Tensorflow_2_x_Basics.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Tensorflow 2.x Basics.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyObeRrzIOOXndjtW01PMU2v",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/mohameddhameem/TensorflowCertification/blob/main/Tensorflow_2_x_Basics.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rU61FlF5UXtG"
      },
      "source": [
        "# Learn basics of Tensorflow"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fipeBlzdUt9c"
      },
      "source": [
        "## Basic difference between Tensorflow v1 and v2"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EFBmZAZxUyRR",
        "outputId": "9ecc95cb-5e38-4321-cac0-0053b9e3c6b6"
      },
      "source": [
        "# V2 code\n",
        "import tensorflow as tf\n",
        "W = tf.Variable(tf.ones(shape=(2,2)), name = \"W\")\n",
        "b = tf.Variable(tf.zeros(shape=(2)), name = \"b\")\n",
        "\n",
        "@tf.function\n",
        "def model(x):\n",
        "  return W * x + b\n",
        "out_a = model([1,0])\n",
        "print(out_a)"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "tf.Tensor(\n",
            "[[1. 0.]\n",
            " [1. 0.]], shape=(2, 2), dtype=float32)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zEUCZdDoV72i"
      },
      "source": [
        "## TensorFlow 2.0 with Keras\n",
        "Simple Sequential Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lEIEqv3DWF8T"
      },
      "source": [
        "from tensorflow import keras\n",
        "NB_CLASSES = 10\n",
        "RESHAPED = 784\n",
        "model = tf.keras.Sequential()\n",
        "model.add(keras.layers.Dense(NB_CLASSES,\n",
        "                             input_shape=(RESHAPED,),\n",
        "                             kernel_initializer='zeros', #others random_uniform, random_normal\n",
        "                             name='dense_layer',\n",
        "                             activation='softmax'))"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Gvu5rjdWXvac"
      },
      "source": [
        "## MNIST Example\n",
        "### Load and prepare dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Su2YSv9LXHXI",
        "outputId": "232c8d4e-6b59-4aeb-8343-a8a9ac7aa0a0"
      },
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "import numpy as np\n",
        "\n",
        "EPOCH = 200\n",
        "BATCH_SIZE = 128\n",
        "VERBOSE = 1\n",
        "NB_CLASSES = 10\n",
        "VALIDATION_SPLIT = 0.2\n",
        "RESHAPED = 784\n",
        "\n",
        "#Load the dataset from kears\n",
        "(X_train, y_train), (X_test, y_test) = keras.datasets.mnist.load_data()\n",
        "print(X_train.shape, X_test.shape)\n",
        "\n",
        "#Lets reshape the input to 784\n",
        "X_train = X_train.reshape(X_train.shape[0], RESHAPED)\n",
        "X_test = X_test.reshape(X_test.shape[0], RESHAPED)\n",
        "X_train = X_train.astype('float32')\n",
        "X_test = X_test.astype('float32')\n",
        "\n",
        "#Normalize the input within [0, 1]\n",
        "X_train /= 255\n",
        "X_test /= 255\n",
        "\n",
        "print(X_train.shape[0], 'train samples')\n",
        "print(X_test.shape[0], 'test samples')\n",
        "# One-hot encoding of labels\n",
        "y_train = tf.keras.utils.to_categorical(y_train, NB_CLASSES)\n",
        "y_test = tf.keras.utils.to_categorical(y_test, NB_CLASSES)"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(60000, 28, 28) (10000, 28, 28)\n",
            "60000 train samples\n",
            "10000 test samples\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RYU_nwXjnJn-"
      },
      "source": [
        "### Model creation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ci36IjjRYipK",
        "outputId": "2c6c56f8-4103-4663-bdeb-f8265f78a596"
      },
      "source": [
        "# we will use Sequential model\n",
        "model = tf.keras.Sequential()\n",
        "model.add(keras.layers.Dense(NB_CLASSES, input_shape=(RESHAPED,),\n",
        "                             name='dense_layer',\n",
        "                             activation='softmax'))\n",
        "model.compile(optimizer='SGD',\n",
        "              loss='categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n",
        "print(model.summary())\n",
        "model.fit(X_train,y_train,\n",
        "          batch_size=BATCH_SIZE,\n",
        "          epochs=EPOCH,\n",
        "          verbose=VERBOSE,\n",
        "          validation_split=VALIDATION_SPLIT)"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_6\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense_layer (Dense)          (None, 10)                7850      \n",
            "=================================================================\n",
            "Total params: 7,850\n",
            "Trainable params: 7,850\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n",
            "Epoch 1/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 1.7477 - accuracy: 0.5108 - val_loss: 0.8896 - val_accuracy: 0.8338\n",
            "Epoch 2/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.8449 - accuracy: 0.8289 - val_loss: 0.6526 - val_accuracy: 0.8621\n",
            "Epoch 3/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.6661 - accuracy: 0.8509 - val_loss: 0.5580 - val_accuracy: 0.8747\n",
            "Epoch 4/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.5832 - accuracy: 0.8621 - val_loss: 0.5057 - val_accuracy: 0.8789\n",
            "Epoch 5/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.5307 - accuracy: 0.8705 - val_loss: 0.4721 - val_accuracy: 0.8838\n",
            "Epoch 6/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.4979 - accuracy: 0.8748 - val_loss: 0.4483 - val_accuracy: 0.8884\n",
            "Epoch 7/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.4702 - accuracy: 0.8817 - val_loss: 0.4304 - val_accuracy: 0.8919\n",
            "Epoch 8/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.4539 - accuracy: 0.8817 - val_loss: 0.4164 - val_accuracy: 0.8950\n",
            "Epoch 9/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.4418 - accuracy: 0.8849 - val_loss: 0.4050 - val_accuracy: 0.8964\n",
            "Epoch 10/200\n",
            "375/375 [==============================] - 1s 2ms/step - loss: 0.4286 - accuracy: 0.8874 - val_loss: 0.3956 - val_accuracy: 0.8977\n",
            "Epoch 11/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.4213 - accuracy: 0.8890 - val_loss: 0.3876 - val_accuracy: 0.8985\n",
            "Epoch 12/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.4114 - accuracy: 0.8902 - val_loss: 0.3807 - val_accuracy: 0.9009\n",
            "Epoch 13/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.4047 - accuracy: 0.8917 - val_loss: 0.3746 - val_accuracy: 0.9022\n",
            "Epoch 14/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3988 - accuracy: 0.8930 - val_loss: 0.3693 - val_accuracy: 0.9018\n",
            "Epoch 15/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3986 - accuracy: 0.8932 - val_loss: 0.3646 - val_accuracy: 0.9035\n",
            "Epoch 16/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3928 - accuracy: 0.8917 - val_loss: 0.3604 - val_accuracy: 0.9051\n",
            "Epoch 17/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3819 - accuracy: 0.8979 - val_loss: 0.3566 - val_accuracy: 0.9055\n",
            "Epoch 18/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3755 - accuracy: 0.8975 - val_loss: 0.3532 - val_accuracy: 0.9061\n",
            "Epoch 19/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3752 - accuracy: 0.8981 - val_loss: 0.3500 - val_accuracy: 0.9062\n",
            "Epoch 20/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3727 - accuracy: 0.8990 - val_loss: 0.3469 - val_accuracy: 0.9072\n",
            "Epoch 21/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3665 - accuracy: 0.9018 - val_loss: 0.3443 - val_accuracy: 0.9070\n",
            "Epoch 22/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3672 - accuracy: 0.8993 - val_loss: 0.3418 - val_accuracy: 0.9072\n",
            "Epoch 23/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3603 - accuracy: 0.9000 - val_loss: 0.3394 - val_accuracy: 0.9091\n",
            "Epoch 24/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3537 - accuracy: 0.9016 - val_loss: 0.3371 - val_accuracy: 0.9093\n",
            "Epoch 25/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3563 - accuracy: 0.9014 - val_loss: 0.3351 - val_accuracy: 0.9097\n",
            "Epoch 26/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3542 - accuracy: 0.9021 - val_loss: 0.3332 - val_accuracy: 0.9097\n",
            "Epoch 27/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3501 - accuracy: 0.9041 - val_loss: 0.3313 - val_accuracy: 0.9101\n",
            "Epoch 28/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3520 - accuracy: 0.9025 - val_loss: 0.3297 - val_accuracy: 0.9110\n",
            "Epoch 29/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3442 - accuracy: 0.9047 - val_loss: 0.3280 - val_accuracy: 0.9109\n",
            "Epoch 30/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3468 - accuracy: 0.9036 - val_loss: 0.3265 - val_accuracy: 0.9112\n",
            "Epoch 31/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3433 - accuracy: 0.9039 - val_loss: 0.3250 - val_accuracy: 0.9112\n",
            "Epoch 32/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3441 - accuracy: 0.9043 - val_loss: 0.3236 - val_accuracy: 0.9120\n",
            "Epoch 33/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3424 - accuracy: 0.9052 - val_loss: 0.3223 - val_accuracy: 0.9121\n",
            "Epoch 34/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3401 - accuracy: 0.9045 - val_loss: 0.3210 - val_accuracy: 0.9121\n",
            "Epoch 35/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3377 - accuracy: 0.9060 - val_loss: 0.3198 - val_accuracy: 0.9118\n",
            "Epoch 36/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3441 - accuracy: 0.9033 - val_loss: 0.3187 - val_accuracy: 0.9122\n",
            "Epoch 37/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3436 - accuracy: 0.9047 - val_loss: 0.3176 - val_accuracy: 0.9122\n",
            "Epoch 38/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3354 - accuracy: 0.9071 - val_loss: 0.3165 - val_accuracy: 0.9129\n",
            "Epoch 39/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3299 - accuracy: 0.9086 - val_loss: 0.3156 - val_accuracy: 0.9134\n",
            "Epoch 40/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3331 - accuracy: 0.9070 - val_loss: 0.3146 - val_accuracy: 0.9133\n",
            "Epoch 41/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3390 - accuracy: 0.9064 - val_loss: 0.3137 - val_accuracy: 0.9137\n",
            "Epoch 42/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3301 - accuracy: 0.9073 - val_loss: 0.3127 - val_accuracy: 0.9137\n",
            "Epoch 43/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3244 - accuracy: 0.9098 - val_loss: 0.3120 - val_accuracy: 0.9142\n",
            "Epoch 44/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3189 - accuracy: 0.9113 - val_loss: 0.3110 - val_accuracy: 0.9146\n",
