CIFAR10.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "CIFAR10.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyOPif4KpfUpvAfX6WldSRXe",
      "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/ORUNSOLU/Practice/blob/main/CIFAR10.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "odFnlrC2F5oW"
      },
      "source": [
        "Attached below is a link to the asssignment\n",
        "\n",
        "https://dphi.tech/challenges/deep-learning-bootcamp-assignment-1-intermediate-object-recognition/28/data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "V5918WspF00b"
      },
      "source": [
        "Import the required libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "j8HZZYL0cDgH"
      },
      "source": [
        "from tensorflow.keras.datasets.cifar10 import load_data\n",
        "from tensorflow.keras.layers import Flatten, Dense, MaxPooling2D, Conv2D\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow import keras\n",
        "\n",
        "import pandas as pd\n",
        "from PIL import Image\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.model_selection import train_test_split\n",
        "from warnings import filterwarnings\n",
        "\n",
        "filterwarnings('ignore')"
      ],
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-bCDqbBEFxJK"
      },
      "source": [
        "Load the data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FtuonwzKFIJp"
      },
      "source": [
        "(x_train, y_train), (_, __) = load_data()\n",
        "test_data = pd.read_csv(\"https://raw.githubusercontent.com/dphi-official/Datasets/master/cifar_image_flattened_pixels.csv\")\n",
        "labels = ['Airplane', 'Automobile', 'Bird', 'Cat', 'Deer', 'Dog', 'Frog', 'Horse', 'Ship', 'Truck']\n",
        "x_train = x_train/255.0"
      ],
      "execution_count": 58,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WGhTRu3JJmwE"
      },
      "source": [
        "Split the data into training and validation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7fxSr-KfOoe6"
      },
      "source": [
        "x_train, x_val, y_train, y_val = train_test_split(x_train, y_train, test_size=0.15)"
      ],
      "execution_count": 59,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "B-Fvq_m-JtgC"
      },
      "source": [
        "Display some images"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 697
        },
        "id": "UZWR7RyvJvn-",
        "outputId": "c9e17636-a23b-459a-ec1b-3db68ca1eabe"
      },
      "source": [
        "plt.figure(figsize=(12, 12))\n",
        "\n",
        "for i in range(9):\n",
        "  image_name = labels[int(y_train[i])]\n",
        "  plot = plt.subplot(3, 3, i+1)\n",
        "  plot.axis('off')\n",
        "  image = Image.fromarray(x_train[i], 'RGB')\n",
        "  plot.set_title(image_name, fontsize=10)\n",
        "  plt.imshow(image)\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x864 with 9 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1NXV8d87f3LR"
      },
      "source": [
        "Build the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "joViOj8vITlr",
        "outputId": "b2aa3298-4d96-463f-9415-921b771527e0"
      },
      "source": [
        "model = Sequential()\n",
        "model.add(Conv2D(16,(3,3),padding='SAME',activation='relu',input_shape=(32,32,3)))\n",
        "model.add(MaxPooling2D(pool_size=(2,2)))\n",
        "\n",
        "model.add(Conv2D(32,(3,3),padding='SAME',activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2,2)))\n",
        "\n",
        "model.add(Conv2D(32,(3,3),padding='SAME',activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2,2)))\n",
        "\n",
        "model.add(Flatten())\n",
        "\n",
        "model.add(Dense(64,activation='relu'))\n",
        "model.add(Dense(10,activation='softmax'))\n",
        "model.summary()"
      ],
      "execution_count": 90,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_17\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_43 (Conv2D)           (None, 32, 32, 16)        448       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_40 (MaxPooling (None, 16, 16, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_44 (Conv2D)           (None, 16, 16, 32)        4640      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_41 (MaxPooling (None, 8, 8, 32)          0         \n",
            "_________________________________________________________________\n",
            "conv2d_45 (Conv2D)           (None, 8, 8, 32)          9248      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_42 (MaxPooling (None, 4, 4, 32)          0         \n",
            "_________________________________________________________________\n",
            "flatten_16 (Flatten)         (None, 512)               0         \n",
            "_________________________________________________________________\n",
            "dense_32 (Dense)             (None, 64)                32832     \n",
            "_________________________________________________________________\n",
            "dense_33 (Dense)             (None, 10)                650       \n",
