cnn-image-recognition.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "cnn image recognition.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyN1hHTJFXxfpX/BEh0PGdYT",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/btkcodedev/ml-projects/blob/main/cnn-image-recognition.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Z_QjFpNE5i0j"
      },
      "source": [
        "- Tasks\n",
        " - Download the dataset and visualize the images\n",
        " - Change the label to one-hot encodings\n",
        " - Scale the image pixel values to take between 0 and 1"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oB7aTkkT5XOl"
      },
      "source": [
        "- CIFAR-10 dataset\n",
        "  - Images to be recognized: Tiny images of 32 * 32 pixels\n",
        "  - Labels: 10 possible labels (airplane, automobile, bird, cat, deer, dog, frog, horse, ship, and truck)\n",
        "  - Dataset size: 60000 images, split into 50000 for training and 10000 for testing"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PiFN6s7M5Kwr"
      },
      "source": [
        "from keras.datasets import cifar10"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cVaj99Gn5OP_",
        "outputId": "5529a0a3-be25-4e18-f770-3e41778f8022"
      },
      "source": [
        "(xtrain,ytrain),(xtest,ytest) = cifar10.load_data()"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz\n",
            "170500096/170498071 [==============================] - 2s 0us/step\n",
            "170508288/170498071 [==============================] - 2s 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NyDLYD9r592w",
        "outputId": "b339f1e7-e1c2-4b5c-ac81-895081d7f92a"
      },
      "source": [
        "print(xtrain.shape, ytrain.shape)\n",
        "# 50000 images\n",
        "# 32 pixels in height\n",
        "# 32 pixels in width\n",
        "# 3 pixels in depth (corresponding to Red, Green and Blue)\n",
        "\n",
        "# 50000 images\n",
        "# 1 label consist of 10 options total"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(50000, 32, 32, 3) (50000, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dme-SLL5-bv-",
        "outputId": "df9ee2d8-638e-451f-dfab-e15244f61c7f"
      },
      "source": [
        "print(xtest.shape, ytest.shape)"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(10000, 32, 32, 3) (10000, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "494vo1Bc6II1",
        "outputId": "407bc074-1ef2-4fea-b87e-3249b96f0162"
      },
      "source": [
        "xtrain[0] # Machine readable format of image"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[[ 59,  62,  63],\n",
              "        [ 43,  46,  45],\n",
              "        [ 50,  48,  43],\n",
              "        ...,\n",
              "        [158, 132, 108],\n",
              "        [152, 125, 102],\n",
              "        [148, 124, 103]],\n",
              "\n",
              "       [[ 16,  20,  20],\n",
              "        [  0,   0,   0],\n",
              "        [ 18,   8,   0],\n",
              "        ...,\n",
              "        [123,  88,  55],\n",
              "        [119,  83,  50],\n",
              "        [122,  87,  57]],\n",
              "\n",
              "       [[ 25,  24,  21],\n",
              "        [ 16,   7,   0],\n",
              "        [ 49,  27,   8],\n",
              "        ...,\n",
              "        [118,  84,  50],\n",
              "        [120,  84,  50],\n",
              "        [109,  73,  42]],\n",
              "\n",
              "       ...,\n",
              "\n",
              "       [[208, 170,  96],\n",
              "        [201, 153,  34],\n",
              "        [198, 161,  26],\n",
              "        ...,\n",
              "        [160, 133,  70],\n",
              "        [ 56,  31,   7],\n",
              "        [ 53,  34,  20]],\n",
              "\n",
              "       [[180, 139,  96],\n",
              "        [173, 123,  42],\n",
              "        [186, 144,  30],\n",
              "        ...,\n",
              "        [184, 148,  94],\n",
              "        [ 97,  62,  34],\n",
              "        [ 83,  53,  34]],\n",
              "\n",
              "       [[177, 144, 116],\n",
              "        [168, 129,  94],\n",
              "        [179, 142,  87],\n",
              "        ...,\n",
              "        [216, 184, 140],\n",
              "        [151, 118,  84],\n",
              "        [123,  92,  72]]], dtype=uint8)"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TJ4Ztaxe6r5_"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline \n",
        "# %matplotlib inline tells the notebook about wishing for the image to display within the image"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 336
        },
        "id": "_oMkTrPr637d",
        "outputId": "d9746121-507f-4ccd-ee68-55bf348a14e7"
      },
      "source": [
        "img = plt.imshow(xtrain[0])\n",
        "plt.xlabel(ytrain[0])\n",
        "plt.show()"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/matplotlib/text.py:1165: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
            "  if s != self._text:\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 266
        },
        "id": "xaTaqErQ7Uzt",
        "outputId": "3436883e-47fd-49d0-b8ca-737c9d84962e"
      },
      "source": [
        "img2 = plt.imshow(xtrain[1])"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5LJvh6HH7iwS"
      },
      "source": [
        "# Preprocessing of Labels\n",
        "\n",
        "# Usage of one-hot-encoding\n",
        "# we need probability of those 10 classes\n",
        "# For 10 class labels, we need 10 output neurons\n",
        "# So for matching each of 10 neurons we need to uniquely identify label\n",
        "import keras"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gwGsGC4W-G-v"
      },
      "source": [
        "from keras.utils import np_utils as npu\n",
        "ytrainonehot = npu.to_categorical(ytrain, 10)\n",
        "ytestonehot = npu.to_categorical(ytest, 10)"
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pOsw9b4K-jMv",