            "Epoch 45/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3210 - accuracy: 0.9100 - val_loss: 0.3103 - val_accuracy: 0.9147\n",
            "Epoch 46/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3248 - accuracy: 0.9110 - val_loss: 0.3095 - val_accuracy: 0.9152\n",
            "Epoch 47/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3250 - accuracy: 0.9090 - val_loss: 0.3087 - val_accuracy: 0.9154\n",
            "Epoch 48/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3259 - accuracy: 0.9105 - val_loss: 0.3078 - val_accuracy: 0.9147\n",
            "Epoch 49/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3288 - accuracy: 0.9075 - val_loss: 0.3072 - val_accuracy: 0.9153\n",
            "Epoch 50/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3233 - accuracy: 0.9100 - val_loss: 0.3065 - val_accuracy: 0.9154\n",
            "Epoch 51/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3203 - accuracy: 0.9107 - val_loss: 0.3059 - val_accuracy: 0.9158\n",
            "Epoch 52/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3187 - accuracy: 0.9119 - val_loss: 0.3052 - val_accuracy: 0.9159\n",
            "Epoch 53/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3169 - accuracy: 0.9105 - val_loss: 0.3046 - val_accuracy: 0.9162\n",
            "Epoch 54/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3152 - accuracy: 0.9119 - val_loss: 0.3040 - val_accuracy: 0.9159\n",
            "Epoch 55/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3166 - accuracy: 0.9127 - val_loss: 0.3036 - val_accuracy: 0.9158\n",
            "Epoch 56/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3191 - accuracy: 0.9103 - val_loss: 0.3028 - val_accuracy: 0.9161\n",
            "Epoch 57/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3140 - accuracy: 0.9118 - val_loss: 0.3023 - val_accuracy: 0.9166\n",
            "Epoch 58/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3133 - accuracy: 0.9118 - val_loss: 0.3017 - val_accuracy: 0.9164\n",
            "Epoch 59/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3153 - accuracy: 0.9109 - val_loss: 0.3012 - val_accuracy: 0.9162\n",
            "Epoch 60/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3178 - accuracy: 0.9110 - val_loss: 0.3007 - val_accuracy: 0.9159\n",
            "Epoch 61/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3187 - accuracy: 0.9109 - val_loss: 0.3002 - val_accuracy: 0.9164\n",
            "Epoch 62/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3149 - accuracy: 0.9123 - val_loss: 0.2998 - val_accuracy: 0.9169\n",
            "Epoch 63/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3135 - accuracy: 0.9133 - val_loss: 0.2993 - val_accuracy: 0.9173\n",
            "Epoch 64/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3124 - accuracy: 0.9117 - val_loss: 0.2988 - val_accuracy: 0.9173\n",
            "Epoch 65/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3142 - accuracy: 0.9129 - val_loss: 0.2983 - val_accuracy: 0.9175\n",
            "Epoch 66/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3096 - accuracy: 0.9139 - val_loss: 0.2978 - val_accuracy: 0.9172\n",
            "Epoch 67/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3087 - accuracy: 0.9153 - val_loss: 0.2974 - val_accuracy: 0.9181\n",
            "Epoch 68/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3081 - accuracy: 0.9151 - val_loss: 0.2970 - val_accuracy: 0.9180\n",
            "Epoch 69/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3075 - accuracy: 0.9158 - val_loss: 0.2965 - val_accuracy: 0.9184\n",
            "Epoch 70/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3143 - accuracy: 0.9126 - val_loss: 0.2962 - val_accuracy: 0.9184\n",
            "Epoch 71/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3064 - accuracy: 0.9140 - val_loss: 0.2959 - val_accuracy: 0.9183\n",
            "Epoch 72/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3081 - accuracy: 0.9139 - val_loss: 0.2954 - val_accuracy: 0.9183\n",
            "Epoch 73/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3054 - accuracy: 0.9153 - val_loss: 0.2950 - val_accuracy: 0.9187\n",
            "Epoch 74/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3122 - accuracy: 0.9126 - val_loss: 0.2947 - val_accuracy: 0.9187\n",
            "Epoch 75/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3026 - accuracy: 0.9168 - val_loss: 0.2943 - val_accuracy: 0.9189\n",
            "Epoch 76/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.3008 - accuracy: 0.9151 - val_loss: 0.2940 - val_accuracy: 0.9188\n",
            "Epoch 77/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3076 - accuracy: 0.9141 - val_loss: 0.2936 - val_accuracy: 0.9187\n",
            "Epoch 78/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.2998 - accuracy: 0.9142 - val_loss: 0.2933 - val_accuracy: 0.9190\n",
            "Epoch 79/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3090 - accuracy: 0.9167 - val_loss: 0.2930 - val_accuracy: 0.9187\n",
            "Epoch 80/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3055 - accuracy: 0.9147 - val_loss: 0.2926 - val_accuracy: 0.9189\n",
            "Epoch 81/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3016 - accuracy: 0.9166 - val_loss: 0.2924 - val_accuracy: 0.9191\n",
            "Epoch 82/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3000 - accuracy: 0.9169 - val_loss: 0.2920 - val_accuracy: 0.9192\n",
            "Epoch 83/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3069 - accuracy: 0.9146 - val_loss: 0.2918 - val_accuracy: 0.9197\n",
            "Epoch 84/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.2941 - accuracy: 0.9178 - val_loss: 0.2914 - val_accuracy: 0.9191\n",
            "Epoch 85/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.3015 - accuracy: 0.9169 - val_loss: 0.2911 - val_accuracy: 0.9197\n",
            "Epoch 86/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3014 - accuracy: 0.9174 - val_loss: 0.2908 - val_accuracy: 0.9193\n",
            "Epoch 87/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3000 - accuracy: 0.9156 - val_loss: 0.2906 - val_accuracy: 0.9193\n",
            "Epoch 88/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2963 - accuracy: 0.9168 - val_loss: 0.2903 - val_accuracy: 0.9194\n",
            "Epoch 89/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2970 - accuracy: 0.9180 - val_loss: 0.2900 - val_accuracy: 0.9193\n",
            "Epoch 90/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2942 - accuracy: 0.9167 - val_loss: 0.2897 - val_accuracy: 0.9197\n",
            "Epoch 91/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2986 - accuracy: 0.9177 - val_loss: 0.2895 - val_accuracy: 0.9202\n",
            "Epoch 92/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2990 - accuracy: 0.9162 - val_loss: 0.2892 - val_accuracy: 0.9193\n",
            "Epoch 93/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2912 - accuracy: 0.9189 - val_loss: 0.2889 - val_accuracy: 0.9196\n",
            "Epoch 94/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2944 - accuracy: 0.9179 - val_loss: 0.2887 - val_accuracy: 0.9199\n",
            "Epoch 95/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2974 - accuracy: 0.9162 - val_loss: 0.2885 - val_accuracy: 0.9196\n",
            "Epoch 96/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2925 - accuracy: 0.9169 - val_loss: 0.2882 - val_accuracy: 0.9197\n",
            "Epoch 97/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2992 - accuracy: 0.9165 - val_loss: 0.2880 - val_accuracy: 0.9194\n",
            "Epoch 98/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2969 - accuracy: 0.9163 - val_loss: 0.2878 - val_accuracy: 0.9195\n",
            "Epoch 99/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2881 - accuracy: 0.9190 - val_loss: 0.2875 - val_accuracy: 0.9202\n",
            "Epoch 100/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2976 - accuracy: 0.9163 - val_loss: 0.2873 - val_accuracy: 0.9199\n",
            "Epoch 101/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.3018 - accuracy: 0.9151 - val_loss: 0.2871 - val_accuracy: 0.9205\n",
            "Epoch 102/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2908 - accuracy: 0.9173 - val_loss: 0.2869 - val_accuracy: 0.9203\n",
            "Epoch 103/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.3000 - accuracy: 0.9171 - val_loss: 0.2866 - val_accuracy: 0.9208\n",
            "Epoch 104/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2900 - accuracy: 0.9193 - val_loss: 0.2864 - val_accuracy: 0.9202\n",
            "Epoch 105/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2972 - accuracy: 0.9164 - val_loss: 0.2862 - val_accuracy: 0.9210\n",
            "Epoch 106/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2945 - accuracy: 0.9173 - val_loss: 0.2860 - val_accuracy: 0.9208\n",
            "Epoch 107/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2850 - accuracy: 0.9197 - val_loss: 0.2860 - val_accuracy: 0.9212\n",
            "Epoch 108/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.2939 - accuracy: 0.9173 - val_loss: 0.2856 - val_accuracy: 0.9204\n",
            "Epoch 109/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2914 - accuracy: 0.9196 - val_loss: 0.2855 - val_accuracy: 0.9212\n",
            "Epoch 110/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.2896 - accuracy: 0.9188 - val_loss: 0.2853 - val_accuracy: 0.9212\n",
            "Epoch 111/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2978 - accuracy: 0.9174 - val_loss: 0.2850 - val_accuracy: 0.9211\n",
            "Epoch 112/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.2931 - accuracy: 0.9171 - val_loss: 0.2848 - val_accuracy: 0.9214\n",
            "Epoch 113/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.2875 - accuracy: 0.9202 - val_loss: 0.2846 - val_accuracy: 0.9215\n",
            "Epoch 114/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2885 - accuracy: 0.9201 - val_loss: 0.2845 - val_accuracy: 0.9213\n",
            "Epoch 115/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.2845 - accuracy: 0.9205 - val_loss: 0.2843 - val_accuracy: 0.9217\n",
            "Epoch 116/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2995 - accuracy: 0.9158 - val_loss: 0.2842 - val_accuracy: 0.9213\n",
            "Epoch 117/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2912 - accuracy: 0.9191 - val_loss: 0.2840 - val_accuracy: 0.9209\n",
            "Epoch 118/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2937 - accuracy: 0.9179 - val_loss: 0.2838 - val_accuracy: 0.9217\n",
            "Epoch 119/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.2894 - accuracy: 0.9192 - val_loss: 0.2836 - val_accuracy: 0.9218\n",
            "Epoch 120/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.2860 - accuracy: 0.9193 - val_loss: 0.2834 - val_accuracy: 0.9219\n",
            "Epoch 121/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.2907 - accuracy: 0.9184 - val_loss: 0.2833 - val_accuracy: 0.9215\n",
            "Epoch 122/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.2888 - accuracy: 0.9193 - val_loss: 0.2832 - val_accuracy: 0.9215\n",