            "=================================================================\n",
            "Total params: 47,818\n",
            "Trainable params: 47,818\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CLHzG5L9G7YE"
      },
      "source": [
        "Add CallBack"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OpT915MBG9UN"
      },
      "source": [
        "checkPoint_cb = keras.callbacks.ModelCheckpoint('model_cp.h5', save_best_only=True)"
      ],
      "execution_count": 89,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VOUzTfLhUQW9"
      },
      "source": [
        "Compile the Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eMbQN-CzZ_pZ",
        "outputId": "cea50e6b-9a79-4ec2-e0a5-83cbf63b1ff7"
      },
      "source": [
        "optimizer = keras.optimizers.Adamax(learning_rate=0.001)\n",
        "model.compile(optimizer=optimizer, loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
        "history = model.fit(x_train, y_train, batch_size=32, epochs=12, validation_data=(x_val, y_val), callbacks=[checkPoint_cb])\n"
      ],
      "execution_count": 92,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/12\n",
            "1329/1329 [==============================] - 6s 4ms/step - loss: 0.5009 - accuracy: 0.8259 - val_loss: 0.8769 - val_accuracy: 0.7164\n",
            "Epoch 2/12\n",
            "1329/1329 [==============================] - 5s 4ms/step - loss: 0.4738 - accuracy: 0.8369 - val_loss: 0.8977 - val_accuracy: 0.7172\n",
            "Epoch 3/12\n",
            "1329/1329 [==============================] - 6s 4ms/step - loss: 0.4591 - accuracy: 0.8411 - val_loss: 0.8924 - val_accuracy: 0.7208\n",
            "Epoch 4/12\n",
            "1329/1329 [==============================] - 6s 4ms/step - loss: 0.4433 - accuracy: 0.8480 - val_loss: 0.9214 - val_accuracy: 0.7153\n",
            "Epoch 5/12\n",
            "1329/1329 [==============================] - 6s 4ms/step - loss: 0.4319 - accuracy: 0.8506 - val_loss: 0.9076 - val_accuracy: 0.7223\n",
            "Epoch 6/12\n",
            "1329/1329 [==============================] - 6s 4ms/step - loss: 0.4199 - accuracy: 0.8539 - val_loss: 0.9128 - val_accuracy: 0.7177\n",
            "Epoch 7/12\n",
            "1329/1329 [==============================] - 5s 4ms/step - loss: 0.4067 - accuracy: 0.8603 - val_loss: 0.9345 - val_accuracy: 0.7205\n",
            "Epoch 8/12\n",
            "1329/1329 [==============================] - 5s 4ms/step - loss: 0.3983 - accuracy: 0.8625 - val_loss: 0.9433 - val_accuracy: 0.7169\n",
            "Epoch 9/12\n",
            "1329/1329 [==============================] - 6s 4ms/step - loss: 0.3870 - accuracy: 0.8669 - val_loss: 0.9408 - val_accuracy: 0.7239\n",
            "Epoch 10/12\n",
            "1329/1329 [==============================] - 5s 4ms/step - loss: 0.3761 - accuracy: 0.8711 - val_loss: 0.9537 - val_accuracy: 0.7205\n",
            "Epoch 11/12\n",
            "1329/1329 [==============================] - 6s 4ms/step - loss: 0.3633 - accuracy: 0.8759 - val_loss: 0.9842 - val_accuracy: 0.7215\n",
            "Epoch 12/12\n",
            "1329/1329 [==============================] - 6s 4ms/step - loss: 0.3556 - accuracy: 0.8773 - val_loss: 0.9853 - val_accuracy: 0.7200\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kYOg-nnrWbcM"
      },
      "source": [
        "VIew the training and Validation loss & accuracy"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "0kZzXvskFdcs",
        "outputId": "1cf48193-6b75-4f35-d40e-0e6763b6fa8c"
      },
      "source": [
        "def show_plots(history):\n",
        "  \"\"\" Useful function to view plot of loss values & accuracies across the various epochs \"\"\"\n",
        "  loss_vals = history['loss']\n",
        "  val_loss_vals = history['val_loss']\n",
        "  epochs = range(1, len(history['accuracy'])+1)\n",
        "    \n",
        "  f, ax = plt.subplots(nrows=1,ncols=2,figsize=(16,4))\n",
        "    \n",
        "  # plot losses on ax[0]\n",
        "  ax[0].plot(epochs, loss_vals, color='navy',marker='o', linestyle=' ', label='Training Loss')\n",
        "  ax[0].plot(epochs, val_loss_vals, color='firebrick', marker='*', label='Validation Loss')\n",
        "  ax[0].set_title('Training & Validation Loss')\n",
        "  ax[0].set_xlabel('Epochs')\n",
        "  ax[0].set_ylabel('Loss')\n",
        "  ax[0].legend(loc='best')\n",
        "  ax[0].grid(True)\n",
        "    \n",
        "  # plot accuracies\n",
        "  acc_vals = history['accuracy']\n",
        "  val_acc_vals = history['val_accuracy']\n",
        "\n",
        "  ax[1].plot(epochs, acc_vals, color='navy', marker='o', ls=' ', label='Training Accuracy')\n",
        "  ax[1].plot(epochs, val_acc_vals, color='firebrick', marker='*', label='Validation Accuracy')\n",
        "  ax[1].set_title('Training & Validation Accuracy')\n",
        "  ax[1].set_xlabel('Epochs')\n",
        "  ax[1].set_ylabel('Accuracy')\n",
        "  ax[1].legend(loc='best')\n",
        "  ax[1].grid(True)\n",
        "    \n",
        "  plt.show()\n",