        "outputId": "7a4a5f44-ca9b-454d-f2e0-cdbe0b4eaf1f"
      },
      "source": [
        "print(ytrainonehot[0])  #Instead of 6 in ytrain[0] "
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0. 0. 0. 0. 0. 0. 1. 0. 0. 0.]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "arGbDpdn_KJR"
      },
      "source": [
        "#Preprocessing of images to make it around 0-1\n",
        "# As pixel already take the values between 0 and 255, we simply need to divide by 255.\n",
        "\n",
        "xtrain = xtrain.astype('float32') # for ensuring decimal points\n",
        "xtest = xtest.astype('float32')\n",
        "xtrain = xtrain/255\n",
        "xtest = xtest/255"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fK_4eH3UAuEw",
        "outputId": "a7e655b7-ec9b-49d7-eee1-84c2ca071cc4"
      },
      "source": [
        "xtrain[0]"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[[0.23137255, 0.24313726, 0.24705882],\n",
              "        [0.16862746, 0.18039216, 0.1764706 ],\n",
              "        [0.19607843, 0.1882353 , 0.16862746],\n",
              "        ...,\n",
              "        [0.61960787, 0.5176471 , 0.42352942],\n",
              "        [0.59607846, 0.49019608, 0.4       ],\n",
              "        [0.5803922 , 0.4862745 , 0.40392157]],\n",
              "\n",
              "       [[0.0627451 , 0.07843138, 0.07843138],\n",
              "        [0.        , 0.        , 0.        ],\n",
              "        [0.07058824, 0.03137255, 0.        ],\n",
              "        ...,\n",
              "        [0.48235294, 0.34509805, 0.21568628],\n",
              "        [0.46666667, 0.3254902 , 0.19607843],\n",
              "        [0.47843137, 0.34117648, 0.22352941]],\n",
              "\n",
              "       [[0.09803922, 0.09411765, 0.08235294],\n",
              "        [0.0627451 , 0.02745098, 0.        ],\n",
              "        [0.19215687, 0.10588235, 0.03137255],\n",
              "        ...,\n",
              "        [0.4627451 , 0.32941177, 0.19607843],\n",
              "        [0.47058824, 0.32941177, 0.19607843],\n",
              "        [0.42745098, 0.28627452, 0.16470589]],\n",
              "\n",
              "       ...,\n",
              "\n",
              "       [[0.8156863 , 0.6666667 , 0.3764706 ],\n",
              "        [0.7882353 , 0.6       , 0.13333334],\n",
              "        [0.7764706 , 0.6313726 , 0.10196079],\n",
              "        ...,\n",
              "        [0.627451  , 0.52156866, 0.27450982],\n",
              "        [0.21960784, 0.12156863, 0.02745098],\n",
              "        [0.20784314, 0.13333334, 0.07843138]],\n",
              "\n",
              "       [[0.7058824 , 0.54509807, 0.3764706 ],\n",
              "        [0.6784314 , 0.48235294, 0.16470589],\n",
              "        [0.7294118 , 0.5647059 , 0.11764706],\n",
              "        ...,\n",
              "        [0.72156864, 0.5803922 , 0.36862746],\n",
              "        [0.38039216, 0.24313726, 0.13333334],\n",
              "        [0.3254902 , 0.20784314, 0.13333334]],\n",
              "\n",
              "       [[0.69411767, 0.5647059 , 0.45490196],\n",
              "        [0.65882355, 0.5058824 , 0.36862746],\n",
              "        [0.7019608 , 0.5568628 , 0.34117648],\n",
              "        ...,\n",
              "        [0.84705883, 0.72156864, 0.54901963],\n",
              "        [0.5921569 , 0.4627451 , 0.32941177],\n",
              "        [0.48235294, 0.36078432, 0.28235295]]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tKXngkeODBxh"
      },
      "source": [
        "# CNN Architecture\n",
        "![image](https://miro.medium.com/max/2000/1*b4neYlVytzv5lfGmHAcqTQ.png)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DVP5oB3mDtPG"
      },
      "source": [
        "Zero Padding(also known as same padding) the layers are neccesary as it provides the dimensions of input and output images to be equal (32%3 = 2: so we need +1 to make it fully divisible)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ihfRZwx1A2mg"
      },
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, MaxPooling2D, Dropout, Flatten, Conv2D"
      ],
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KG-KN6G_G2j8"
      },
      "source": [
        "model = Sequential()"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ggsk9cAgG7Gx"
      },
      "source": [
        "#First Layer\n",
        "model.add(\n",
        "    Conv2D(\n",
        "        32,                   #Depth\n",
        "        (3,3),                #Filter width, height\n",
        "        activation='relu',    #activation\n",
        "        padding='same',       #zero padding\n",
        "        input_shape=(32,32,3))\n",
        ")"
      ],
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_gv2NRZMKOyO"
      },
      "source": [
        "#Second Layer:: No input required\n",
        "model.add(Conv2D(32, (3, 3), activation='relu', padding='same'))"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MfplT-O4LdF5"
      },
      "source": [
        "#Third Layer\n",
        "model.add(\n",
        "    MaxPooling2D(pool_size=(2,2))\n",
        ")"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eoZWA4V8LBGS"
      },
      "source": [
        "#Fourth layer:: Dropout with probability 0.25\n",
        "model.add(Dropout(0.25))"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MrSYenQPLFg4"
      },
      "source": [
        "#Next Four Layers\n",
        "model.add(Conv2D(64, (3, 3), activation='relu', padding='same'))\n",
        "model.add(Conv2D(64, (3, 3), activation='relu', padding='same'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Dropout(0.25))"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lhNCWcFQLw02"
      },
      "source": [
        "# Nineth layer:: FC:: Prerequisite\n",
        "# Our neurons are spatially arranged in a cube-like format rather than in just one row. \n",
        "# To make this cube-like format of neurons into one row, we have to first flatten it\n",
        "model.add(Flatten())"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oar1SlffMG7Z"
      },
      "source": [