            "Epoch 123/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.2904 - accuracy: 0.9174 - val_loss: 0.2829 - val_accuracy: 0.9217\n",
            "Epoch 124/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2877 - accuracy: 0.9196 - val_loss: 0.2828 - val_accuracy: 0.9221\n",
            "Epoch 125/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2900 - accuracy: 0.9182 - val_loss: 0.2827 - val_accuracy: 0.9219\n",
            "Epoch 126/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2918 - accuracy: 0.9176 - val_loss: 0.2825 - val_accuracy: 0.9219\n",
            "Epoch 127/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2865 - accuracy: 0.9203 - val_loss: 0.2823 - val_accuracy: 0.9221\n",
            "Epoch 128/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2876 - accuracy: 0.9205 - val_loss: 0.2822 - val_accuracy: 0.9223\n",
            "Epoch 129/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2900 - accuracy: 0.9197 - val_loss: 0.2821 - val_accuracy: 0.9219\n",
            "Epoch 130/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2809 - accuracy: 0.9219 - val_loss: 0.2819 - val_accuracy: 0.9219\n",
            "Epoch 131/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.2842 - accuracy: 0.9216 - val_loss: 0.2818 - val_accuracy: 0.9223\n",
            "Epoch 132/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2903 - accuracy: 0.9185 - val_loss: 0.2816 - val_accuracy: 0.9227\n",
            "Epoch 133/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2851 - accuracy: 0.9198 - val_loss: 0.2815 - val_accuracy: 0.9222\n",
            "Epoch 134/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2805 - accuracy: 0.9206 - val_loss: 0.2813 - val_accuracy: 0.9223\n",
            "Epoch 135/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2834 - accuracy: 0.9205 - val_loss: 0.2812 - val_accuracy: 0.9225\n",
            "Epoch 136/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2849 - accuracy: 0.9203 - val_loss: 0.2811 - val_accuracy: 0.9224\n",
            "Epoch 137/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2859 - accuracy: 0.9197 - val_loss: 0.2810 - val_accuracy: 0.9223\n",
            "Epoch 138/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2898 - accuracy: 0.9193 - val_loss: 0.2808 - val_accuracy: 0.9226\n",
            "Epoch 139/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2856 - accuracy: 0.9214 - val_loss: 0.2807 - val_accuracy: 0.9225\n",
            "Epoch 140/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2845 - accuracy: 0.9210 - val_loss: 0.2805 - val_accuracy: 0.9222\n",
            "Epoch 141/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2854 - accuracy: 0.9206 - val_loss: 0.2804 - val_accuracy: 0.9225\n",
            "Epoch 142/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2852 - accuracy: 0.9197 - val_loss: 0.2803 - val_accuracy: 0.9230\n",
            "Epoch 143/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.2891 - accuracy: 0.9190 - val_loss: 0.2802 - val_accuracy: 0.9226\n",
            "Epoch 144/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2845 - accuracy: 0.9203 - val_loss: 0.2800 - val_accuracy: 0.9227\n",
            "Epoch 145/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2849 - accuracy: 0.9209 - val_loss: 0.2800 - val_accuracy: 0.9223\n",
            "Epoch 146/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2769 - accuracy: 0.9218 - val_loss: 0.2799 - val_accuracy: 0.9226\n",
            "Epoch 147/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2825 - accuracy: 0.9191 - val_loss: 0.2797 - val_accuracy: 0.9227\n",
            "Epoch 148/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2790 - accuracy: 0.9222 - val_loss: 0.2796 - val_accuracy: 0.9230\n",
            "Epoch 149/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2759 - accuracy: 0.9213 - val_loss: 0.2795 - val_accuracy: 0.9227\n",
            "Epoch 150/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2860 - accuracy: 0.9195 - val_loss: 0.2794 - val_accuracy: 0.9224\n",
            "Epoch 151/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2927 - accuracy: 0.9177 - val_loss: 0.2792 - val_accuracy: 0.9226\n",
            "Epoch 152/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2870 - accuracy: 0.9200 - val_loss: 0.2792 - val_accuracy: 0.9226\n",
            "Epoch 153/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2811 - accuracy: 0.9201 - val_loss: 0.2791 - val_accuracy: 0.9229\n",
            "Epoch 154/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2810 - accuracy: 0.9204 - val_loss: 0.2789 - val_accuracy: 0.9231\n",
            "Epoch 155/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2804 - accuracy: 0.9221 - val_loss: 0.2789 - val_accuracy: 0.9232\n",
            "Epoch 156/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2895 - accuracy: 0.9188 - val_loss: 0.2788 - val_accuracy: 0.9233\n",
            "Epoch 157/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2786 - accuracy: 0.9216 - val_loss: 0.2786 - val_accuracy: 0.9233\n",
            "Epoch 158/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2813 - accuracy: 0.9215 - val_loss: 0.2784 - val_accuracy: 0.9233\n",
            "Epoch 159/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2871 - accuracy: 0.9199 - val_loss: 0.2783 - val_accuracy: 0.9229\n",
            "Epoch 160/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2832 - accuracy: 0.9211 - val_loss: 0.2783 - val_accuracy: 0.9226\n",
            "Epoch 161/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2818 - accuracy: 0.9217 - val_loss: 0.2782 - val_accuracy: 0.9231\n",
            "Epoch 162/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2840 - accuracy: 0.9210 - val_loss: 0.2781 - val_accuracy: 0.9229\n",
            "Epoch 163/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2804 - accuracy: 0.9215 - val_loss: 0.2780 - val_accuracy: 0.9233\n",
            "Epoch 164/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2796 - accuracy: 0.9210 - val_loss: 0.2779 - val_accuracy: 0.9233\n",
            "Epoch 165/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2799 - accuracy: 0.9224 - val_loss: 0.2778 - val_accuracy: 0.9232\n",
            "Epoch 166/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2830 - accuracy: 0.9205 - val_loss: 0.2777 - val_accuracy: 0.9234\n",
            "Epoch 167/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2825 - accuracy: 0.9206 - val_loss: 0.2776 - val_accuracy: 0.9233\n",
            "Epoch 168/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2808 - accuracy: 0.9197 - val_loss: 0.2775 - val_accuracy: 0.9234\n",
            "Epoch 169/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2799 - accuracy: 0.9222 - val_loss: 0.2774 - val_accuracy: 0.9236\n",
            "Epoch 170/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2771 - accuracy: 0.9234 - val_loss: 0.2773 - val_accuracy: 0.9238\n",
            "Epoch 171/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2723 - accuracy: 0.9241 - val_loss: 0.2772 - val_accuracy: 0.9238\n",
            "Epoch 172/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2758 - accuracy: 0.9228 - val_loss: 0.2771 - val_accuracy: 0.9235\n",
            "Epoch 173/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2739 - accuracy: 0.9221 - val_loss: 0.2770 - val_accuracy: 0.9236\n",
            "Epoch 174/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2794 - accuracy: 0.9234 - val_loss: 0.2769 - val_accuracy: 0.9236\n",
            "Epoch 175/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2832 - accuracy: 0.9202 - val_loss: 0.2769 - val_accuracy: 0.9238\n",
            "Epoch 176/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2774 - accuracy: 0.9229 - val_loss: 0.2769 - val_accuracy: 0.9237\n",
            "Epoch 177/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2777 - accuracy: 0.9227 - val_loss: 0.2766 - val_accuracy: 0.9237\n",
            "Epoch 178/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2756 - accuracy: 0.9228 - val_loss: 0.2766 - val_accuracy: 0.9240\n",
            "Epoch 179/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2754 - accuracy: 0.9231 - val_loss: 0.2765 - val_accuracy: 0.9237\n",
            "Epoch 180/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2728 - accuracy: 0.9247 - val_loss: 0.2764 - val_accuracy: 0.9237\n",
            "Epoch 181/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2814 - accuracy: 0.9212 - val_loss: 0.2763 - val_accuracy: 0.9238\n",
            "Epoch 182/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2779 - accuracy: 0.9231 - val_loss: 0.2763 - val_accuracy: 0.9237\n",
            "Epoch 183/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2805 - accuracy: 0.9220 - val_loss: 0.2762 - val_accuracy: 0.9240\n",
            "Epoch 184/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2735 - accuracy: 0.9233 - val_loss: 0.2761 - val_accuracy: 0.9243\n",
            "Epoch 185/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2792 - accuracy: 0.9226 - val_loss: 0.2761 - val_accuracy: 0.9241\n",
            "Epoch 186/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2733 - accuracy: 0.9237 - val_loss: 0.2760 - val_accuracy: 0.9237\n",
            "Epoch 187/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2737 - accuracy: 0.9235 - val_loss: 0.2759 - val_accuracy: 0.9233\n",
            "Epoch 188/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2726 - accuracy: 0.9236 - val_loss: 0.2758 - val_accuracy: 0.9237\n",
            "Epoch 189/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2747 - accuracy: 0.9230 - val_loss: 0.2757 - val_accuracy: 0.9239\n",
            "Epoch 190/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2781 - accuracy: 0.9241 - val_loss: 0.2757 - val_accuracy: 0.9239\n",
            "Epoch 191/200\n",
            "375/375 [==============================] - 1s 4ms/step - loss: 0.2788 - accuracy: 0.9227 - val_loss: 0.2757 - val_accuracy: 0.9243\n",
            "Epoch 192/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2773 - accuracy: 0.9238 - val_loss: 0.2756 - val_accuracy: 0.9239\n",
            "Epoch 193/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2769 - accuracy: 0.9228 - val_loss: 0.2755 - val_accuracy: 0.9240\n",
            "Epoch 194/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2793 - accuracy: 0.9226 - val_loss: 0.2753 - val_accuracy: 0.9237\n",
            "Epoch 195/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2760 - accuracy: 0.9233 - val_loss: 0.2753 - val_accuracy: 0.9238\n",
            "Epoch 196/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2729 - accuracy: 0.9228 - val_loss: 0.2752 - val_accuracy: 0.9239\n",
            "Epoch 197/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2784 - accuracy: 0.9229 - val_loss: 0.2752 - val_accuracy: 0.9240\n",
            "Epoch 198/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2782 - accuracy: 0.9226 - val_loss: 0.2751 - val_accuracy: 0.9241\n",
            "Epoch 199/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2788 - accuracy: 0.9225 - val_loss: 0.2751 - val_accuracy: 0.9241\n",
            "Epoch 200/200\n",
