CIFAR_Dataset_With_Tensorflow.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.8"
    },
    "colab": {
      "name": "CIFAR Dataset With Tensorflow.ipynb",
      "provenance": [],
      "include_colab_link": true
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/aayushkubb/Deep_Learning_Tutorial/blob/master/CIFAR_Dataset_With_Tensorflow.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "k0VLKrycGnf7"
      },
      "source": [
        "### Import the Libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9KqrRsoPGnf-"
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "sns.set()\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Zqs2_OQJGngB"
      },
      "source": [
        "### Load the Dataset and Prepare testing and training data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Yx0MAiHIGngC",
        "outputId": "d6be3e3f-1b79-4359-cd86-7f3a1203f4d4"
      },
      "source": [
        "from keras.datasets import cifar10"
      ],
      "execution_count": null,
      "outputs": [
        {
          "ename": "ModuleNotFoundError",
          "evalue": "No module named 'tensorflow'",
          "output_type": "error",
          "traceback": [
            "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[1;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
            "\u001b[1;32m<ipython-input-8-cc3044134c45>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[1;32mfrom\u001b[0m \u001b[0mkeras\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdatasets\u001b[0m \u001b[1;32mimport\u001b[0m \u001b[0mcifar10\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[1;32m~\\anaconda3\\lib\\site-packages\\keras\\__init__.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     19\u001b[0m \"\"\"\n\u001b[0;32m     20\u001b[0m \u001b[1;31m# pylint: disable=unused-import\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 21\u001b[1;33m \u001b[1;32mfrom\u001b[0m \u001b[0mtensorflow\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpython\u001b[0m \u001b[1;32mimport\u001b[0m \u001b[0mtf2\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     22\u001b[0m \u001b[1;32mfrom\u001b[0m \u001b[0mkeras\u001b[0m \u001b[1;32mimport\u001b[0m \u001b[0mdistribute\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     23\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
            "\u001b[1;31mModuleNotFoundError\u001b[0m: No module named 'tensorflow'"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2wmYQlcSGngE"
      },
      "source": [
        "(x_train,y_train),(x_test,y_test)=cifar10.load_data()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dc_7CrHzGngF"
      },
      "source": [
        "print(x_train.shape)\n",
        "print(x_test.shape)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "p-ZLu91JGngG"
      },
      "source": [
        "### Change the Datatype to Float and Normalize it"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KaDygvY5GngH"
      },
      "source": [
        "x_train=x_train.astype('float64')/255\n",
        "x_test=x_test.astype('float64')/255"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FU_UM6yjGngI"
      },
      "source": [
        "x_train.dtype\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NPvvfi6lGngJ"
      },
      "source": [
        "print(y_train.shape)\n",
        "print(y_test.shape)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LC_32anPGngJ"
      },
      "source": [
        "## Do the one hot Encoding of the Dependent Variable using Keras.utils"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xIDDy-r9GngK"
      },
      "source": [
        "from keras.utils import np_utils\n",
        "from keras.utils import to_categorical"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NIQ1kVULGngK"
      },
      "source": [
        "num_classes= len(np.unique(y_train))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CAT_jdvVGngL"
      },
      "source": [
        "num_classes"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TwFaShjtGngL"
      },
      "source": [
        "print(y_train)\n",
        "len(y_train)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VjSKoa9gGngM"
      },
      "source": [
        "y_train= to_categorical(y_train,num_classes)\n",
        "y_test=to_categorical(y_test,num_classes)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Lr8-oJyuGngM"
      },
      "source": [
        "### Divide the Training Dataset into  validation and Training"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "W_yuEdRzGngN"
      },
      "source": [
        "(y_train, y_valid) = y_train[5000:], y_train[:5000]\n",
        "(x_train, x_valid) = x_train[5000:], x_train[:5000]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "H-ysP0y-GngN",
        "outputId": "ea74eb75-bd86-40fc-8da2-19c0bc3c32d3"
      },
      "source": [
        "print(y_train.shape)\n",
        "print(x_train.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "ename": "NameError",
          "evalue": "name 'y_train' is not defined",
          "output_type": "error",
          "traceback": [
            "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[1;32m<ipython-input-3-ad7c0ec5537e>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0my_train\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx_train\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
            "\u001b[1;31mNameError\u001b[0m: name 'y_train' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fU7E0SjWGngN",