        "#Nineth Layer:: FC\n",
        "model.add(Sequential(Dense(512, activation='relu')))"
      ],
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7Sra1tfdMYKc"
      },
      "source": [
        "#Tenth Layer:: DropOut\n",
        "model.add(Dropout(0.5))"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gOBNXukHMmWc"
      },
      "source": [
        "#Eleventh + Twevelth Layer:: FC + Softmax\n",
        "model.add(Dense(10, activation=\"softmax\"))"
      ],
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KEa5KTGuM79X",
        "outputId": "3fe67e60-7d8a-4f4f-b9e8-6b85b06b054a"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d (Conv2D)              (None, 32, 32, 32)        896       \n",
            "_________________________________________________________________\n",
            "conv2d_1 (Conv2D)            (None, 32, 32, 32)        9248      \n",
            "_________________________________________________________________\n",
            "max_pooling2d (MaxPooling2D) (None, 16, 16, 32)        0         \n",
            "_________________________________________________________________\n",
            "dropout (Dropout)            (None, 16, 16, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 16, 16, 64)        18496     \n",
            "_________________________________________________________________\n",
            "conv2d_3 (Conv2D)            (None, 16, 16, 64)        36928     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 8, 8, 64)          0         \n",
            "_________________________________________________________________\n",
            "dropout_1 (Dropout)          (None, 8, 8, 64)          0         \n",
            "_________________________________________________________________\n",
            "flatten (Flatten)            (None, 4096)              0         \n",
            "_________________________________________________________________\n",
            "sequential_1 (Sequential)    (None, 512)               2097664   \n",
            "_________________________________________________________________\n",
            "dropout_2 (Dropout)          (None, 512)               0         \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 10)                5130      \n",
            "=================================================================\n",
            "Total params: 2,168,362\n",
            "Trainable params: 2,168,362\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PyK3-FmeM91l"
      },
      "source": [
        "model.compile(\n",
        "    loss = \"categorical_crossentropy\", #Used mainly for classification problems\n",
        "    optimizer = \"adam\",  #Adam is simply a type of stochastic gradient descent\n",
        "    metrics = ['accuracy']\n",
        ")"
      ],
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5xjsNVO0NUXr",
        "outputId": "f641a623-6d12-44a4-94ef-686d86698413"
      },
      "source": [
        "hist = model.fit(xtrain, ytrainonehot, \n",
        "           batch_size=32, epochs=20, \n",
        "           validation_split= 0.2)\n",
        "#20% of training dataset is used for validation"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "1250/1250 [==============================] - 51s 17ms/step - loss: 1.5462 - accuracy: 0.4341 - val_loss: 1.1620 - val_accuracy: 0.5870\n",
            "Epoch 2/20\n",
            "1250/1250 [==============================] - 20s 16ms/step - loss: 1.1190 - accuracy: 0.6001 - val_loss: 0.9245 - val_accuracy: 0.6698\n",
            "Epoch 3/20\n",
            "1250/1250 [==============================] - 19s 16ms/step - loss: 0.9559 - accuracy: 0.6617 - val_loss: 0.8948 - val_accuracy: 0.6867\n",
            "Epoch 4/20\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.8561 - accuracy: 0.6994 - val_loss: 0.7628 - val_accuracy: 0.7340\n",
            "Epoch 5/20\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.7869 - accuracy: 0.7223 - val_loss: 0.7625 - val_accuracy: 0.7333\n",
            "Epoch 6/20\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.7279 - accuracy: 0.7440 - val_loss: 0.7214 - val_accuracy: 0.7502\n",
            "Epoch 7/20\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.6829 - accuracy: 0.7592 - val_loss: 0.6827 - val_accuracy: 0.7652\n",
            "Epoch 8/20\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.6430 - accuracy: 0.7717 - val_loss: 0.7067 - val_accuracy: 0.7602\n",
            "Epoch 9/20\n",
            "1250/1250 [==============================] - 18s 15ms/step - loss: 0.6112 - accuracy: 0.7854 - val_loss: 0.7213 - val_accuracy: 0.7526\n",
            "Epoch 10/20\n",
            "1250/1250 [==============================] - 18s 15ms/step - loss: 0.5826 - accuracy: 0.7943 - val_loss: 0.6909 - val_accuracy: 0.7700\n",
            "Epoch 11/20\n",
            "1250/1250 [==============================] - 18s 15ms/step - loss: 0.5515 - accuracy: 0.8059 - val_loss: 0.6653 - val_accuracy: 0.7774\n",
            "Epoch 12/20\n",
            "1250/1250 [==============================] - 18s 15ms/step - loss: 0.5222 - accuracy: 0.8152 - val_loss: 0.6749 - val_accuracy: 0.7807\n",
            "Epoch 13/20\n",
            "1250/1250 [==============================] - 18s 15ms/step - loss: 0.5094 - accuracy: 0.8207 - val_loss: 0.6926 - val_accuracy: 0.7705\n",
            "Epoch 14/20\n",
            "1250/1250 [==============================] - 18s 15ms/step - loss: 0.4872 - accuracy: 0.8276 - val_loss: 0.6700 - val_accuracy: 0.7791\n",
            "Epoch 15/20\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.4691 - accuracy: 0.8346 - val_loss: 0.6723 - val_accuracy: 0.7828\n",
            "Epoch 16/20\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.4534 - accuracy: 0.8373 - val_loss: 0.6985 - val_accuracy: 0.7791\n",
            "Epoch 17/20\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.4268 - accuracy: 0.8494 - val_loss: 0.7245 - val_accuracy: 0.7772\n",
            "Epoch 18/20\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.4251 - accuracy: 0.8495 - val_loss: 0.6806 - val_accuracy: 0.7823\n",