            "375/375 [==============================] - 1s 3ms/step - loss: 0.2755 - accuracy: 0.9235 - val_loss: 0.2750 - val_accuracy: 0.9241\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7fcb4faa60f0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9VVJK4H3qKs4"
      },
      "source": [
        "### Model evaluation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "02gbrqnZpl-a",
        "outputId": "9254625f-aff6-474b-99d1-d183dbd508ed"
      },
      "source": [
        "test_loss, test_accuracy = model.evaluate(X_test, y_test)\n",
        "print('Test Loss :', test_loss, ' Test Accuracy:', test_accuracy)"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 1s 2ms/step - loss: 0.2772 - accuracy: 0.9224\n",
            "Test Loss : 0.27716150879859924  Test Accuracy: 0.9223999977111816\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "opKcnPATsSTX"
      },
      "source": [
        "### Finetune model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "twYorLMRqav3",
        "outputId": "91b6c17f-1424-4aed-ef71-6cb61aef92ff"
      },
      "source": [
        "N_HIDDEN = 128\n",
        "model = tf.keras.models.Sequential()\n",
        "model.add(keras.layers.Dense(N_HIDDEN,\n",
        "                             input_shape=(RESHAPED,),\n",
        "                             name='dense_layer_1',\n",
        "                             activation='relu'))\n",
        "model.add(keras.layers.Dense(N_HIDDEN,\n",
        "                             name='dense_layer_2',\n",
        "                             activation='relu'))\n",
        "#final layer\n",
        "model.add(keras.layers.Dense(NB_CLASSES,\n",
        "                             name='dense_layer_final',\n",
        "                             activation='softmax'))\n",
        "model.summary()\n",
        "model.compile(optimizer='SGD',\n",
        "              loss='categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n",
        "model.fit(X_train,y_train,\n",
        "          batch_size=BATCH_SIZE,\n",
        "          epochs=EPOCH,\n",
        "          verbose=VERBOSE,\n",
        "          validation_split=VALIDATION_SPLIT)\n",
        "test_loss, test_accuracy = model.evaluate(X_test,y_test)\n",
        "print('Test Loss :', test_loss, ' Test Accuracy:', test_accuracy)"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_7\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense_layer_1 (Dense)        (None, 128)               100480    \n",
            "_________________________________________________________________\n",
            "dense_layer_2 (Dense)        (None, 128)               16512     \n",
            "_________________________________________________________________\n",
            "dense_layer_final (Dense)    (None, 10)                1290      \n",
            "=================================================================\n",
            "Total params: 118,282\n",
            "Trainable params: 118,282\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/200\n",
            "375/375 [==============================] - 5s 10ms/step - loss: 1.8373 - accuracy: 0.4738 - val_loss: 0.7028 - val_accuracy: 0.8504\n",
            "Epoch 2/200\n",
            "375/375 [==============================] - 3s 8ms/step - loss: 0.6484 - accuracy: 0.8450 - val_loss: 0.4482 - val_accuracy: 0.8817\n",
            "Epoch 3/200\n",
            "375/375 [==============================] - 3s 7ms/step - loss: 0.4524 - accuracy: 0.8766 - val_loss: 0.3737 - val_accuracy: 0.8975\n",
            "Epoch 4/200\n",
            "375/375 [==============================] - 2s 7ms/step - loss: 0.3903 - accuracy: 0.8900 - val_loss: 0.3387 - val_accuracy: 0.9043\n",
            "Epoch 5/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.3505 - accuracy: 0.9012 - val_loss: 0.3173 - val_accuracy: 0.9090\n",
            "Epoch 6/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.3327 - accuracy: 0.9033 - val_loss: 0.3003 - val_accuracy: 0.9140\n",
            "Epoch 7/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.3177 - accuracy: 0.9097 - val_loss: 0.2845 - val_accuracy: 0.9192\n",
            "Epoch 8/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.3025 - accuracy: 0.9130 - val_loss: 0.2723 - val_accuracy: 0.9227\n",
            "Epoch 9/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.2885 - accuracy: 0.9177 - val_loss: 0.2623 - val_accuracy: 0.9247\n",
            "Epoch 10/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.2690 - accuracy: 0.9238 - val_loss: 0.2528 - val_accuracy: 0.9265\n",
            "Epoch 11/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.2620 - accuracy: 0.9253 - val_loss: 0.2452 - val_accuracy: 0.9310\n",
            "Epoch 12/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.2503 - accuracy: 0.9285 - val_loss: 0.2375 - val_accuracy: 0.9327\n",
            "Epoch 13/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.2432 - accuracy: 0.9316 - val_loss: 0.2296 - val_accuracy: 0.9351\n",
            "Epoch 14/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.2342 - accuracy: 0.9332 - val_loss: 0.2224 - val_accuracy: 0.9362\n",
            "Epoch 15/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.2258 - accuracy: 0.9370 - val_loss: 0.2172 - val_accuracy: 0.9373\n",
            "Epoch 16/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.2193 - accuracy: 0.9368 - val_loss: 0.2112 - val_accuracy: 0.9401\n",
            "Epoch 17/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.2138 - accuracy: 0.9390 - val_loss: 0.2061 - val_accuracy: 0.9417\n",
            "Epoch 18/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.2104 - accuracy: 0.9404 - val_loss: 0.2012 - val_accuracy: 0.9431\n",
            "Epoch 19/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1948 - accuracy: 0.9451 - val_loss: 0.1959 - val_accuracy: 0.9442\n",
            "Epoch 20/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.1967 - accuracy: 0.9436 - val_loss: 0.1916 - val_accuracy: 0.9466\n",
            "Epoch 21/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.1841 - accuracy: 0.9467 - val_loss: 0.1886 - val_accuracy: 0.9472\n",
            "Epoch 22/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.1825 - accuracy: 0.9462 - val_loss: 0.1842 - val_accuracy: 0.9484\n",
            "Epoch 23/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1777 - accuracy: 0.9486 - val_loss: 0.1803 - val_accuracy: 0.9494\n",
            "Epoch 24/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1733 - accuracy: 0.9501 - val_loss: 0.1758 - val_accuracy: 0.9508\n",
            "Epoch 25/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1679 - accuracy: 0.9521 - val_loss: 0.1726 - val_accuracy: 0.9519\n",
            "Epoch 26/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1606 - accuracy: 0.9535 - val_loss: 0.1706 - val_accuracy: 0.9514\n",
            "Epoch 27/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1606 - accuracy: 0.9534 - val_loss: 0.1657 - val_accuracy: 0.9532\n",
            "Epoch 28/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.1587 - accuracy: 0.9541 - val_loss: 0.1640 - val_accuracy: 0.9540\n",
            "Epoch 29/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.1525 - accuracy: 0.9556 - val_loss: 0.1610 - val_accuracy: 0.9548\n",
            "Epoch 30/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.1508 - accuracy: 0.9568 - val_loss: 0.1584 - val_accuracy: 0.9565\n",
            "Epoch 31/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.1446 - accuracy: 0.9582 - val_loss: 0.1566 - val_accuracy: 0.9565\n",
            "Epoch 32/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1448 - accuracy: 0.9569 - val_loss: 0.1534 - val_accuracy: 0.9578\n",
            "Epoch 33/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1389 - accuracy: 0.9604 - val_loss: 0.1506 - val_accuracy: 0.9583\n",
            "Epoch 34/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.1363 - accuracy: 0.9606 - val_loss: 0.1483 - val_accuracy: 0.9594\n",
            "Epoch 35/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.1314 - accuracy: 0.9620 - val_loss: 0.1473 - val_accuracy: 0.9593\n",
            "Epoch 36/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.1327 - accuracy: 0.9615 - val_loss: 0.1443 - val_accuracy: 0.9604\n",
            "Epoch 37/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1292 - accuracy: 0.9633 - val_loss: 0.1438 - val_accuracy: 0.9595\n",
            "Epoch 38/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1240 - accuracy: 0.9637 - val_loss: 0.1416 - val_accuracy: 0.9613\n",
            "Epoch 39/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1221 - accuracy: 0.9637 - val_loss: 0.1399 - val_accuracy: 0.9621\n",
            "Epoch 40/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1181 - accuracy: 0.9654 - val_loss: 0.1379 - val_accuracy: 0.9628\n",
            "Epoch 41/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1193 - accuracy: 0.9654 - val_loss: 0.1359 - val_accuracy: 0.9636\n",
            "Epoch 42/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1167 - accuracy: 0.9658 - val_loss: 0.1346 - val_accuracy: 0.9630\n",
            "Epoch 43/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1138 - accuracy: 0.9675 - val_loss: 0.1333 - val_accuracy: 0.9643\n",
            "Epoch 44/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1081 - accuracy: 0.9686 - val_loss: 0.1321 - val_accuracy: 0.9638\n",
            "Epoch 45/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.1091 - accuracy: 0.9694 - val_loss: 0.1304 - val_accuracy: 0.9652\n",
            "Epoch 46/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1084 - accuracy: 0.9695 - val_loss: 0.1299 - val_accuracy: 0.9648\n",
            "Epoch 47/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1072 - accuracy: 0.9689 - val_loss: 0.1284 - val_accuracy: 0.9650\n",
            "Epoch 48/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1010 - accuracy: 0.9719 - val_loss: 0.1266 - val_accuracy: 0.9655\n",
            "Epoch 49/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.1032 - accuracy: 0.9702 - val_loss: 0.1253 - val_accuracy: 0.9662\n",
            "Epoch 50/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0961 - accuracy: 0.9737 - val_loss: 0.1248 - val_accuracy: 0.9653\n",
            "Epoch 51/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0998 - accuracy: 0.9715 - val_loss: 0.1237 - val_accuracy: 0.9663\n",
            "Epoch 52/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0965 - accuracy: 0.9732 - val_loss: 0.1223 - val_accuracy: 0.9672\n",
            "Epoch 53/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0946 - accuracy: 0.9728 - val_loss: 0.1220 - val_accuracy: 0.9661\n",
            "Epoch 54/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0932 - accuracy: 0.9735 - val_loss: 0.1211 - val_accuracy: 0.9671\n",
            "Epoch 55/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0902 - accuracy: 0.9747 - val_loss: 0.1193 - val_accuracy: 0.9680\n",
            "Epoch 56/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0905 - accuracy: 0.9755 - val_loss: 0.1180 - val_accuracy: 0.9680\n",