        "outputId": "62bf1397-51b6-4637-f72b-6c5824b207cb"
      },
      "source": [
        "print(y_valid.shape)\n",
        "print(x_valid.shape)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "ename": "NameError",
          "evalue": "name 'y_valid' is not defined",
          "output_type": "error",
          "traceback": [
            "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[1;32m<ipython-input-4-c8a1d3879a87>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0my_valid\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mx_valid\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
            "\u001b[1;31mNameError\u001b[0m: name 'y_valid' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Q-2AWsbHGngN"
      },
      "source": [
        "### Import the Deep Learning Libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mhWctTHPGngO"
      },
      "source": [
        "import tensorflow as tf\n",
        "  "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BGQQZm4kGngO",
        "outputId": "04b4e907-c6ee-4049-9812-540fdc4779f5"
      },
      "source": [
        "tf.__version__"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "'2.2.0'"
            ]
          },
          "execution_count": 17,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_uLASd8LGngO"
      },
      "source": [
        "### Build a CNN Architecture with Tensorflow"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xTAYSn28GngP"
      },
      "source": [
        "model= tf.keras.Sequential()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fWHR_4v_GngP"
      },
      "source": [
        "model.add(tf.keras.layers.Conv2D(filters=16,kernel_size=3,activation='relu',input_shape=(32,32,3)))\n",
        "model.add(tf.keras.layers.MaxPooling2D(pool_size=2))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9zAHbIOqGngP"
      },
      "source": [
        "model.add(tf.keras.layers.Conv2D(filters=16,kernel_size=3,activation='relu'))\n",
        "model.add(tf.keras.layers.MaxPooling2D(pool_size=2))\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wyner22TGngP"
      },
      "source": [
        "model.add(tf.keras.layers.Conv2D(filters=16,kernel_size=3,activation='relu'))\n",
        "model.add(tf.keras.layers.MaxPooling2D(pool_size=2))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HDkbHslSGngQ",
        "outputId": "00d84f38-52f7-45f6-c1ce-38c10d4fda82"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d (Conv2D)              (None, 30, 30, 16)        448       \n",
            "_________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D) (None, 15, 15, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_1 (Conv2D)            (None, 13, 13, 16)        2320      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 6, 6, 16)          0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 4, 4, 16)          2320      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2 (None, 2, 2, 16)          0         \n",
            "=================================================================\n",
            "Total params: 5,088\n",
            "Trainable params: 5,088\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oFPauRNLGngQ"
      },
      "source": [
        "model.add(tf.keras.layers.Flatten())"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "scrolled": true,
        "id": "eidqO-UUGngQ",
        "outputId": "dae3664c-32fd-403b-cbd8-0dd122ec638b"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d (Conv2D)              (None, 30, 30, 16)        448       \n",
            "_________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D) (None, 15, 15, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_1 (Conv2D)            (None, 13, 13, 16)        2320      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 6, 6, 16)          0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 4, 4, 16)          2320      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2 (None, 2, 2, 16)          0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 64)                0         \n",
            "=================================================================\n",
            "Total params: 5,088\n",
            "Trainable params: 5,088\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "y089Dj67GngR"
      },
      "source": [
        "### Stack an ANN on top of CNN"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FYVUY2bgGngS"
      },
      "source": [
        "model.add(tf.keras.layers.Dense(units=512,activation='relu',kernel_initializer='uniform'))\n",
        "model.add(tf.keras.layers.Dense(units=512,activation='relu',kernel_initializer='uniform'))\n",
        "model.add(tf.keras.layers.Dense(units=512,activation='relu',kernel_initializer='uniform'))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4oBc2YVfGngS"
      },
      "source": [
        "model.add(tf.keras.layers.Dense(units=512,activation='relu',kernel_initializer='uniform'))\n",
        "model.add(tf.keras.layers.Dense(units=512,activation='relu',kernel_initializer='uniform'))\n",
        "model.add(tf.keras.layers.Dense(units=512,activation='relu',kernel_initializer='uniform'))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2-4z7QZCGngS"
      },
      "source": [
        "model.add(tf.keras.layers.Dense(units=10,activation='softmax',kernel_initializer='uniform'))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EqQXnvqxGngS"
      },
      "source": [