            "Epoch 19/20\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.4131 - accuracy: 0.8546 - val_loss: 0.6700 - val_accuracy: 0.7870\n",
            "Epoch 20/20\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.4069 - accuracy: 0.8572 - val_loss: 0.6803 - val_accuracy: 0.7832\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "fMQt20IdN4ui",
        "outputId": "974f8ed5-1e3b-4395-c0e5-2960ebbc93b9"
      },
      "source": [
        "plt.plot(hist.history['loss'])\n",
        "plt.plot(hist.history['val_loss'])\n",
        "plt.title('Model loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Val'], loc='upper right')\n",
        "plt.show()"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "d18B3NozgGIw",
        "outputId": "895c2aa3-8cb6-4f90-e392-697379d48772"
      },
      "source": [
        "plt.plot(hist.history['accuracy'])\n",
        "plt.plot(hist.history['val_accuracy'])\n",
        "plt.title('Model Accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Val'], loc='upper right')\n",
        "plt.show()"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dJNp4Z_ZgTSv",
        "outputId": "d39eb71f-a037-410a-fd0c-69022f385ec4"
      },
      "source": [
        "model.evaluate(xtest,ytestonehot)"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 [==============================] - 2s 6ms/step - loss: 0.6897 - accuracy: 0.7780\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.6897390484809875, 0.777999997138977]"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PNpZz4EVgaaj"
      },
      "source": [
        "model.save('1cifar10_model.h5') #model will be saved in a file format called HDF5"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rVDGz83bglHQ",
        "outputId": "db27e994-966c-4f75-d30b-49a202268fe4"
      },
      "source": [
        "model2 = Sequential()\n",
        "model2.add(\n",
        "    Conv2D(\n",
        "        32,                   #Depth\n",
        "        (3,3),                #Filter width, height\n",
        "        activation='relu',    #activation\n",
        "        padding='same',       #zero padding\n",
        "        input_shape=(32,32,3))\n",
        ")\n",
        "#Second Layer:: No input required\n",
        "model2.add(Conv2D(32, (3, 3), activation='relu', padding='same'))\n",
        "#Third Layer\n",
        "model2.add(\n",
        "    MaxPooling2D(pool_size=(2,2))\n",
        ")\n",
        "#Fourth layer:: Dropout with probability 0.25\n",
        "model2.add(Dropout(0.25))\n",
        "#Next Four Layers\n",
        "model2.add(Conv2D(64, (3, 3), activation='relu', padding='same'))\n",
        "model2.add(Conv2D(64, (3, 3), activation='relu', padding='same'))\n",
        "model2.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model2.add(Dropout(0.25))\n",
        "# Nineth layer:: FC:: Prerequisite\n",
        "# Our neurons are spatially arranged in a cube-like format rather than in just one row. \n",
        "# To make this cube-like format of neurons into one row, we have to first flatten it\n",
        "model2.add(Flatten())\n",
        "#Nineth Layer:: FC\n",
        "model2.add(Sequential(Dense(512, activation='relu')))\n",
        "#Tenth Layer:: DropOut\n",
        "model2.add(Dropout(0.5))\n",
        "#Eleventh + Twevelth Layer:: FC + Softmax\n",
        "model2.add(Dense(10, activation=\"softmax\"))\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "model2.compile(\n",
        "    loss = \"categorical_crossentropy\", #Used mainly for classification problems\n",
        "    optimizer = \"adam\",  #Adam is simply a type of stochastic gradient descent\n",
        "    metrics = ['accuracy']\n",
        ")\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "hist2 = model2.fit(xtrain, ytrainonehot, \n",
        "           batch_size=32, epochs=10, #Number of passess entire training dataset has been completed\n",
        "           validation_split= 0.2)"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "1250/1250 [==============================] - 22s 17ms/step - loss: 1.5476 - accuracy: 0.4353 - val_loss: 1.1557 - val_accuracy: 0.5880\n",
            "Epoch 2/10\n",
            "1250/1250 [==============================] - 20s 16ms/step - loss: 1.1291 - accuracy: 0.5985 - val_loss: 0.9773 - val_accuracy: 0.6503\n",
            "Epoch 3/10\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.9686 - accuracy: 0.6579 - val_loss: 0.8840 - val_accuracy: 0.6858\n",
            "Epoch 4/10\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.8719 - accuracy: 0.6934 - val_loss: 0.7988 - val_accuracy: 0.7197\n",
            "Epoch 5/10\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.7957 - accuracy: 0.7222 - val_loss: 0.7637 - val_accuracy: 0.7340\n",
            "Epoch 6/10\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.7388 - accuracy: 0.7411 - val_loss: 0.7527 - val_accuracy: 0.7359\n",
            "Epoch 7/10\n",
            "1250/1250 [==============================] - 19s 15ms/step - loss: 0.6910 - accuracy: 0.7562 - val_loss: 0.7271 - val_accuracy: 0.7509\n",
            "Epoch 8/10\n",
            "1250/1250 [==============================] - 18s 15ms/step - loss: 0.6472 - accuracy: 0.7729 - val_loss: 0.7013 - val_accuracy: 0.7627\n",
            "Epoch 9/10\n",
            "1250/1250 [==============================] - 18s 15ms/step - loss: 0.6154 - accuracy: 0.7847 - val_loss: 0.6966 - val_accuracy: 0.7608\n",
            "Epoch 10/10\n",
            "1250/1250 [==============================] - 18s 14ms/step - loss: 0.5805 - accuracy: 0.7942 - val_loss: 0.6648 - val_accuracy: 0.7731\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VdC7jgm2HayA",
        "outputId": "8b6de76b-73e5-43a1-f11d-819c0472762f"
      },
      "source": [
        "model2.evaluate(xtest,ytestonehot)"
      ],
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 [==============================] - 2s 6ms/step - loss: 0.6897 - accuracy: 0.7674\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.6897116303443909, 0.7674000263214111]"
            ]
          },
          "metadata": {},
          "execution_count": 72
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "OIDPJnmwhABc",
        "outputId": "d102b2b8-3d79-4c65-a5b8-dbd0e6019e2f"