            "Epoch 57/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0892 - accuracy: 0.9759 - val_loss: 0.1176 - val_accuracy: 0.9675\n",
            "Epoch 58/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0893 - accuracy: 0.9755 - val_loss: 0.1162 - val_accuracy: 0.9678\n",
            "Epoch 59/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0854 - accuracy: 0.9765 - val_loss: 0.1158 - val_accuracy: 0.9674\n",
            "Epoch 60/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0836 - accuracy: 0.9770 - val_loss: 0.1157 - val_accuracy: 0.9687\n",
            "Epoch 61/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0828 - accuracy: 0.9767 - val_loss: 0.1142 - val_accuracy: 0.9688\n",
            "Epoch 62/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0812 - accuracy: 0.9779 - val_loss: 0.1134 - val_accuracy: 0.9681\n",
            "Epoch 63/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0781 - accuracy: 0.9786 - val_loss: 0.1131 - val_accuracy: 0.9691\n",
            "Epoch 64/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0784 - accuracy: 0.9784 - val_loss: 0.1129 - val_accuracy: 0.9687\n",
            "Epoch 65/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0768 - accuracy: 0.9788 - val_loss: 0.1119 - val_accuracy: 0.9688\n",
            "Epoch 66/200\n",
            "375/375 [==============================] - 3s 8ms/step - loss: 0.0750 - accuracy: 0.9789 - val_loss: 0.1108 - val_accuracy: 0.9691\n",
            "Epoch 67/200\n",
            "375/375 [==============================] - 3s 7ms/step - loss: 0.0749 - accuracy: 0.9797 - val_loss: 0.1108 - val_accuracy: 0.9691\n",
            "Epoch 68/200\n",
            "375/375 [==============================] - 3s 8ms/step - loss: 0.0744 - accuracy: 0.9795 - val_loss: 0.1095 - val_accuracy: 0.9693\n",
            "Epoch 69/200\n",
            "375/375 [==============================] - 3s 8ms/step - loss: 0.0714 - accuracy: 0.9807 - val_loss: 0.1095 - val_accuracy: 0.9695\n",
            "Epoch 70/200\n",
            "375/375 [==============================] - 3s 8ms/step - loss: 0.0714 - accuracy: 0.9806 - val_loss: 0.1086 - val_accuracy: 0.9691\n",
            "Epoch 71/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0703 - accuracy: 0.9802 - val_loss: 0.1083 - val_accuracy: 0.9692\n",
            "Epoch 72/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0700 - accuracy: 0.9814 - val_loss: 0.1073 - val_accuracy: 0.9701\n",
            "Epoch 73/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0678 - accuracy: 0.9817 - val_loss: 0.1080 - val_accuracy: 0.9693\n",
            "Epoch 74/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0692 - accuracy: 0.9818 - val_loss: 0.1063 - val_accuracy: 0.9696\n",
            "Epoch 75/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0701 - accuracy: 0.9814 - val_loss: 0.1051 - val_accuracy: 0.9701\n",
            "Epoch 76/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0657 - accuracy: 0.9821 - val_loss: 0.1049 - val_accuracy: 0.9700\n",
            "Epoch 77/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0623 - accuracy: 0.9835 - val_loss: 0.1047 - val_accuracy: 0.9705\n",
            "Epoch 78/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0620 - accuracy: 0.9831 - val_loss: 0.1043 - val_accuracy: 0.9702\n",
            "Epoch 79/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0635 - accuracy: 0.9829 - val_loss: 0.1039 - val_accuracy: 0.9703\n",
            "Epoch 80/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0604 - accuracy: 0.9837 - val_loss: 0.1043 - val_accuracy: 0.9699\n",
            "Epoch 81/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0628 - accuracy: 0.9828 - val_loss: 0.1038 - val_accuracy: 0.9703\n",
            "Epoch 82/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0594 - accuracy: 0.9839 - val_loss: 0.1036 - val_accuracy: 0.9711\n",
            "Epoch 83/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0582 - accuracy: 0.9840 - val_loss: 0.1024 - val_accuracy: 0.9709\n",
            "Epoch 84/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0585 - accuracy: 0.9844 - val_loss: 0.1025 - val_accuracy: 0.9707\n",
            "Epoch 85/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0583 - accuracy: 0.9844 - val_loss: 0.1022 - val_accuracy: 0.9702\n",
            "Epoch 86/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0561 - accuracy: 0.9852 - val_loss: 0.1026 - val_accuracy: 0.9712\n",
            "Epoch 87/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0572 - accuracy: 0.9848 - val_loss: 0.1011 - val_accuracy: 0.9706\n",
            "Epoch 88/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0573 - accuracy: 0.9842 - val_loss: 0.1014 - val_accuracy: 0.9707\n",
            "Epoch 89/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0553 - accuracy: 0.9854 - val_loss: 0.1005 - val_accuracy: 0.9706\n",
            "Epoch 90/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0532 - accuracy: 0.9863 - val_loss: 0.1000 - val_accuracy: 0.9710\n",
            "Epoch 91/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0557 - accuracy: 0.9852 - val_loss: 0.0995 - val_accuracy: 0.9708\n",
            "Epoch 92/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0521 - accuracy: 0.9859 - val_loss: 0.0995 - val_accuracy: 0.9718\n",
            "Epoch 93/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0518 - accuracy: 0.9865 - val_loss: 0.0994 - val_accuracy: 0.9715\n",
            "Epoch 94/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0520 - accuracy: 0.9867 - val_loss: 0.0986 - val_accuracy: 0.9718\n",
            "Epoch 95/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0506 - accuracy: 0.9871 - val_loss: 0.0990 - val_accuracy: 0.9717\n",
            "Epoch 96/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0474 - accuracy: 0.9875 - val_loss: 0.0985 - val_accuracy: 0.9715\n",
            "Epoch 97/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0475 - accuracy: 0.9872 - val_loss: 0.0974 - val_accuracy: 0.9717\n",
            "Epoch 98/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0491 - accuracy: 0.9875 - val_loss: 0.0976 - val_accuracy: 0.9718\n",
            "Epoch 99/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0476 - accuracy: 0.9872 - val_loss: 0.0987 - val_accuracy: 0.9715\n",
            "Epoch 100/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0463 - accuracy: 0.9881 - val_loss: 0.0966 - val_accuracy: 0.9722\n",
            "Epoch 101/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0474 - accuracy: 0.9880 - val_loss: 0.0967 - val_accuracy: 0.9715\n",
            "Epoch 102/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0450 - accuracy: 0.9883 - val_loss: 0.0973 - val_accuracy: 0.9723\n",
            "Epoch 103/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0451 - accuracy: 0.9879 - val_loss: 0.0967 - val_accuracy: 0.9719\n",
            "Epoch 104/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0454 - accuracy: 0.9887 - val_loss: 0.0961 - val_accuracy: 0.9722\n",
            "Epoch 105/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0434 - accuracy: 0.9886 - val_loss: 0.0960 - val_accuracy: 0.9722\n",
            "Epoch 106/200\n",
            "375/375 [==============================] - 2s 7ms/step - loss: 0.0422 - accuracy: 0.9895 - val_loss: 0.0962 - val_accuracy: 0.9723\n",
            "Epoch 107/200\n",
            "375/375 [==============================] - 3s 7ms/step - loss: 0.0445 - accuracy: 0.9884 - val_loss: 0.0959 - val_accuracy: 0.9717\n",
            "Epoch 108/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0438 - accuracy: 0.9883 - val_loss: 0.0961 - val_accuracy: 0.9721\n",
            "Epoch 109/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.0425 - accuracy: 0.9888 - val_loss: 0.0955 - val_accuracy: 0.9725\n",
            "Epoch 110/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0429 - accuracy: 0.9894 - val_loss: 0.0963 - val_accuracy: 0.9721\n",
            "Epoch 111/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0412 - accuracy: 0.9895 - val_loss: 0.0949 - val_accuracy: 0.9725\n",
            "Epoch 112/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0410 - accuracy: 0.9893 - val_loss: 0.0956 - val_accuracy: 0.9728\n",
            "Epoch 113/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0406 - accuracy: 0.9892 - val_loss: 0.0958 - val_accuracy: 0.9721\n",
            "Epoch 114/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0398 - accuracy: 0.9903 - val_loss: 0.0959 - val_accuracy: 0.9722\n",
            "Epoch 115/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0389 - accuracy: 0.9899 - val_loss: 0.0943 - val_accuracy: 0.9732\n",
            "Epoch 116/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0384 - accuracy: 0.9906 - val_loss: 0.0955 - val_accuracy: 0.9732\n",
            "Epoch 117/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0386 - accuracy: 0.9902 - val_loss: 0.0947 - val_accuracy: 0.9726\n",
            "Epoch 118/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0380 - accuracy: 0.9906 - val_loss: 0.0953 - val_accuracy: 0.9723\n",
            "Epoch 119/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0400 - accuracy: 0.9904 - val_loss: 0.0943 - val_accuracy: 0.9732\n",
            "Epoch 120/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0358 - accuracy: 0.9914 - val_loss: 0.0948 - val_accuracy: 0.9729\n",
            "Epoch 121/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0361 - accuracy: 0.9918 - val_loss: 0.0946 - val_accuracy: 0.9721\n",
            "Epoch 122/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0363 - accuracy: 0.9914 - val_loss: 0.0948 - val_accuracy: 0.9732\n",
            "Epoch 123/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0346 - accuracy: 0.9912 - val_loss: 0.0950 - val_accuracy: 0.9730\n",
            "Epoch 124/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0334 - accuracy: 0.9923 - val_loss: 0.0948 - val_accuracy: 0.9735\n",
            "Epoch 125/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0363 - accuracy: 0.9918 - val_loss: 0.0941 - val_accuracy: 0.9729\n",
            "Epoch 126/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0316 - accuracy: 0.9931 - val_loss: 0.0944 - val_accuracy: 0.9726\n",
            "Epoch 127/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0318 - accuracy: 0.9926 - val_loss: 0.0946 - val_accuracy: 0.9736\n",
            "Epoch 128/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0327 - accuracy: 0.9921 - val_loss: 0.0935 - val_accuracy: 0.9739\n",
            "Epoch 129/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0347 - accuracy: 0.9920 - val_loss: 0.0948 - val_accuracy: 0.9726\n",
            "Epoch 130/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0338 - accuracy: 0.9916 - val_loss: 0.0939 - val_accuracy: 0.9729\n",
            "Epoch 131/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0341 - accuracy: 0.9919 - val_loss: 0.0941 - val_accuracy: 0.9732\n",
            "Epoch 132/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0314 - accuracy: 0.9929 - val_loss: 0.0944 - val_accuracy: 0.9726\n",
            "Epoch 133/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0307 - accuracy: 0.9927 - val_loss: 0.0940 - val_accuracy: 0.9741\n",