        "model.compile(optimizer='adam',loss=\"categorical_crossentropy\",metrics=['accuracy'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VSbGryQkGngT",
        "outputId": "b20993d8-fa4d-4c7f-ee71-124c381911c9"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d (Conv2D)              (None, 30, 30, 16)        448       \n",
            "_________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D) (None, 15, 15, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_1 (Conv2D)            (None, 13, 13, 16)        2320      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 6, 6, 16)          0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 4, 4, 16)          2320      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2 (None, 2, 2, 16)          0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 64)                0         \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 512)               33280     \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 512)               262656    \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 512)               262656    \n",
            "_________________________________________________________________\n",
            "dense_3 (Dense)              (None, 512)               262656    \n",
            "_________________________________________________________________\n",
            "dense_4 (Dense)              (None, 512)               262656    \n",
            "_________________________________________________________________\n",
            "dense_5 (Dense)              (None, 512)               262656    \n",
            "_________________________________________________________________\n",
            "dense_6 (Dense)              (None, 10)                5130      \n",
            "=================================================================\n",
            "Total params: 1,356,778\n",
            "Trainable params: 1,356,778\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Zo_duTPBGngT"
      },
      "source": [
        "### Fit the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Kvq9nncaGngT",
        "outputId": "fe0969f5-3fbc-440c-dbf6-6d75f95d2318"
      },
      "source": [
        "answers=model.fit(x_train,y_train,validation_data=(x_valid,y_valid),shuffle=True,epochs=10,verbose=2,batch_size=500)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "90/90 - 16s - loss: 2.2747 - accuracy: 0.1198 - val_loss: 2.0853 - val_accuracy: 0.1982\n",
            "Epoch 2/10\n",
            "90/90 - 16s - loss: 1.9473 - accuracy: 0.2209 - val_loss: 1.8934 - val_accuracy: 0.2322\n",
            "Epoch 3/10\n",
            "90/90 - 14s - loss: 1.7968 - accuracy: 0.2859 - val_loss: 1.7837 - val_accuracy: 0.3082\n",
            "Epoch 4/10\n",
            "90/90 - 15s - loss: 1.6869 - accuracy: 0.3437 - val_loss: 1.6695 - val_accuracy: 0.3568\n",
            "Epoch 5/10\n",
            "90/90 - 17s - loss: 1.6057 - accuracy: 0.3858 - val_loss: 1.6137 - val_accuracy: 0.3914\n",
            "Epoch 6/10\n",
            "90/90 - 15s - loss: 1.5234 - accuracy: 0.4302 - val_loss: 1.5191 - val_accuracy: 0.4344\n",
            "Epoch 7/10\n",
            "90/90 - 15s - loss: 1.4559 - accuracy: 0.4619 - val_loss: 1.4247 - val_accuracy: 0.4778\n",
            "Epoch 8/10\n",
            "90/90 - 15s - loss: 1.4076 - accuracy: 0.4814 - val_loss: 1.4209 - val_accuracy: 0.4794\n",
            "Epoch 9/10\n",
            "90/90 - 15s - loss: 1.3547 - accuracy: 0.5045 - val_loss: 1.3543 - val_accuracy: 0.5132\n",
            "Epoch 10/10\n",
            "90/90 - 16s - loss: 1.3127 - accuracy: 0.5198 - val_loss: 1.3269 - val_accuracy: 0.5192\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mRlyB9T-GngU"
      },
      "source": [
        "### Evaluate the Results"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OdozmyaTGngU",
        "outputId": "ff654988-befd-4558-fd8c-3a854e4f7edf"
      },
      "source": [
        "results=model.evaluate(x_test,y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 1s 5ms/step - loss: 1.3413 - accuracy: 0.5075\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TrDpTWsqGngV",
        "outputId": "8f1d91c0-3451-4ac8-c8d0-3994fa44729f"
      },
      "source": [
        "results"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[1.3413255214691162, 0.5074999928474426]"
            ]
          },
          "execution_count": 32,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e8IlWNeyGngV",
        "outputId": "372a885b-2ce2-4ae1-ff34-b82522b5e80a"
      },
      "source": [
        "answers.history.keys()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])"
            ]
          },
          "execution_count": 33,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6m1ATwwzGngW",
        "outputId": "583a7c0d-52f4-4a04-98a5-b96c7939e95e"
      },
      "source": [
        "plt.plot(answers.history['val_loss'],label='Validation Loss');\n",
        "plt.plot(answers.history['loss'],label='Training Loss');\n",
        "plt.legend();"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ufiGZY_DGngX",
        "outputId": "2f9dc149-2cea-4bef-ed24-855647bb265f"
      },
      "source": [
        "plt.plot(answers.history['val_accuracy'],label='Validation Accuracy');\n",
        "plt.plot(answers.history['accuracy'],label='Training Accuracy');\n",
        "plt.legend;"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P6FKpF4EGngX"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}