      },
      "source": [
        "plt.plot(hist2.history['loss'])\n",
        "plt.plot(hist2.history['val_loss'])\n",
        "plt.title('Model loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Val'], loc='upper right')\n",
        "plt.show()"
      ],
      "execution_count": 73,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXhV1bnH8e+bkwkykonMhDCPYQgRBAWcqohiqyjROtTZDvbW21q1t7d6r62d7FU7eB1KqRPgdWilaEUtqIDMEiDMc0ICmcgAmXPW/WOfQMAAScg+O8l5P8+TJzl7n3PWmzyQX9Zea68lxhiUUkr5Lj+nC1BKKeUsDQKllPJxGgRKKeXjNAiUUsrHaRAopZSP0yBQSikfp0GgVBuISJqIGBHxb8Nz7xCR5ef7Pkp5iwaB6nFEZL+I1ItIzGnHv/T8Ek5zpjKluiYNAtVT7QOymx+IyCigt3PlKNV1aRConupV4LYWj28HXmn5BBGJEJFXRKRYRA6IyH+IiJ/nnEtEfisiJSKyF7i6ldf+WUQKReSQiDwpIq72FikiiSLynoiUichuEbmnxbksEVknIpUickREfuc5Hiwir4lIqYiUi8haEenb3raVaqZBoHqqVUC4iAzz/IKeA7x22nN+D0QA6cBUrOD4lufcPcBMYCyQCdxw2mvnAY3AQM9zrgDu7kCdC4B8INHTxi9E5BLPuWeBZ40x4cAA4E3P8ds9dacA0cD9QE0H2lYK0CBQPVtzr+ByYBtwqPlEi3B41BhTZYzZDzwN3Op5yo3AM8aYPGNMGfBUi9f2BWYA/2aMOW6MKQL+x/N+bSYiKcBk4MfGmFpjzEbgZU72ZBqAgSISY4w5ZoxZ1eJ4NDDQGNNkjFlvjKlsT9tKtaRBoHqyV4GbgTs47bIQEAMEAAdaHDsAJHm+TgTyTjvXrJ/ntYWeSzPlwAtAXDvrSwTKjDFVZ6jhLmAwsN1z+Wdmi+/rQ2CBiBSIyK9FJKCdbSt1ggaB6rGMMQewBo1nAO+cdroE6y/rfi2OpXKy11CIdeml5blmeUAdEGOMifR8hBtjRrSzxAIgSkTCWqvBGLPLGJONFTC/At4SkRBjTIMx5gljzHDgQqxLWLehVAdpEKie7i7gEmPM8ZYHjTFNWNfcfy4iYSLSD3iIk+MIbwIPikiyiPQBHmnx2kJgCfC0iISLiJ+IDBCRqe0pzBiTB6wEnvIMAI/21PsagIh8U0RijTFuoNzzMreITBeRUZ7LW5VYgeZuT9tKtaRBoHo0Y8weY8y6M5z+HnAc2AssB94A5nrOvYR1+SUH2MBXexS3AYHAVuAo8BaQ0IESs4E0rN7Bu8DPjDEfe85dCeSKyDGsgeM5xpgaIN7TXiXW2MenWJeLlOoQ0Y1plFLKt2mPQCmlfJwGgVJK+TgNAqWU8nEaBEop5eO63VK4MTExJi0tzekylFKqW1m/fn2JMSa2tXPdLgjS0tJYt+5MswGVUkq1RkQOnOmcXhpSSikfp0GglFI+ToNAKaV8XLcbI1BKqfZqaGggPz+f2tpap0uxXXBwMMnJyQQEtH1BWg0CpVSPl5+fT1hYGGlpaYiI0+XYxhhDaWkp+fn59O/fv82v00tDSqker7a2lujo6B4dAgAiQnR0dLt7PhoESimf0NNDoFlHvk+fCYLdRcd4YlEu9Y26bLtSSrXkM0GQV1bNX1bs55NtR5wuRSnlY0pLSxkzZgxjxowhPj6epKSkE4/r6+vP+tp169bx4IMP2lqfzwwWXzw4lsSIYN5Yc5CrRnVk/xCllOqY6OhoNm7cCMDjjz9OaGgoP/zhD0+cb2xsxN+/9V/HmZmZZGZm2lqfz/QIXH7C7MwUlu8uIa+s2ulylFI+7o477uD+++/nggsu4OGHH2bNmjVMmjSJsWPHcuGFF7Jjxw4Ali1bxsyZMwErRO68806mTZtGeno6zz33XKfU4jM9AoAbJ6Tw3L928ea6PP79iiFOl6OUcsATi3LZWlDZqe85PDGcn10zot2vy8/PZ+XKlbhcLiorK/n888/x9/fn448/5rHHHuPtt9/+ymu2b9/O0qVLqaqqYsiQITzwwAPtumegNT4VBEmRvZg2OJY31+Xx/UsH4e/ymQ6RUqoLmj17Ni6XC4CKigpuv/12du3ahYjQ0NDQ6muuvvpqgoKCCAoKIi4ujiNHjpCcnHxedfhUEADMyUrlvlfXs2xHMZcN7+t0OUopL+vIX+52CQkJOfH1T3/6U6ZPn867777L/v37mTZtWquvCQoKOvG1y+WisbHxvOvwuT+JLxkaR2xYEPPXHHS6FKWUOqGiooKkpCQA5s2b59W2fS4IAlx+zB6fzNIdRRRW1DhdjlJKAfDwww/z6KOPMnbs2E75K789xBjj1QbPV2ZmpjnfjWkOlB5n6m+W8dDlg3nw0kGdVJlSqqvatm0bw4YNc7oMr2nt+xWR9caYVueh+lyPAKBfdAhTBsawcG0ebnf3CkKllOpsPhkEAHOyUjhUXsPnu0ucLkUppRzls0Fw+fC+RIUEMn+1DhorpXybzwZBkL+L68cl8fG2IxRX1TldjlJKOcZngwDgpgmpNLoNb63Pd7oUpZRyjE8HwcC4ULLSoli49iDdbfaUUkp1Fp8OAoDsC1LYX1rNF3tLnS5FKdVDTZ8+nQ8//PCUY8888wwPPPBAq8+fNm0a5ztNvj18PgiuGplAeLA/C9bkOV2KUqqHys7OZsGCBaccW7BgAdnZ2Q5VdCqfD4LgABffGJfMP7ccpuz42TeIUEqpjrjhhhtYvHjxiU1o9u/fT0FBAfPnzyczM5MRI0bws5/9zLH6fG7RudbMyUph3sr9vLMhn7svSne6HKWUnT54BA5v7tz3jB8FV/3yjKejoqLIysrigw8+YNasWSxYsIAbb7yRxx57jKioKJqamrj00kvZtGkTo0eP7tza2sDnewQAQ+PDGZMSyYK1eTporJSyRcvLQ82Xhd58803GjRvH2LFjyc3NZevWrY7Upj0Cj5uzUnn47U2sP3CUzLQop8tRStnlLH+522nWrFn84Ac/YMOGDVRXVxMVFcVvf/tb1q5dS58+fbjjjjuora11pDbtEXjMzEggNMif+TporJSyQWhoKNOnT+fOO+8kOzubyspKQkJCiIiI4MiRI3zwwQeO1aZB4NE70J9rxySyeHMBFTWt7wyklFLnIzs7m5ycHLKzs8nIyGDs2LEMHTqUm2++mcmTJztWl14aaiF7QipvrD7I3zce4rZJaU6Xo5TqYa677rpTxiHPtAHNsmXLvFOQh/YIWhiVHMGIxHDmr9FBY6WU79AgOM2crFS2FVayKb/C6VKUUsorNAhOM2tMIr0CXCxYq8tTK9WT+EovvyPfp21BICJzRaRIRLac43kTRKRRRG6wq5b2CA8OYOboBP6+sYBjdd7dN1QpZY/g4GBKS0t7fBgYYygtLSU4OLhdr7NzsHge8AfglTM9QURcwK+AJTbW0W5zslL5v/X5LMopIDsr1elylFLnKTk5mfz8fIqLi50uxXbBwcEkJye36zW2BYEx5jMRSTvH074HvA1MsKuOjhiXGsngvqEsWHNQg0CpHiAgIID+/fs7XUaX5dgYgYgkAV8Hnm/Dc+8VkXUiss4biS4izJmQSk5+BVsLKm1vTymlnOTkYPEzwI+NMe5zPdEY86IxJtMYkxkbG+uF0uAb45II9PfTQWOlVI/nZBBkAgtEZD9wA/AnEbnOwXpOEdk7kBkj43n3y0PU1Dc5XY5SStnGsSAwxvQ3xqQZY9KAt4BvG2P+5lQ9rZmTlUpVbSOLNxc6XYpSStnGzumj84EvgCEiki8id4nI/SJyv11tdrYL+keRHhPCgjV6eUgp1XPZOWuozXuwGWPusKuO8yEi3DQhhac+2M6uI1UM6hvmdElKKdXp9M7ic7h+fDIBLmHBWl2eWinVM2kQnENMaBBXDI/nnQ351DbooLFSqufRIGiDOVkpHK1u4MPcw06XopRSnU6DoA0mD4ghJaoXC3T3MqVUD6RB0AZ+fsJNmSl8sbeU/SXHnS5HKaU6lQZBG83OTMHlp4PGSqmeR4OgjfqGB3PJ0DjeWp9HfeM5V8VQSqluQ4OgHbKzUig5Vs8n2444XYpSSnUaDYJ2mDo4joSIYObr5SGlVA+iQdAOLj9hdmYKn+8qJq+s2ulylFKqU2gQtNONmdbOP/+3TnsFSqmeQYOgnZL79Gbq4FjeXJdPY5MOGiuluj8Ngg6YMyGVw5W1LNvR8/c/VUr1fBoEHXDpsDhiQoN09zKlVI+gQdABAS4/Zmcm86/tRRyuqHW6HKWUOi8aBB00Z0IKbqODxkqp7k+DoIP6RYdw4YBoFq7Lw+02TpejlFIdpkFwHrKzUsk/WsPy3SVOl6KUUh2mQXAerhjRlz69A5ivexorpboxDYLzEOTv4vpxyXy09QjFVXVOl6OUUh2iQXCe5mSl0Og2vL0h3+lSlFKqQzQIztPAuDAmpPVh4do8jNFBY6VU96NB0Amys1LZV3KcVXvLnC5FKaXaTYOgE8wYlUB4sL8OGiuluiUNgk4QHODi62OT+OeWwxw9Xu90OUop1S4aBJ1kTlYq9U1u3vnykNOlKKVUu2gQdJJhCeFkpESyYM1BHTRWSnUrGgSd6OasFHYVHWPDwaNOl6KUUm2mQdCJZo5OJCTQxRurdSE6pVT3oUHQiUKC/Ll2TBKLNxdQUdPgdDlKKdUmGgSdLDsrhdoGN+9t1EFjpVT3oEHQyUYlRTA8IZz5a/ROY6VU92BbEIjIXBEpEpEtZzh/i4hsEpHNIrJSRDLsqsWbRITsrBS2Flay+VCF0+UopdQ52dkjmAdceZbz+4CpxphRwH8DL9pYi1fNGptEcICf3mmslOoWbAsCY8xnwBkX3zHGrDTGNM+zXAUk21WLt4UHBzBzdCLvbSzgeF2j0+UopdRZdZUxgruAD850UkTuFZF1IrKuuLjYi2V1XHZWCsfrm1iUU+B0KUopdVaOB4GITMcKgh+f6TnGmBeNMZnGmMzY2FjvFXcexqX2YVBcKPPX6j0FSqmuzdEgEJHRwMvALGNMqZO1dDYRYU5WKjl55WwrrHS6HKWUOiPHgkBEUoF3gFuNMTudqsNO3xibRKC/Hwt00Fgp1YXZOX10PvAFMERE8kXkLhG5X0Tu9zzlP4Fo4E8islFE1tlVi1P6hARy1ch43vnyEDX1TU6Xo5RSrfK3642NMdnnOH83cLdd7XcVcyak8veNBby/uZDrx/eYiVFKqR7E8cHinm5iehT9Y0JYsFYvDymluiYNApuJCDdNSGHt/qPsLqpyuhyllPoK3wmC+uOw8g/gdnu96evHJePvJyxYo1NJlVJdj+8EQe7fYMlP4J+PgJcXg4sNC+KKEX15e0M+dY06aKyU6lp8JwjG3AyTvgtrXoBPf+X15udMSOVodQMf5h7xettKKXU2vhMEInDFkzDmm7DsKVj9glebnzIwhuQ+vfSeAqVUl+M7QQBWGFzzLAydCR88DDkLvda0n59wU2YKK/eUcqD0uNfaVUqpc/GtIABw+cP1f4b+F8PfHoAdZ1zrrtPNzkzBT2CBrj+klOpCfC8IAAKCYc4bkJAB/3cH7F/hlWbjI4K5ZGhfFq7N42BptVfaVEqpc/HNIAAICoNb3oLIfjB/DhRs9EqzP/zaYNzGcNOLX7C3+JhX2lRKqbPx3SAACImGW9+F4Ah47Xoo2W17k0Pjw5l/z0TqG93c9OIqdh3Rm8yUUs7y7SAAiEiCW/9mDSS/eh1U5Nve5LCEcBbeNxEBbnpxFVsLdJlqpZRzNAgAYgbCN9+G2gp49etw3P6tEQbGhbHwvkkE+fuR/dIqNuWX296mUkq1RoOgWUIG3LwQyg/C69dDrf1/pfePCeHN+yYRFuzPLS+tZv2Bo+d+kVJKdbI2BYGIhIiIn+frwSJyrYgE2FuaA/pdCDe+Aoc3w4KboaHW9iZTonqz8L5JRIcGctufV7N6b4/aqE0p1Q20tUfwGRAsIknAEuBWYJ5dRTlq8Nfguudh/+fw1p3Q1Gh7k0mRvVh43yTiI4K5/S9rWL6rxPY2lVKqWVuDQIwx1cA3gD8ZY2YDI+wry2Gjb4SrfgM7FsN73/PKiqV9w4NZeN8k0qJDuPOva1m6vcj2NpVSCtoRBCIyCbgFWOw55rKnpC7ignth2mOQ84a1aqkXViyNCQ1i/j0TGdw3lHtfXceS3MO2t6mUUm0Ngn8DHgXeNcbkikg6sNS+srqIqQ/DBQ/Aqj/BZ7/1SpN9QgJ5/e6JjEiM4Nuvb+Afmwq80q5Syne1ac9iY8ynwKcAnkHjEmPMg3YW1iWIwNd+ATVHYemT0CsSsu6xvdmIXgG8elcWd85by4Pzv6Shyc3Xx+p+x0ope7R11tAbIhIuIiHAFmCriPzI3tK6CD8/mPUHGDID3v8RbH7LK82GBQfw1zuzmJgezUNv5rBQ9zxWStmkrZeGhhtjKoHrgA+A/lgzh3yDKwBu+Av0mwzv3gc7l3il2d6B/sy9YwIXD4rlx29v5tUv9nulXaWUb2lrEAR47hu4DnjPGNMAeHe/R6cFBEP2fOg7At68FQ584ZVmgwNcvHjbeC4bFsdP/57Ly5/v9Uq7Sinf0dYgeAHYD4QAn4lIP8D3FsgJDodvvgMRKfDGTVC4ySvNBvm7+NMt45kxKp4nF2/jj0vtXxxPKeU72hQExpjnjDFJxpgZxnIAmG5zbV1TSIy1YmlQGLz2DSjd45VmA/39eG7OWGaNSeQ3H+7gfz7aifHClFalVM/X1sHiCBH5nYis83w8jdU78E2RKXDb38C44ZXroNI7Uzz9XX787sYxzB6fzLOf7OLXH+7QMFBKnbe2XhqaC1QBN3o+KoG/2FVUtxAzyFqxtOaotWJpdZlXmnX5Cb+6fjS3XJDK88v28N//2KZhoJQ6L20NggHGmJ8ZY/Z6Pp4A0u0srFtIHAs3L4CyffD6DVDnnU1m/PyEJ68bybcmpzF3xT5++vctuN0aBkqpjmlrENSIyJTmByIyGaixp6RuJm0KzJ5nbXW54BZorPNKsyLCf84czn1T03lt1UEeeWcTTRoGSqkOaNOdxcD9wCsiEuF5fBS43Z6SuqGhM2DWH+Fv98Pbd8EN88DV1h9tx4kIj1w5lCB/F899sov6Rje/nZ2Bv0u3mVBKtV1bl5jIATJEJNzzuFJE/g3wzvzJ7mBMNtSWwz8fgX98H679g7VEhc1EhIcuH0yQvx+/+XAH9U1unp0zlgANA6VUG7Xrz1bP3cXNHgKe6dxyurmJD1iDx5/+Cnr1gcv/2ythAPCd6QMJ8vfjycXbqG/cwB9vGUuQf89eIFYp1TnO58/Gs/6GE5G5IlIkIlvOcF5E5DkR2S0im0Rk3HnU0nVMexSy7oWVv4fl/+PVpu++KJ3/mjWCj7cd4d5X1lPb0OTV9pVS3dP5BMG5RibnAVee5fxVwCDPx73A8+dRS9chAlf+CkbNhk+egHVzvdr8bZPS+OU3RvHZrmLunLeW6nr7d1hTSnVvZw0CEakSkcpWPqqAxLO91hjzGXC2yfWzgFc8dyqvAiJFJKHd30FX5OdnbXc56Gvwj4dgy9tebX5OVipPz85g1d5Sbp+7hqraBq+2r5TqXs4aBMaYMGNMeCsfYcaY850WkwTktXic7zn2FSJyb/NdzcXFxefZrJe4AqxppamT4J37YPfHXm3+G+OSeS57LBsOlnPrn9dQUaNhoJRqXbeYWmKMedEYk2mMyYyNjXW6nLYL7G3dcBY3FBbeCgdXe7X5maMT+dMt48gtqOCWl1dx9Hi9V9tXSnUPTgbBISClxeNkz7GeJTjCWrE0LAHemA2HWx07t83XRsTz4q2Z7DxyjOyXVlFyzDs3vCmlug8ng+A94DbP7KGJQIUxptDBeuwTGmctUhcQYq1YWubdPQWmD41j7u0T2F96nJte+IIjlbVebV8p1bXZFgQiMh/4AhgiIvkicpeI3C8i93ue8j6wF9gNvAR8265auoTIVGv56qYG+MvVkPsueHGxuCmDYvjrt7I4XFHLTS98QUG5rhCilLJId1u5MjMz06xbt87pMjquMAfefQCKciFlInztF5A83mvNrz9wlDvmriGidwDz75lISlRvr7WtlHKOiKw3xmS2dq5bDBb3KAkZcP/ncM2zULYHXr4E3r4HKvK90vz4fn14/Z4LqKpt5OrnPmfein00Nrm90rZSqmvSIHCCnwvG3wEPfgkX/Tts/Tv8fjz860moO2Z786OTI3nn2xcyOjmSxxdt5ernlrNyd4nt7Sqluia9NNQVlB+Ejx+3bjwL7QuX/BTG3GwFho2MMSzZeoQnF28lr6yGq0bG89iMYXq5SKke6GyXhjQIupK8NfDhY5C/FuJHWeMH/S+2vdnahiZe/nwvf1y6B7cx3Dd1AA9MHUCvQF20TqmeQoOgOzHG6hl8/DhU5MGQGdYqpjEDbW+6sKKGp97fzns5BSRGBPOTq4czY1Q84qUVVJVS9tEg6I4aamDV8/D576CxBibcA1Mfht5Rtje9Zl8ZP3svl22FlVzQP4rHrx3BsIRw29tVStlHg6A7O1YES38OG16BoHCY9ghk3gX+gbY22+Q2LFh7kN9+uIOKmgZuuaAfD10+mD4h9rarlLKHBkFPcCQXPvwJ7F0KUQPgiidhyFW2b3xTXl3PMx/v4tVVBwgL9uffrxjCzVmpuPz0cpFS3YkGQU9hDOz6CJb8BEp2QtpF1oBywmjbm95+uJIn3tvKF3tLGRofxuPXjmBierTt7SqlOocGQU/T1ADr58HSX1hbY469xZpyGhZva7PGGP655TBPLt7GofIaZo5O4LEZw0iM7GVru0qp86dB0FPVlMNnv4HVL4ArEKb8ACZ9x1r+2s5m65t44bM9PL9sDyLw7WkDuffidIIDdLqpUl2VBkFPV7YXPvpP2LYIwpPgssdh5A3WTmk2yj9azVPvb2fx5kKS+/TiP64extdG6HRTpboiDQJfsX+FdUNa4UZIHGeNH/SbZHuzX+wp5YlFuWw/XMXkgdH87JoRDO4bZnu7Sqm20yDwJW43bFoInzwBVYUw/DqrhxDV39ZmG5vcvLHmIE8v2cmxukZundiPH1w2mIjeAba2q5RqGw0CX1R/HFb+HlY8C+5GuOB+uPiH1o5pNjp6vJ6nP9rBG6sPEtk7kB9eMYSbJqTodFOlHKZB4MsqC6xVTTe+Ab2jYfpjMO52cPnb2uzWgkoeX5TLmn1ljEgM54lrR5CZZv9d0Uqp1mkQKCjYaN2QdmA5xA6FK34Ogy6ztUljDP/YVMgv3t9GYUUts8Yk8uhVw4iPCLa1XaXUV2kQKIsxsH0xfPRTa6ZR+nTr/gObd0irrm/kf5ft4X8/24u/n/Cd6QO5a0p/nW6qlBdpEKhTNdbD2pfg86ehutRa4XT6Y9bS1zbKK6vm54u38c/cw6RG9eanM4dz2bA4nW6qlBdoEKjW1VXB6v+FFb+HugoY8XWY9ijEDrG12eW7SnhiUS67io5x0aAYHpsxTFc3VcpmGgTq7GqOwhd/tJa9bqiGUTfCtB9DVLptTTY0uXlt1QF+99FOqmobmTYklnsvTmdSerT2EJSygQaBapvjpbDiGVjzEjTVw9hvwsU/gsgU25osr67ntVUHmLdyPyXH6hmdHMF9Fw/gypHxOuVUqU6kQaDap+qwtSHO+r9Yj8d/Cy56yNZF7WobmnhnwyFe+nwv+0qO0y+6N3dflM7s8ck6qKxUJ9AgUB1TnmctarfxdfALgKy7YfIPIMS+5aeb3IaPth7m+U/3kpNXTnRIILdfmMatE/vppjhKnQcNAnV+yvbCp7+2lq4I6A0TH4BJ34VekbY1aYxhzb4yXvhsL//aXkSvABc3TUjhrin9SYmyd3VVpXoiDQLVOYp3wLKnIPddCIqAC78HE++HIHsXmNt5pIoXP9vL3zcewm3g6lEJ3HtxOiOT7F0uQ6meRINAda7Dm2HpU7BjMfSKsvZBmHC37fsgFFbU8JcV+3lj9UGO1TUyZWAM901NZ8rAGJ1ppNQ5aBAoe+Svh6U/hz2fQGhfuOiHMP528A+ytdnK2gbeWH2Qucv3UVRVx/CEcO6bms7VoxLwd9m7B4NS3ZUGgbLXgZXWwnYHVkB4Mkx9GMbcDC57l6Cua2zi718W8MJne9hTfJzkPr24e0p/bpyQQu9AexfVU6q70SBQ9jMG9i6zegj5a6FPmnWX8qjZ4Gfv9E+32/DJ9iJe+HQP6w4cJbJ3ALdNSuP2Sf2IDrW3d6JUd6FBoLzHGNi1xOohHN4EMYOtQBh+ne1bZwKsP1DGC5/uZcnWIwT5+zE7M5m7p6STFhNie9tKdWUaBMr73G7Y/g+rh1C8HfqOhOk/gSFXgRcGdncXHePlz/fyzoZDNLrdXDXSmmmUkWLflFelujLHgkBErgSeBVzAy8aYX552PhX4KxDpec4jxpj3z/aeGgTdjLsJtrwDy35h3Y+QOA4u+Q8YcIlXAqGospZ5K/fz6qoDVNU2MjE9ivumDmDa4FidaaR8iiNBICIuYCdwOZAPrAWyjTFbWzznReBLY8zzIjIceN8Yk3a299Ug6KaaGiFnvnVjWsVBSL3QCoS0yV5p/lhdIwvWHOTPy/dRWFHL0Pgw7r04nWsyEgnQmUbKB5wtCOz8H5AF7DbG7DXG1AMLgFmnPccAzesPRwAFNtajnOTyh3G3wvfWw9VPw9F9MG8GvDLLukGtPM8aX7BJaJA/d1+Uzqc/ms7TszNwG8NDb+Yw9ddLefnzvRyra7StbaW6Ojt7BDcAVxpj7vY8vhW4wBjz3RbPSQCWAH2AEOAyY8z6Vt7rXuBegNTU1PEHDhywpWblRQ01sG6utbhddYl1LCQOkjMhaRwkjbcuI9m0jIUxhmU7ivnfT/ewel8Z4cH+XDsmkWtGJzIhLQo/XflU9TBOXRpqSxA85KnhaRGZBPwZGGmMcZ/pffXSUA/TWA9HtsCh9Sc/SnaePB89yAqF5o/4kZ1+w9rGvHLmLt/Hkq2HqW1wEx8ezNWjE7gmI5GM5AgdS1A9glNBMAl43BjzNc/jRwGMMU+1eE4uVljkeR7vBYd6KPkAABEASURBVCYaY4rO9L4aBD6gphwKvvQEwwY4tA6OHbHOuQKtLTWTxkNSpvU5Kr1TpqYer2vk421H+MemQj7dUUx9k5vUqN5ck5HAzNGJDI0P01BQ3ZZTQeCPNVh8KXAIa7D4ZmNMbovnfAAsNMbME5FhwCdAkjlLURoEPsgYqDx0sseQv94Kiobj1vngCOsyUnLmyZ5DaNx5NVlR08CHuYdZlFPAyj2lNLkNA+NCuTYjkZmjE0iPDe2Eb0wp73Fy+ugM4BmsqaFzjTE/F5H/AtYZY97zzBR6CQjFGjh+2Biz5GzvqUGgAGtaavEOTzissz4f2QqmyTofkXLqJaWEDAjq2C/vkmN1fLDFCoW1+8swBkYmhXPN6ESuHp1Ach9dFlt1fXpDmfIN9dVQmNNivGEdlB+0zokfxA0/ORCdlAmxQ63ZTO1QWFHD4k2FLNpUSE5eOQDj+/XhmtEJzBidQFxYcGd/V0p1Cg0C5buOFUPBBs8lJU/Podb6BU5Ab0gYczIcwpOsWUrBkdbncwxKHyytZtGmAhblFLD9cBV+AhPTo7kmI5ErR8TrjmqqS9EgUKqZMdYdzi1nKRVugqa6rz7Xvxf06nNqOJzhc15NIJ/sr2fRjmo2lwluv0AuGhTDNRmJXD68L2HB9q7EqtS5aBAodTaN9VC8DY4VWTOWasvP/bn+2Fnfst4vmHJ3CGXu3lRJCIEhUUTFxJHQNx7/kKizhEof8NeehOp8ZwsCXbRdKf9AazC5PZoaoLailaA4CjXlBNaWE1tzlICyYoLLiqk/dgi/AzuoO3Acf6k9+3tHpELMQGvl1mjP55jBEBbvlfWZlO/RIFCqI1wBEBJjfZyBYN0y3wdochtW7yvlDzmFfLQ5D3dNBUnBtVyRHsz0fgEMi3TjqquA6lIo3WPdVLfh1ZNTZAECw1oExCCIGWR9HZUOATpIrTpOLw0p5WUNTW6W7y5hUU4BS3KPcKyukeiQQGaMSmDm6AQy06Jw+Yk1nlFVaIVCyS7Px04o3Q0VeS3eUaBPP084DPYEhOfrkFjtRShAxwiU6rJqG5pYtqOYRZsK+GTbEWob3ET2DuDiQbFMHxrLxYNiW99lrf64FQinBMQuKNkNjTUnnxcc0XpA9OmvYxE+RoNAqW7geF0jS3cUsXR7MZ/uLKLkWD0ikJEcySVD45g+JI4RieFnXxDP7bbuwm7uRZTuOvl1VeHJ54nL2k40ZvDJy03Nl5xCom3/XpX3aRAo1c243YYtBRUs3V7M0h1F5OSXYwzEhAYxbUgs04fEMWVQDBG92jEttbbyZC/iREDsto61nD7bK6pFD2Lwya8j+7X7BjzVdWgQKNXNlR6r47NdxZ7eQjEVNQ24/ITx/fowfUgc04fGMqRvBxfFczdZd2CX7vaEQ4sxieMt1n90BVoD06cHRPQgCA4/8/urLkGDQKkepLHJTU5++YneQm5BJQAJEcFMGxLH9CGxTB4YQ0hQJ/z1XnPU6jWcEhA7rZvymtd1AghLaD0gwpM6ZWVYdf40CJTqwY5U1vLpjmL+tb2I5btLOFbXSKDLj6z+UdZlpKFxpMeEdO4S2o31cHT/qQFRuguKd0JdxcnnBfQ+9V6I5rCIHgABvTqvHnVOGgRK+Yj6RjfrDpSxbEcxS7cXsavIugM6Nao304fEMm1oHJPSowkOcNlTgDFwvPirPYiSndZ2pDT/vhGITG0REC3CQqe82kKDQCkflVdWzbKdxSzbXsSKPSXUNrgJ8vfjwgHRTPfMREqJ8tIy2vXVULbnqwHR2pTX5lAIT7IW//MPtj67Alt8DramwLqCzn3MFeDz4aJBoJSitqGJ1fvKWLq9iKU7ijhQWg3AgNgQpg+J45KhcWSmRRHo7+Vr+m43VOafDIWWvYljhzuvnVPCofnroFOD4yvHPEESFg/xGZAw+rw3PXKKBoFS6iv2lRw/EQqr95ZR3+QmJNDFlEExTBsSx5SBMd7rLZyJ2w1N9db01sbmz56PU441f65t5Vh7n19vHW/5dfPS5QCh8dbaVAmjIX609TmyX5fvcWgQKKXOqrq+kZW7Sz03tBVRUGEtjJca1ZvJA2OYPDCaCwfEEOWreyzUlMPhzXB4k7Vs+eFN1g55zTOngiM8oZBx8nPMIPCzaSymAzQIlFJtZoxhd9ExVuwuYcWeUlbtKaWqrhGAEYnhTBkYw4UDY8hKi6JXYNf5Red1DTXW9qiHc6xwKMyBoq1WDwKs/Sz6jrB6DM0BETfcsQUCNQiUUh3W2ORm06EKVu4uYfnuEjYcKKe+yU2gy49x/SJPBMPopAj8XT5+z0BTozW+cdgTDM29hzrrXg/8/CFmyKmXluJHeeWGPA0CpVSnqa5vZO3+oyeCofmGtrAgfyYOiGbygGimDIphQGxo59670F253VC+/2QoNPceWt61HZV+crwhIcMamA6N7dQyNAiUUrYpO17PF3tKWb67hBW7SzhYZs1G6hseZI0vDIhh8sAY4iN0z4RTVB32hEPOyd5D+YGT58MSTo43NPceIlM7PCitQaCU8pq8smpWeHoLK/eUUna8HoCBcaHWZaQB0UwcEE247uP8Vc2D0oU5J3sPJTvAuK3zE78DV/6iQ2+tQaCUcoTbbdh+uOpEMKzZV0ZNQxN+AhkpkSd6C+P6RRLk78MDz2dTX20NQhfmWIPN/SZ16G00CJRSXUJ9o5svDx49MSNpY145TW5DcIAfWf2t8YXJA2MYnnCOfRdUu2kQKKW6pKraBlbvLfNcRiph5xFrbaQ+vQO4cEAMF6RHMTIpgmHx4b49VbUTnC0IdJcJpZRjwoIDuGx4Xy4b3hewVlJduaeE5btKWbG7hMWbrV3V/ATSY0MZkRjOiMRwRiZGMDwxnMjePnqDWyfTHoFSqksyxlBYUcuWQxXkFlSSW1DJ1oKKE3c9AyRF9vKEQwQjk6zPfcODdNpqK7RHoJTqdkSExMheJEb24ooR8SeOlx2vJ7fgZDjkHqrgo21HaP6bNjokkOGnhUO/qN465nAWGgRKqW4lKiSQiwbFctGgkzdcHa9rZFuhFQzNPYg/L99LQ5OVDqFB/gxLCGNEYsSJHsSgvqEE+Pqd0B4aBEqpbi8kyJ/MtCgy06JOHKtrbGLXkWOn9B7eXJdHdb21UFygy4/B8aGMSLB6DsMTIxiWEEbvQN/7teh737FSyicE+bsYmRTByKSIE8ea3IZ9JcfJLahgqycclmw9zMJ1eYA1KN0/JuSUy0ojfGBQWoNAKeUzXH7CwLhQBsaFMmtMEmANShdU1JLbYlB67f4y3sspOPG65D69yEiJZExyJBkpkYxMCu9RPQdbvxMRuRJ4FnABLxtjftnKc24EHsfazDTHGHOznTUppVRLIkJSZC+SzjAoveWQNe6w8WA5izednM46uG8YY1KsYMhIjmRw39Buu/qqbdNHRcQF7AQuB/KBtUC2MWZri+cMAt4ELjHGHBWROGNMUatv6KHTR5VSTimuqmNTfjk5eeVszK8gJ6+cipoGAIID/BiVFEGGp9cwJiWS5D69usxUVqemj2YBu40xez1FLABmAVtbPOce4I/GmKMA5woBpZRyUmxYEJcO68ulw6wb4IwxHCitJie/nI15VkC8suoA9cv3AdYMp4zkCKvX4Ok5dMVd3uwMgiQgr8XjfOCC054zGEBEVmBdPnrcGPPP099IRO4F7gVITU21pVillGovESEtJoS0mJATYw71jW52Hqk6EQw5+eUs21l84j6H1KjenlCIYExKJCMSIxxfPsPp0Q5/YBAwDUgGPhORUcaY8pZPMsa8CLwI1qUhbxeplFJtFejvd2K20jcn9gPgWF0jm/MryPFcVlq/v4xFnsFol58wpG+Y53KS1XsYFBeGy4s3wNkZBIeAlBaPkz3HWsoHVhtjGoB9IrITKxjW2liXUkp5VWiQP5MGRDNpQPSJY0WVteR4xhly8stZvKmA+WsOAtA70MXIxAgyPMGQkWzveIOdg8X+WIPFl2IFwFrgZmNMbovnXIk1gHy7iMQAXwJjjDGlZ3pfHSxWSvVEbrfhQFm1NRDtCYfcgkrqG61NaaJDArl/6gDuuTi9Q+/vyGCxMaZRRL4LfIh1/X+uMSZXRP4LWGeMec9z7goR2Qo0AT86WwgopVRP5ecn9I8JoX9MCNeNPTnesONwFRs9l5T62rTdp64+qpRSPuBsPYLuefeDUkqpTqNBoJRSPk6DQCmlfJwGgVJK+TgNAqWU8nEaBEop5eM0CJRSysdpECillI/rdjeUiUgxcKCDL48BSjqxnO5Ofx6n0p/HSfqzOFVP+Hn0M8bEtnai2wXB+RCRdWe6s84X6c/jVPrzOEl/Fqfq6T8PvTSklFI+ToNAKaV8nK8FwYtOF9DF6M/jVPrzOEl/Fqfq0T8PnxojUEop9VW+1iNQSil1Gg0CpZTycT4TBCJypYjsEJHdIvKI0/U4SURSRGSpiGwVkVwR+b7TNTlNRFwi8qWI/MPpWpwmIpEi8paIbBeRbSIyyemanCIiP/D8H9kiIvNFxJ4twhzmE0EgIi7gj8BVwHAgW0SGO1uVoxqBfzfGDAcmAt/x8Z8HwPeBbU4X0UU8C/zTGDMUyMBHfy4ikgQ8CGQaY0Zibbk7x9mq7OETQQBkAbuNMXuNMfXAAmCWwzU5xhhTaIzZ4Pm6Cus/epKzVTlHRJKBq4GXna7FaSISAVwM/BnAGFNvjCl3tipH+QO9RMQf6A0UOFyPLXwlCJKAvBaP8/HhX3wtiUgaMBZY7WwljnoGeBhwO11IF9AfKAb+4rlU9rKIhDhdlBOMMYeA3wIHgUKgwhizxNmq7OErQaBaISKhwNvAvxljKp2uxwkiMhMoMsasd7qWLsIfGAc8b4wZCxwHfHJMTUT6YF056A8kAiEi8k1nq7KHrwTBISClxeNkzzGfJSIBWCHwujHmHafrcdBk4FoR2Y91yfASEXnN2ZIclQ/kG2Oae4hvYQWDL7oM2GeMKTbGNADvABc6XJMtfCUI1gKDRKS/iARiDfi853BNjhERwboGvM0Y8zun63GSMeZRY0yyMSYN69/Fv4wxPfKvvrYwxhwG8kRkiOfQpcBWB0ty0kFgooj09vyfuZQeOnDu73QB3mCMaRSR7wIfYo38zzXG5DpclpMmA7cCm0Vko+fYY8aY9x2sSXUd3wNe9/zRtBf4lsP1OMIYs1pE3gI2YM20+5IeutSELjGhlFI+zlcuDSmllDoDDQKllPJxGgRKKeXjNAiUUsrHaRAopZSP0yBQ6jQi0iQiG1t8dNqdtSKSJiJbOuv9lOoMPnEfgVLtVGOMGeN0EUp5i/YIlGojEdkvIr8Wkc0iskZEBnqOp4nIv0Rkk4h8IiKpnuN9ReRdEcnxfDQvT+ASkZc869wvEZFejn1TSqFBoFRrep12aeimFucqjDGjgD9grVoK8Hvgr8aY0cDrwHOe488BnxpjMrDW62m+m30Q8EdjzAigHLje5u9HqbPSO4uVOo2IHDPGhLZyfD9wiTFmr2fRvsPGmGgRKQESjDENnuOFxpgYESkGko0xdS3eIw34yBgzyPP4x0CAMeZJ+78zpVqnPQKl2sec4ev2qGvxdRM6VqccpkGgVPvc1OLzF56vV3JyC8NbgM89X38CPAAn9kSO8FaRSrWH/iWi1Ff1arEqK1j79zZPIe0jIpuw/qrP9hz7HtaOXj/C2t2rebXO7wMvishdWH/5P4C105VSXYqOESjVRp4xgkxjTInTtSjVmfTSkFJK+TjtESillI/THoFSSvk4DQKllPJxGgRKKeXjNAiUUsrHaRAopZSP+3+BFSSvxyQ4MgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "Cz5cY5nFlRs9",
        "outputId": "f5c8fb5e-4f79-48c3-9838-cf6616ab7acd"
      },
      "source": [
        "plt.plot(hist2.history['accuracy'])\n",
        "plt.plot(hist2.history['val_accuracy'])\n",
        "plt.title('Model Accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Val'], loc='upper right')\n",
        "plt.show()"
      ],
      "execution_count": 74,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 266
        },
        "id": "3aXoeROoltcZ",
        "outputId": "971633ed-2a7e-4579-eb3b-3da1191553ae"
      },
      "source": [
        "my_image = plt.imread(\"/content/cat.png\")\n",
        "from skimage.transform import resize\n",
        "resized = resize(my_image,(32,32,3))\n",
        "image = plt.imshow(resized)"
      ],
      "execution_count": 75,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "--1FveS_mcSf"
      },
      "source": [
        "import numpy as np"
      ],
      "execution_count": 76,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mek2h_jbmfUZ"
      },
      "source": [
        "probabilities = model.predict(\n",
        "    np.array([resized, ])\n",
        "    )\n",
        "#model.predict expects a 4-D array instead of a 3-D array\n",
        "#np.array(...) code is there to change our current array of my_image_resized into a 4-D array \n",
        "#before applying the model.predict function."
      ],
      "execution_count": 77,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "icoT6mgym1sp",