            "Epoch 134/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0320 - accuracy: 0.9929 - val_loss: 0.0936 - val_accuracy: 0.9728\n",
            "Epoch 135/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0318 - accuracy: 0.9924 - val_loss: 0.0938 - val_accuracy: 0.9732\n",
            "Epoch 136/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0294 - accuracy: 0.9937 - val_loss: 0.0943 - val_accuracy: 0.9731\n",
            "Epoch 137/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0302 - accuracy: 0.9935 - val_loss: 0.0939 - val_accuracy: 0.9736\n",
            "Epoch 138/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0295 - accuracy: 0.9935 - val_loss: 0.0939 - val_accuracy: 0.9734\n",
            "Epoch 139/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0292 - accuracy: 0.9935 - val_loss: 0.0944 - val_accuracy: 0.9736\n",
            "Epoch 140/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0279 - accuracy: 0.9939 - val_loss: 0.0934 - val_accuracy: 0.9736\n",
            "Epoch 141/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0281 - accuracy: 0.9936 - val_loss: 0.0938 - val_accuracy: 0.9732\n",
            "Epoch 142/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0284 - accuracy: 0.9941 - val_loss: 0.0939 - val_accuracy: 0.9740\n",
            "Epoch 143/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0280 - accuracy: 0.9938 - val_loss: 0.0937 - val_accuracy: 0.9743\n",
            "Epoch 144/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0269 - accuracy: 0.9940 - val_loss: 0.0943 - val_accuracy: 0.9733\n",
            "Epoch 145/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0290 - accuracy: 0.9938 - val_loss: 0.0936 - val_accuracy: 0.9740\n",
            "Epoch 146/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0263 - accuracy: 0.9942 - val_loss: 0.0937 - val_accuracy: 0.9735\n",
            "Epoch 147/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0272 - accuracy: 0.9943 - val_loss: 0.0941 - val_accuracy: 0.9739\n",
            "Epoch 148/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0276 - accuracy: 0.9943 - val_loss: 0.0935 - val_accuracy: 0.9739\n",
            "Epoch 149/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0259 - accuracy: 0.9945 - val_loss: 0.0941 - val_accuracy: 0.9738\n",
            "Epoch 150/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0252 - accuracy: 0.9951 - val_loss: 0.0932 - val_accuracy: 0.9743\n",
            "Epoch 151/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0258 - accuracy: 0.9942 - val_loss: 0.0937 - val_accuracy: 0.9740\n",
            "Epoch 152/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0256 - accuracy: 0.9942 - val_loss: 0.0937 - val_accuracy: 0.9743\n",
            "Epoch 153/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0246 - accuracy: 0.9952 - val_loss: 0.0935 - val_accuracy: 0.9744\n",
            "Epoch 154/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0256 - accuracy: 0.9951 - val_loss: 0.0938 - val_accuracy: 0.9742\n",
            "Epoch 155/200\n",
            "375/375 [==============================] - 2s 6ms/step - loss: 0.0250 - accuracy: 0.9954 - val_loss: 0.0948 - val_accuracy: 0.9737\n",
            "Epoch 156/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0239 - accuracy: 0.9954 - val_loss: 0.0940 - val_accuracy: 0.9737\n",
            "Epoch 157/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0242 - accuracy: 0.9957 - val_loss: 0.0935 - val_accuracy: 0.9743\n",
            "Epoch 158/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0239 - accuracy: 0.9954 - val_loss: 0.0952 - val_accuracy: 0.9734\n",
            "Epoch 159/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0229 - accuracy: 0.9957 - val_loss: 0.0943 - val_accuracy: 0.9746\n",
            "Epoch 160/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0226 - accuracy: 0.9960 - val_loss: 0.0940 - val_accuracy: 0.9742\n",
            "Epoch 161/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0225 - accuracy: 0.9960 - val_loss: 0.0942 - val_accuracy: 0.9740\n",
            "Epoch 162/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0224 - accuracy: 0.9963 - val_loss: 0.0947 - val_accuracy: 0.9746\n",
            "Epoch 163/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0221 - accuracy: 0.9963 - val_loss: 0.0942 - val_accuracy: 0.9744\n",
            "Epoch 164/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0208 - accuracy: 0.9965 - val_loss: 0.0943 - val_accuracy: 0.9741\n",
            "Epoch 165/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0215 - accuracy: 0.9961 - val_loss: 0.0936 - val_accuracy: 0.9745\n",
            "Epoch 166/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0218 - accuracy: 0.9960 - val_loss: 0.0943 - val_accuracy: 0.9739\n",
            "Epoch 167/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.0206 - accuracy: 0.9963 - val_loss: 0.0946 - val_accuracy: 0.9740\n",
            "Epoch 168/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.0217 - accuracy: 0.9962 - val_loss: 0.0935 - val_accuracy: 0.9746\n",
            "Epoch 169/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0209 - accuracy: 0.9964 - val_loss: 0.0946 - val_accuracy: 0.9737\n",
            "Epoch 170/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0216 - accuracy: 0.9962 - val_loss: 0.0940 - val_accuracy: 0.9744\n",
            "Epoch 171/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.0197 - accuracy: 0.9968 - val_loss: 0.0941 - val_accuracy: 0.9747\n",
            "Epoch 172/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0192 - accuracy: 0.9967 - val_loss: 0.0941 - val_accuracy: 0.9747\n",
            "Epoch 173/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0196 - accuracy: 0.9969 - val_loss: 0.0942 - val_accuracy: 0.9747\n",
            "Epoch 174/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.0189 - accuracy: 0.9974 - val_loss: 0.0949 - val_accuracy: 0.9746\n",
            "Epoch 175/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0196 - accuracy: 0.9969 - val_loss: 0.0950 - val_accuracy: 0.9747\n",
            "Epoch 176/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.0190 - accuracy: 0.9972 - val_loss: 0.0947 - val_accuracy: 0.9747\n",
            "Epoch 177/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0182 - accuracy: 0.9971 - val_loss: 0.0946 - val_accuracy: 0.9745\n",
            "Epoch 178/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0194 - accuracy: 0.9967 - val_loss: 0.0947 - val_accuracy: 0.9746\n",
            "Epoch 179/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0187 - accuracy: 0.9972 - val_loss: 0.0943 - val_accuracy: 0.9746\n",
            "Epoch 180/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0199 - accuracy: 0.9971 - val_loss: 0.0952 - val_accuracy: 0.9742\n",
            "Epoch 181/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0189 - accuracy: 0.9972 - val_loss: 0.0947 - val_accuracy: 0.9743\n",
            "Epoch 182/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0191 - accuracy: 0.9970 - val_loss: 0.0949 - val_accuracy: 0.9747\n",
            "Epoch 183/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0177 - accuracy: 0.9975 - val_loss: 0.0953 - val_accuracy: 0.9742\n",
            "Epoch 184/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0173 - accuracy: 0.9976 - val_loss: 0.0951 - val_accuracy: 0.9750\n",
            "Epoch 185/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0181 - accuracy: 0.9974 - val_loss: 0.0950 - val_accuracy: 0.9746\n",
            "Epoch 186/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0170 - accuracy: 0.9974 - val_loss: 0.0958 - val_accuracy: 0.9744\n",
            "Epoch 187/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0178 - accuracy: 0.9973 - val_loss: 0.0957 - val_accuracy: 0.9746\n",
            "Epoch 188/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0158 - accuracy: 0.9979 - val_loss: 0.0960 - val_accuracy: 0.9743\n",
            "Epoch 189/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0161 - accuracy: 0.9979 - val_loss: 0.0952 - val_accuracy: 0.9744\n",
            "Epoch 190/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0163 - accuracy: 0.9976 - val_loss: 0.0950 - val_accuracy: 0.9747\n",
            "Epoch 191/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0169 - accuracy: 0.9977 - val_loss: 0.0955 - val_accuracy: 0.9750\n",
            "Epoch 192/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0155 - accuracy: 0.9981 - val_loss: 0.0963 - val_accuracy: 0.9749\n",
            "Epoch 193/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0154 - accuracy: 0.9981 - val_loss: 0.0959 - val_accuracy: 0.9743\n",
            "Epoch 194/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.0161 - accuracy: 0.9975 - val_loss: 0.0962 - val_accuracy: 0.9743\n",
            "Epoch 195/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0163 - accuracy: 0.9980 - val_loss: 0.0967 - val_accuracy: 0.9743\n",
            "Epoch 196/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0152 - accuracy: 0.9980 - val_loss: 0.0964 - val_accuracy: 0.9743\n",
            "Epoch 197/200\n",
            "375/375 [==============================] - 2s 5ms/step - loss: 0.0157 - accuracy: 0.9977 - val_loss: 0.0960 - val_accuracy: 0.9747\n",
            "Epoch 198/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.0159 - accuracy: 0.9982 - val_loss: 0.0960 - val_accuracy: 0.9747\n",
            "Epoch 199/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.0141 - accuracy: 0.9983 - val_loss: 0.0958 - val_accuracy: 0.9747\n",
            "Epoch 200/200\n",
            "375/375 [==============================] - 2s 4ms/step - loss: 0.0156 - accuracy: 0.9978 - val_loss: 0.0969 - val_accuracy: 0.9745\n",
            "313/313 [==============================] - 0s 1ms/step - loss: 0.0829 - accuracy: 0.9749\n",
            "Test Loss : 0.08294537663459778  Test Accuracy: 0.9749000072479248\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_0Z5izuQwXF8"
      },
      "source": [
        "### Model with Dropout"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z-cu-HnEupHp",
        "outputId": "d4c1ecce-a88c-4ba6-d36b-75f3f7f6475e"
      },
      "source": [
        "N_HIDDEN = 128\n",
        "DROPOUT = 0.3\n",
        "EPOCH = 50\n",
        "BATCH_SIZE = 512 # We are using TPU. we can increase the Batch size\n",
        "model = tf.keras.models.Sequential()\n",
        "model.add(keras.layers.Dense(N_HIDDEN,\n",
        "                             input_shape=(RESHAPED,),\n",
        "                             name='dense_layer_1',\n",
        "                             activation='relu'))\n",
        "model.add(keras.layers.Dropout(DROPOUT))\n",
        "model.add(keras.layers.Dense(N_HIDDEN,\n",
        "                             name='dense_layer_2',\n",
        "                             activation='relu'))\n",
        "model.add(keras.layers.Dropout(DROPOUT))\n",
        "#final layer\n",
        "model.add(keras.layers.Dense(NB_CLASSES,\n",
        "                             name='dense_layer_final',\n",
        "                             activation='softmax'))\n",
