 - val_loss: 2.5857 - val_accuracy: 0.5215\n",
      "\n",
      "Epoch 00061: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-61_val_acc-0.52.h5\n",
      "Epoch 62/75\n",
      "99000/99000 [==============================] - 28s 280us/step - loss: 0.2780 - accuracy: 0.8695 - val_loss: 2.6060 - val_accuracy: 0.5333\n",
      "\n",
      "Epoch 00062: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-62_val_acc-0.53.h5\n",
      "Epoch 63/75\n",
      "99000/99000 [==============================] - 27s 277us/step - loss: 0.3959 - accuracy: 0.8309 - val_loss: 2.5763 - val_accuracy: 0.5094\n",
      "\n",
      "Epoch 00063: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-63_val_acc-0.51.h5\n",
      "Epoch 64/75\n",
      "99000/99000 [==============================] - 28s 283us/step - loss: 0.3124 - accuracy: 0.8571 - val_loss: 2.5432 - val_accuracy: 0.5305\n",
      "\n",
      "Epoch 00064: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-64_val_acc-0.53.h5\n",
      "Epoch 65/75\n",
      "99000/99000 [==============================] - 28s 283us/step - loss: 0.2567 - accuracy: 0.8841 - val_loss: 2.6149 - val_accuracy: 0.5374\n",
      "\n",
      "Epoch 00065: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-65_val_acc-0.54.h5\n",
      "Epoch 66/75\n",
      "99000/99000 [==============================] - 28s 283us/step - loss: 0.2237 - accuracy: 0.8984 - val_loss: 2.5598 - val_accuracy: 0.5475\n",
      "\n",
      "Epoch 00066: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-66_val_acc-0.55.h5\n",
      "Epoch 67/75\n",
      "99000/99000 [==============================] - 27s 277us/step - loss: 0.2102 - accuracy: 0.9039 - val_loss: 2.5648 - val_accuracy: 0.5454\n",
      "\n",
      "Epoch 00067: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-67_val_acc-0.55.h5\n",
      "Epoch 68/75\n",
      "99000/99000 [==============================] - 28s 280us/step - loss: 0.1905 - accuracy: 0.9132 - val_loss: 2.8930 - val_accuracy: 0.5475\n",
      "\n",
      "Epoch 00068: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-68_val_acc-0.55.h5\n",
      "Epoch 69/75\n",
      "99000/99000 [==============================] - 27s 275us/step - loss: 0.1800 - accuracy: 0.9173 - val_loss: 2.9158 - val_accuracy: 0.5443\n",
      "\n",
      "Epoch 00069: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-69_val_acc-0.54.h5\n",
      "Epoch 70/75\n",
      "99000/99000 [==============================] - 28s 282us/step - loss: 0.1743 - accuracy: 0.9177 - val_loss: 2.7106 - val_accuracy: 0.5569\n",
      "\n",
      "Epoch 00070: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-70_val_acc-0.56.h5\n",
      "Epoch 71/75\n",
      "99000/99000 [==============================] - 27s 274us/step - loss: 0.1738 - accuracy: 0.9184 - val_loss: 3.2983 - val_accuracy: 0.5517\n",
      "\n",
      "Epoch 00071: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-71_val_acc-0.55.h5\n",
      "Epoch 72/75\n",
      "99000/99000 [==============================] - 28s 286us/step - loss: 0.1646 - accuracy: 0.9225 - val_loss: 3.0927 - val_accuracy: 0.5557\n",
      "\n",
      "Epoch 00072: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-72_val_acc-0.56.h5\n",
      "Epoch 73/75\n",
      "99000/99000 [==============================] - 28s 282us/step - loss: 0.1553 - accuracy: 0.9267 - val_loss: 3.2683 - val_accuracy: 0.5533\n",
      "\n",
      "Epoch 00073: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-73_val_acc-0.55.h5\n",
      "Epoch 74/75\n",
      "99000/99000 [==============================] - 28s 281us/step - loss: 0.1508 - accuracy: 0.9277 - val_loss: 3.2824 - val_accuracy: 0.5641\n",
      "\n",
      "Epoch 00074: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-74_val_acc-0.56.h5\n",
      "Epoch 75/75\n",
      "99000/99000 [==============================] - 28s 279us/step - loss: 1395.8732 - accuracy: 0.2424 - val_loss: 778.0947 - val_accuracy: 0.1112\n",
      "\n",
      "Epoch 00075: saving model to /content/gdrive/My Drive/Project-EE5611/Model_files/models/scrnn_highsnr/model_epoch-75_val_acc-0.11.h5\n"
     ]
    }
   ],
   "source": [
    "history = m1.fit(x=x_tr,\n",
    "       y=y_tr,\n",
    "       validation_data=[x_ts,y_ts],\n",
    "       batch_size=128,\n",
    "       epochs=75,\n",
    "       shuffle='batch',\n",
    "       callbacks = [checkpoint],\n",
    "       verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 981
    },
    "colab_type": "code",
    "id": "VyZkwyAgMkKd",
    "outputId": "7b1fdd5d-80ca-4d23-85e8-7f5548c5f5f1"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "acc = history.history['accuracy']\n",
    "val_acc = history.history['val_accuracy']\n",
    "loss = history.history['loss']\n",
    "val_loss = history.history['val_loss']\n",
    "epochs = range(1, len(acc) + 1)\n",
    "plt.figure(figsize=(8,8))\n",
    "plt.plot(epochs, acc,label='Training acc')\n",
    "plt.plot(epochs, val_acc,label='Validation acc')\n",
    "plt.title('Training and validation accuracy for Base SCRNN trained on high SNR Range')\n",
    "plt.legend()\n",
    "plt.ylim([0,1])\n",
    "plt.xlim([0,70])\n",
    "plt.figure(figsize=(8,8))\n",
    "plt.plot(epochs, loss,label='Training loss')\n",
    "plt.plot(epochs, val_loss, label='Validation loss')\n",
    "plt.title('Training and validation loss for Base SCRNN trained on high SNR Range')\n",
    "plt.legend()\n",
    "plt.ylim([0,10])\n",
    "plt.xlim([0,70])\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "scrnn_highsnr.ipynb",
   "provenance": []
  },
  "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}