        "outputId": "b9c909a5-b630-42e3-a5a5-60acc41c1a91"
      },
      "source": [
        "probabilities"
      ],
      "execution_count": 78,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.02030825, 0.00180207, 0.00274953, 0.5945772 , 0.04064691,\n",
              "        0.06330916, 0.01720748, 0.21726166, 0.01481535, 0.02732233]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 78
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xQMdiS8-m5t9",
        "outputId": "1a9db451-6296-4be3-faaf-6cb6abcc05a5"
      },
      "source": [
        "number_to_class = ['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']\n",
        "index = np.argsort(probabilities[0,:])\n",
        "print(\"Most likely class:\", number_to_class[index[9]], \"-- Probability:\", probabilities[0,index[9]])\n",
        "print(\"Second most likely class:\", number_to_class[index[8]], \"-- Probability:\", probabilities[0,index[8]])\n",
        "print(\"Third most likely class:\", number_to_class[index[7]], \"-- Probability:\", probabilities[0,index[7]])\n",
        "print(\"Fourth most likely class:\", number_to_class[index[6]], \"-- Probability:\", probabilities[0,index[6]])\n",
        "print(\"Fifth most likely class:\", number_to_class[index[5]], \"-- Probability:\", probabilities[0,index[5]])"
      ],
      "execution_count": 79,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Most likely class: cat -- Probability: 0.5945772\n",
            "Second most likely class: horse -- Probability: 0.21726166\n",
            "Third most likely class: dog -- Probability: 0.06330916\n",
            "Fourth most likely class: deer -- Probability: 0.040646907\n",
            "Fifth most likely class: truck -- Probability: 0.027322328\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UiY2fmSGnBPt",
        "outputId": "89856218-9c90-4ffc-c259-c9bde87f2a8f"
      },
      "source": [
        "probabilities2 = model2.predict(\n",
        "    np.array([resized, ])\n",
        "    )\n",
        "probabilities2"
      ],
      "execution_count": 85,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[5.3403061e-04, 1.1388100e-04, 6.3860500e-03, 2.3458794e-01,\n",
              "        1.9892299e-01, 4.0009025e-01, 2.1736661e-02, 1.3656318e-01,\n",
              "        4.6083113e-04, 6.0418126e-04]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 85
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7lBVRfvHFU2D",
        "outputId": "a5242b25-ceed-4c93-da81-c2801fb2c88f"
      },
      "source": [
        "number_to_class = ['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']\n",
        "index = np.argsort(probabilities2[0,:])\n",
        "print(\"Most likely class:\", number_to_class[index[9]], \"-- Probability:\", probabilities2[0,index[9]])\n",
        "print(\"Second most likely class:\", number_to_class[index[8]], \"-- Probability:\", probabilities2[0,index[8]])\n",
        "print(\"Third most likely class:\", number_to_class[index[7]], \"-- Probability:\", probabilities2[0,index[7]])\n",
        "print(\"Fourth most likely class:\", number_to_class[index[6]], \"-- Probability:\", probabilities2[0,index[6]])\n",
        "print(\"Fifth most likely class:\", number_to_class[index[5]], \"-- Probability:\", probabilities2[0,index[5]])"
      ],
      "execution_count": 81,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Most likely class: dog -- Probability: 0.40009025\n",
            "Second most likely class: cat -- Probability: 0.23458794\n",
            "Third most likely class: deer -- Probability: 0.19892299\n",
            "Fourth most likely class: horse -- Probability: 0.13656318\n",
            "Fifth most likely class: frog -- Probability: 0.021736661\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z-tHnsbRlbcn"
      },
      "source": [
        "from keras.models import load_model\n",
        "model3 = load_model('/content/my_cifar10_model.h5')"
      ],
      "execution_count": 83,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IK4O-OrhFeKn",
        "outputId": "e3b3b8f3-dfca-4217-c4b0-3be6c909eb29"
      },
      "source": [
        "model3.evaluate(xtest,ytestonehot)"
      ],
      "execution_count": 84,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 [==============================] - 2s 6ms/step - loss: 0.7219 - accuracy: 0.7709\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.7218727469444275, 0.7709000110626221]"
            ]
          },
          "metadata": {},
          "execution_count": 84
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S590P7kgIEAJ"
      },
      "source": [
        "probabilities3 = model.predict(\n",
        "    np.array([resized, ])\n",
        "    )"
      ],
      "execution_count": 87,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-EIfWApdIJys",
        "outputId": "8a1eb418-9f41-44b6-c824-e69906edfbb0"
      },
      "source": [
        "number_to_class = ['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']\n",
        "index = np.argsort(probabilities3[0,:])\n",
        "print(\"Most likely class:\", number_to_class[index[9]], \"-- Probability:\", probabilities3[0,index[9]])\n",
        "print(\"Second most likely class:\", number_to_class[index[8]], \"-- Probability:\", probabilities3[0,index[8]])\n",
        "print(\"Third most likely class:\", number_to_class[index[7]], \"-- Probability:\", probabilities3[0,index[7]])\n",
        "print(\"Fourth most likely class:\", number_to_class[index[6]], \"-- Probability:\", probabilities3[0,index[6]])\n",
        "print(\"Fifth most likely class:\", number_to_class[index[5]], \"-- Probability:\", probabilities3[0,index[5]])"
      ],
      "execution_count": 88,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Most likely class: cat -- Probability: 0.5945772\n",
            "Second most likely class: horse -- Probability: 0.21726166\n",
            "Third most likely class: dog -- Probability: 0.06330916\n",
            "Fourth most likely class: deer -- Probability: 0.040646907\n",
            "Fifth most likely class: truck -- Probability: 0.027322328\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XOdN-GR_IR6N"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}