        "model.summary()\n",
        "model.compile(optimizer='SGD',\n",
        "              loss='categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n",
        "model.fit(X_train,y_train,\n",
        "          batch_size=BATCH_SIZE,\n",
        "          epochs=EPOCH,\n",
        "          verbose=VERBOSE,\n",
        "          validation_split=VALIDATION_SPLIT)\n",
        "test_loss, test_accuracy = model.evaluate(X_test,y_test)\n",
        "print('Test Loss :', test_loss, ' Test Accuracy:', test_accuracy)"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_9\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "dense_layer_1 (Dense)        (None, 128)               100480    \n",
            "_________________________________________________________________\n",
            "dropout_2 (Dropout)          (None, 128)               0         \n",
            "_________________________________________________________________\n",
            "dense_layer_2 (Dense)        (None, 128)               16512     \n",
            "_________________________________________________________________\n",
            "dropout_3 (Dropout)          (None, 128)               0         \n",
            "_________________________________________________________________\n",
            "dense_layer_final (Dense)    (None, 10)                1290      \n",
            "=================================================================\n",
            "Total params: 118,282\n",
            "Trainable params: 118,282\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/50\n",
            "94/94 [==============================] - 3s 21ms/step - loss: 2.3091 - accuracy: 0.1376 - val_loss: 1.9770 - val_accuracy: 0.5318\n",
            "Epoch 2/50\n",
            "94/94 [==============================] - 2s 17ms/step - loss: 1.9638 - accuracy: 0.3823 - val_loss: 1.5564 - val_accuracy: 0.6987\n",
            "Epoch 3/50\n",
            "94/94 [==============================] - 2s 18ms/step - loss: 1.6068 - accuracy: 0.5339 - val_loss: 1.1649 - val_accuracy: 0.7660\n",
            "Epoch 4/50\n",
            "94/94 [==============================] - 1s 14ms/step - loss: 1.3108 - accuracy: 0.6130 - val_loss: 0.9064 - val_accuracy: 0.8034\n",
            "Epoch 5/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 1.1061 - accuracy: 0.6609 - val_loss: 0.7508 - val_accuracy: 0.8240\n",
            "Epoch 6/50\n",
            "94/94 [==============================] - 1s 14ms/step - loss: 0.9790 - accuracy: 0.6962 - val_loss: 0.6546 - val_accuracy: 0.8402\n",
            "Epoch 7/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.8850 - accuracy: 0.7246 - val_loss: 0.5879 - val_accuracy: 0.8503\n",
            "Epoch 8/50\n",
            "94/94 [==============================] - 2s 17ms/step - loss: 0.8223 - accuracy: 0.7429 - val_loss: 0.5391 - val_accuracy: 0.8627\n",
            "Epoch 9/50\n",
            "94/94 [==============================] - 1s 15ms/step - loss: 0.7688 - accuracy: 0.7591 - val_loss: 0.5022 - val_accuracy: 0.8690\n",
            "Epoch 10/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.7202 - accuracy: 0.7771 - val_loss: 0.4731 - val_accuracy: 0.8758\n",
            "Epoch 11/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.6970 - accuracy: 0.7863 - val_loss: 0.4489 - val_accuracy: 0.8800\n",
            "Epoch 12/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.6561 - accuracy: 0.7992 - val_loss: 0.4291 - val_accuracy: 0.8842\n",
            "Epoch 13/50\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.6354 - accuracy: 0.8070 - val_loss: 0.4127 - val_accuracy: 0.8885\n",
            "Epoch 14/50\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.6077 - accuracy: 0.8138 - val_loss: 0.3982 - val_accuracy: 0.8917\n",
            "Epoch 15/50\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.5867 - accuracy: 0.8212 - val_loss: 0.3848 - val_accuracy: 0.8952\n",
            "Epoch 16/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.5694 - accuracy: 0.8266 - val_loss: 0.3736 - val_accuracy: 0.8982\n",
            "Epoch 17/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.5559 - accuracy: 0.8322 - val_loss: 0.3636 - val_accuracy: 0.8998\n",
            "Epoch 18/50\n",
            "94/94 [==============================] - 1s 14ms/step - loss: 0.5379 - accuracy: 0.8362 - val_loss: 0.3546 - val_accuracy: 0.9028\n",
            "Epoch 19/50\n",
            "94/94 [==============================] - 1s 14ms/step - loss: 0.5292 - accuracy: 0.8386 - val_loss: 0.3470 - val_accuracy: 0.9047\n",
            "Epoch 20/50\n",
            "94/94 [==============================] - 1s 15ms/step - loss: 0.5125 - accuracy: 0.8439 - val_loss: 0.3395 - val_accuracy: 0.9064\n",
            "Epoch 21/50\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.5028 - accuracy: 0.8500 - val_loss: 0.3332 - val_accuracy: 0.9085\n",
            "Epoch 22/50\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.4967 - accuracy: 0.8504 - val_loss: 0.3258 - val_accuracy: 0.9101\n",
            "Epoch 23/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.4900 - accuracy: 0.8545 - val_loss: 0.3200 - val_accuracy: 0.9122\n",
            "Epoch 24/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.4804 - accuracy: 0.8594 - val_loss: 0.3156 - val_accuracy: 0.9133\n",
            "Epoch 25/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.4790 - accuracy: 0.8591 - val_loss: 0.3100 - val_accuracy: 0.9144\n",
            "Epoch 26/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.4619 - accuracy: 0.8637 - val_loss: 0.3052 - val_accuracy: 0.9151\n",
            "Epoch 27/50\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.4536 - accuracy: 0.8667 - val_loss: 0.3006 - val_accuracy: 0.9160\n",
            "Epoch 28/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.4475 - accuracy: 0.8669 - val_loss: 0.2966 - val_accuracy: 0.9165\n",
            "Epoch 29/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.4384 - accuracy: 0.8700 - val_loss: 0.2925 - val_accuracy: 0.9179\n",
            "Epoch 30/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.4319 - accuracy: 0.8720 - val_loss: 0.2886 - val_accuracy: 0.9183\n",
            "Epoch 31/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.4257 - accuracy: 0.8743 - val_loss: 0.2852 - val_accuracy: 0.9186\n",
            "Epoch 32/50\n",
            "94/94 [==============================] - 1s 14ms/step - loss: 0.4282 - accuracy: 0.8731 - val_loss: 0.2813 - val_accuracy: 0.9199\n",
            "Epoch 33/50\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.4167 - accuracy: 0.8767 - val_loss: 0.2780 - val_accuracy: 0.9201\n",
            "Epoch 34/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.4236 - accuracy: 0.8764 - val_loss: 0.2752 - val_accuracy: 0.9203\n",
            "Epoch 35/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.4099 - accuracy: 0.8791 - val_loss: 0.2716 - val_accuracy: 0.9209\n",
            "Epoch 36/50\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.4035 - accuracy: 0.8787 - val_loss: 0.2692 - val_accuracy: 0.9222\n",
            "Epoch 37/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.3944 - accuracy: 0.8841 - val_loss: 0.2660 - val_accuracy: 0.9232\n",
            "Epoch 38/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.3939 - accuracy: 0.8861 - val_loss: 0.2633 - val_accuracy: 0.9235\n",
            "Epoch 39/50\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.3913 - accuracy: 0.8861 - val_loss: 0.2603 - val_accuracy: 0.9242\n",
            "Epoch 40/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.3880 - accuracy: 0.8857 - val_loss: 0.2579 - val_accuracy: 0.9247\n",
            "Epoch 41/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.3801 - accuracy: 0.8878 - val_loss: 0.2559 - val_accuracy: 0.9248\n",
            "Epoch 42/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.3782 - accuracy: 0.8881 - val_loss: 0.2527 - val_accuracy: 0.9263\n",
            "Epoch 43/50\n",
            "94/94 [==============================] - 1s 14ms/step - loss: 0.3826 - accuracy: 0.8883 - val_loss: 0.2507 - val_accuracy: 0.9260\n",
            "Epoch 44/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.3774 - accuracy: 0.8886 - val_loss: 0.2481 - val_accuracy: 0.9267\n",
            "Epoch 45/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.3638 - accuracy: 0.8897 - val_loss: 0.2458 - val_accuracy: 0.9266\n",
            "Epoch 46/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.3633 - accuracy: 0.8921 - val_loss: 0.2436 - val_accuracy: 0.9285\n",
            "Epoch 47/50\n",
            "94/94 [==============================] - 1s 14ms/step - loss: 0.3582 - accuracy: 0.8964 - val_loss: 0.2415 - val_accuracy: 0.9283\n",
            "Epoch 48/50\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.3503 - accuracy: 0.8964 - val_loss: 0.2394 - val_accuracy: 0.9288\n",
            "Epoch 49/50\n",
            "94/94 [==============================] - 1s 15ms/step - loss: 0.3572 - accuracy: 0.8948 - val_loss: 0.2372 - val_accuracy: 0.9305\n",
            "Epoch 50/50\n",
            "94/94 [==============================] - 1s 15ms/step - loss: 0.3459 - accuracy: 0.8986 - val_loss: 0.2356 - val_accuracy: 0.9308\n",
            "313/313 [==============================] - 1s 2ms/step - loss: 0.2376 - accuracy: 0.9305\n",
            "Test Loss : 0.2375728040933609  Test Accuracy: 0.9304999709129333\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pV74tqCZyPBz"
      },
      "source": [
        "### Optimizer RMSProp"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VIx8MDqnw08l",
        "outputId": "64955de0-2324-43c3-9d16-2b0cf13f634d"
      },
      "source": [
        "model.compile(optimizer='RMSProp',\n",
        "              loss='categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n",
        "model.fit(X_train,y_train,\n",
        "          batch_size=BATCH_SIZE,\n",
        "          epochs=10,\n",
        "          verbose=VERBOSE,\n",
        "          validation_split=VALIDATION_SPLIT)\n",
        "test_loss, test_accuracy = model.evaluate(X_test,y_test)\n",
        "print('Test Loss :', test_loss, ' Test Accuracy:', test_accuracy)"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "94/94 [==============================] - 2s 17ms/step - loss: 0.3362 - accuracy: 0.8992 - val_loss: 0.1703 - val_accuracy: 0.9504\n",
            "Epoch 2/10\n",
            "94/94 [==============================] - 1s 14ms/step - loss: 0.2388 - accuracy: 0.9294 - val_loss: 0.1480 - val_accuracy: 0.9563\n",
            "Epoch 3/10\n",
            "94/94 [==============================] - 1s 15ms/step - loss: 0.2027 - accuracy: 0.9407 - val_loss: 0.1243 - val_accuracy: 0.9622\n",
            "Epoch 4/10\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.1710 - accuracy: 0.9489 - val_loss: 0.1104 - val_accuracy: 0.9683\n",
            "Epoch 5/10\n",
            "94/94 [==============================] - 1s 14ms/step - loss: 0.1463 - accuracy: 0.9555 - val_loss: 0.1031 - val_accuracy: 0.9697\n",
            "Epoch 6/10\n",
            "94/94 [==============================] - 1s 14ms/step - loss: 0.1294 - accuracy: 0.9610 - val_loss: 0.0982 - val_accuracy: 0.9703\n",
            "Epoch 7/10\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.1162 - accuracy: 0.9638 - val_loss: 0.0949 - val_accuracy: 0.9725\n",
            "Epoch 8/10\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.1112 - accuracy: 0.9657 - val_loss: 0.0892 - val_accuracy: 0.9743\n",
            "Epoch 9/10\n",
            "94/94 [==============================] - 1s 14ms/step - loss: 0.1028 - accuracy: 0.9686 - val_loss: 0.0880 - val_accuracy: 0.9748\n",
            "Epoch 10/10\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.0957 - accuracy: 0.9700 - val_loss: 0.0846 - val_accuracy: 0.9747\n",
            "313/313 [==============================] - 0s 1ms/step - loss: 0.0777 - accuracy: 0.9765\n",
            "Test Loss : 0.0776589959859848  Test Accuracy: 0.9764999747276306\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KCZkWBnRyj87"
      },
      "source": [
        "### Optimizer ADAM"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RfnCv9aZydva",
        "outputId": "c96219f5-be00-4cad-c6b7-3691853bf9cb"
      },
      "source": [
        "model.compile(optimizer='Adam',\n",
        "              loss='categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n",
        "model.fit(X_train,y_train,\n",
        "          batch_size=BATCH_SIZE,\n",
        "          epochs=10,\n",
        "          verbose=VERBOSE,\n",
        "          validation_split=VALIDATION_SPLIT)\n",
        "test_loss, test_accuracy = model.evaluate(X_test,y_test)\n",
        "print('Test Loss :', test_loss, ' Test Accuracy:', test_accuracy)"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "94/94 [==============================] - 2s 14ms/step - loss: 0.0893 - accuracy: 0.9717 - val_loss: 0.0849 - val_accuracy: 0.9765\n",
            "Epoch 2/10\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.0831 - accuracy: 0.9747 - val_loss: 0.0806 - val_accuracy: 0.9767\n",
            "Epoch 3/10\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.0820 - accuracy: 0.9741 - val_loss: 0.0839 - val_accuracy: 0.9757\n",
            "Epoch 4/10\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.0713 - accuracy: 0.9779 - val_loss: 0.0785 - val_accuracy: 0.9774\n",
            "Epoch 5/10\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.0725 - accuracy: 0.9762 - val_loss: 0.0795 - val_accuracy: 0.9772\n",
            "Epoch 6/10\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.0677 - accuracy: 0.9778 - val_loss: 0.0796 - val_accuracy: 0.9772\n",
            "Epoch 7/10\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.0662 - accuracy: 0.9786 - val_loss: 0.0771 - val_accuracy: 0.9778\n",
            "Epoch 8/10\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.0631 - accuracy: 0.9801 - val_loss: 0.0787 - val_accuracy: 0.9781\n",
            "Epoch 9/10\n",
            "94/94 [==============================] - 1s 13ms/step - loss: 0.0577 - accuracy: 0.9815 - val_loss: 0.0787 - val_accuracy: 0.9773\n",
            "Epoch 10/10\n",
            "94/94 [==============================] - 1s 12ms/step - loss: 0.0561 - accuracy: 0.9819 - val_loss: 0.0805 - val_accuracy: 0.9774\n",
            "313/313 [==============================] - 0s 1ms/step - loss: 0.0753 - accuracy: 0.9784\n",
            "Test Loss : 0.07528786361217499  Test Accuracy: 0.9783999919891357\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wOdTONd40_G0"
      },
      "source": [
        "# Sentiment Analysis"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fNr0Rr60yq37",
        "outputId": "639d5c8f-d7fa-46ed-a45c-afc36a13809c"
      },
      "source": [
        "from tensorflow.keras import datasets, layers, models, preprocessing\n",
        "import tensorflow_datasets as tfds\n",
        "\n",
        "\n",
        "MAX_LEN = 200\n",
        "N_WORDS = 10000\n",
        "DIM_EMBEDDING = 256\n",
        "EPOCH = 20\n",
        "BATCH_SIZE = 500\n",
        "\n",
        "def load_data():\n",
        "  (X_train, y_train), (X_test, y_test) = datasets.imdb.load_data(num_words=N_WORDS)\n",
        "  # pad with 0 till max_len\n",
        "  X_train = preprocessing.sequence.pad_sequences(X_train, maxlen=MAX_LEN)\n",
        "  X_test = preprocessing.sequence.pad_sequences(X_test, maxlen=MAX_LEN)\n",
        "  return (X_train, y_train), (X_test, y_test)\n",
        "\n",
        "def build_model():\n",
        "  model = models.Sequential()\n",
        "  model.add(layers.Embedding(\n",
        "      N_WORDS, DIM_EMBEDDING, input_length=MAX_LEN\n",
        "  ))\n",
        "  model.add(layers.Dropout(0.3))\n",
        "  # Takes the maximum value of either feature vector from each of     # the n_words features.\n",
        "  model.add(layers.GlobalAveragePooling1D())\n",
        "  model.add(layers.Dense(128, activation='relu'))\n",
        "  model.add(layers.Dropout(0.5))\n",
        "  model.add(layers.Dense(1, activation='sigmoid'))\n",
        "  return model\n",
        "\n",
        "(X_train, y_train), (X_test, y_test) = load_data()\n",
        "model = build_model()\n",
        "model.summary()\n",
        "model.compile(optimizer=\"adam\", loss=\"binary_crossentropy\", metrics=[\"accuracy\"])\n",
        "score = model.fit(X_train, y_train,\n",
        "                  epochs= EPOCH,\n",
        "                  batch_size = BATCH_SIZE,\n",
        "                  #validation_data = (X_test,y_test)\n",
        "                  validation_split=VALIDATION_SPLIT\n",
        "                  )\n",
        "score = model.evaluate(X_test, y_test, batch_size=BATCH_SIZE)\n",
        "print(\"\\nTest Score: \", score[0])\n",
        "print(\"Test accuracy: \", score[1])"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<string>:6: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
            "/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/datasets/imdb.py:159: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
            "  x_train, y_train = np.array(xs[:idx]), np.array(labels[:idx])\n",
            "/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/datasets/imdb.py:160: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
            "  x_test, y_test = np.array(xs[idx:]), np.array(labels[idx:])\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_14\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding_4 (Embedding)      (None, 200, 256)          2560000   \n",
            "_________________________________________________________________\n",
            "dropout_6 (Dropout)          (None, 200, 256)          0         \n",
            "_________________________________________________________________\n",
            "global_average_pooling1d_4 ( (None, 256)               0         \n",
            "_________________________________________________________________\n",
            "dense_12 (Dense)             (None, 128)               32896     \n",
            "_________________________________________________________________\n",
            "dropout_7 (Dropout)          (None, 128)               0         \n",
            "_________________________________________________________________\n",
            "dense_13 (Dense)             (None, 1)                 129       \n",
            "=================================================================\n",
            "Total params: 2,593,025\n",
            "Trainable params: 2,593,025\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/20\n",
            "40/40 [==============================] - 18s 442ms/step - loss: 0.6880 - accuracy: 0.5695 - val_loss: 0.6346 - val_accuracy: 0.7538\n",
            "Epoch 2/20\n",
            "40/40 [==============================] - 18s 447ms/step - loss: 0.5855 - accuracy: 0.7785 - val_loss: 0.4199 - val_accuracy: 0.8368\n",
            "Epoch 3/20\n",
            "40/40 [==============================] - 18s 449ms/step - loss: 0.3724 - accuracy: 0.8598 - val_loss: 0.3244 - val_accuracy: 0.8670\n",
            "Epoch 4/20\n",
            "40/40 [==============================] - 17s 435ms/step - loss: 0.2749 - accuracy: 0.8940 - val_loss: 0.2976 - val_accuracy: 0.8780\n",
            "Epoch 5/20\n",
            "40/40 [==============================] - 16s 396ms/step - loss: 0.2179 - accuracy: 0.9191 - val_loss: 0.2911 - val_accuracy: 0.8804\n",
            "Epoch 6/20\n",
            "40/40 [==============================] - 16s 400ms/step - loss: 0.1822 - accuracy: 0.9347 - val_loss: 0.2917 - val_accuracy: 0.8814\n",
            "Epoch 7/20\n",
            "40/40 [==============================] - 15s 373ms/step - loss: 0.1612 - accuracy: 0.9443 - val_loss: 0.2966 - val_accuracy: 0.8808\n",
            "Epoch 8/20\n",
            "40/40 [==============================] - 15s 372ms/step - loss: 0.1312 - accuracy: 0.9573 - val_loss: 0.3083 - val_accuracy: 0.8812\n",
            "Epoch 9/20\n",
            "40/40 [==============================] - 15s 367ms/step - loss: 0.1186 - accuracy: 0.9636 - val_loss: 0.3242 - val_accuracy: 0.8790\n",
            "Epoch 10/20\n",
            "40/40 [==============================] - 15s 363ms/step - loss: 0.1045 - accuracy: 0.9686 - val_loss: 0.3394 - val_accuracy: 0.8772\n",
            "Epoch 11/20\n",
            "40/40 [==============================] - 15s 377ms/step - loss: 0.0864 - accuracy: 0.9750 - val_loss: 0.3542 - val_accuracy: 0.8744\n",
            "Epoch 12/20\n",
            "40/40 [==============================] - 15s 370ms/step - loss: 0.0783 - accuracy: 0.9799 - val_loss: 0.3726 - val_accuracy: 0.8742\n",
            "Epoch 13/20\n",
            "40/40 [==============================] - 15s 374ms/step - loss: 0.0720 - accuracy: 0.9815 - val_loss: 0.3927 - val_accuracy: 0.8732\n",
            "Epoch 14/20\n",
            "40/40 [==============================] - 15s 371ms/step - loss: 0.0581 - accuracy: 0.9866 - val_loss: 0.4181 - val_accuracy: 0.8704\n",
            "Epoch 15/20\n",
            "40/40 [==============================] - 15s 374ms/step - loss: 0.0526 - accuracy: 0.9890 - val_loss: 0.4339 - val_accuracy: 0.8694\n",
            "Epoch 16/20\n",
            "40/40 [==============================] - 14s 359ms/step - loss: 0.0453 - accuracy: 0.9911 - val_loss: 0.4571 - val_accuracy: 0.8678\n",
            "Epoch 17/20\n",
            "40/40 [==============================] - 14s 355ms/step - loss: 0.0398 - accuracy: 0.9934 - val_loss: 0.4843 - val_accuracy: 0.8654\n",
            "Epoch 18/20\n",
            "40/40 [==============================] - 15s 364ms/step - loss: 0.0378 - accuracy: 0.9941 - val_loss: 0.5063 - val_accuracy: 0.8648\n",
            "Epoch 19/20\n",
            "40/40 [==============================] - 15s 368ms/step - loss: 0.0300 - accuracy: 0.9955 - val_loss: 0.5294 - val_accuracy: 0.8646\n",
            "Epoch 20/20\n",
            "40/40 [==============================] - 14s 350ms/step - loss: 0.0280 - accuracy: 0.9963 - val_loss: 0.5529 - val_accuracy: 0.8656\n",
            "50/50 [==============================] - 2s 37ms/step - loss: 0.6278 - accuracy: 0.8437\n",
            "\n",
            "Test Score:  0.6278123259544373\n",
            "Test accuracy:  0.8436800241470337\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2cW9b_iC2GxT"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}