        "  plt.close()\n",
        "\n",
        "show_plots(history.history)"
      ],
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pU3p5foEWqu-"
      },
      "source": [
        "The chart above shows the model is almost overfitting as the training accuracy has a much higher variance than the validation accuracy."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9Qq_i7a9XG3O"
      },
      "source": [
        "Make some predictions"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 256
        },
        "id": "PnaMaREwXNWg",
        "outputId": "7a9e9088-b75b-4669-bf45-033c737338b5"
      },
      "source": [
        "#view the Test dataset\n",
        "test_data.head()"
      ],
      "execution_count": 94,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "      <th>12</th>\n",
              "      <th>13</th>\n",
              "      <th>14</th>\n",
              "      <th>15</th>\n",
              "      <th>16</th>\n",
              "      <th>17</th>\n",
              "      <th>18</th>\n",
              "      <th>19</th>\n",
              "      <th>20</th>\n",
              "      <th>21</th>\n",
              "      <th>22</th>\n",
              "      <th>23</th>\n",
              "      <th>24</th>\n",
              "      <th>25</th>\n",
              "      <th>26</th>\n",
              "      <th>27</th>\n",
              "      <th>28</th>\n",
              "      <th>29</th>\n",
              "      <th>30</th>\n",
              "      <th>31</th>\n",
              "      <th>32</th>\n",
              "      <th>33</th>\n",
              "      <th>34</th>\n",
              "      <th>35</th>\n",
              "      <th>36</th>\n",
              "      <th>37</th>\n",
              "      <th>38</th>\n",
              "      <th>39</th>\n",
              "      <th>...</th>\n",
              "      <th>3032</th>\n",
              "      <th>3033</th>\n",
              "      <th>3034</th>\n",
              "      <th>3035</th>\n",
              "      <th>3036</th>\n",
              "      <th>3037</th>\n",
              "      <th>3038</th>\n",
              "      <th>3039</th>\n",
              "      <th>3040</th>\n",
              "      <th>3041</th>\n",
              "      <th>3042</th>\n",
              "      <th>3043</th>\n",
              "      <th>3044</th>\n",
              "      <th>3045</th>\n",
              "      <th>3046</th>\n",
              "      <th>3047</th>\n",
              "      <th>3048</th>\n",
              "      <th>3049</th>\n",
              "      <th>3050</th>\n",
              "      <th>3051</th>\n",
              "      <th>3052</th>\n",
              "      <th>3053</th>\n",
              "      <th>3054</th>\n",
              "      <th>3055</th>\n",
              "      <th>3056</th>\n",
              "      <th>3057</th>\n",
              "      <th>3058</th>\n",
              "      <th>3059</th>\n",
              "      <th>3060</th>\n",
              "      <th>3061</th>\n",
              "      <th>3062</th>\n",
              "      <th>3063</th>\n",
              "      <th>3064</th>\n",
              "      <th>3065</th>\n",
              "      <th>3066</th>\n",
              "      <th>3067</th>\n",
              "      <th>3068</th>\n",
              "      <th>3069</th>\n",
              "      <th>3070</th>\n",
              "      <th>3071</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>98</td>\n",
              "      <td>105</td>\n",
              "      <td>108</td>\n",
              "      <td>92</td>\n",
              "      <td>101</td>\n",
              "      <td>106</td>\n",
              "      <td>91</td>\n",
              "      <td>101</td>\n",
              "      <td>107</td>\n",
              "      <td>93</td>\n",
              "      <td>104</td>\n",
              "      <td>108</td>\n",
              "      <td>89</td>\n",
              "      <td>100</td>\n",
              "      <td>104</td>\n",
              "      <td>88</td>\n",
              "      <td>99</td>\n",
              "      <td>103</td>\n",
              "      <td>90</td>\n",
              "      <td>101</td>\n",
              "      <td>106</td>\n",
              "      <td>95</td>\n",
              "      <td>105</td>\n",
              "      <td>110</td>\n",
              "      <td>99</td>\n",
              "      <td>108</td>\n",
              "      <td>110</td>\n",
              "      <td>109</td>\n",
              "      <td>117</td>\n",
              "      <td>116</td>\n",
              "      <td>119</td>\n",
              "      <td>125</td>\n",
              "      <td>123</td>\n",
              "      <td>125</td>\n",
              "      <td>130</td>\n",
              "      <td>128</td>\n",
              "      <td>134</td>\n",
              "      <td>140</td>\n",
              "      <td>138</td>\n",
              "      <td>140</td>\n",
              "      <td>...</td>\n",
              "      <td>184</td>\n",
              "      <td>178</td>\n",
              "      <td>177</td>\n",
              "      <td>172</td>\n",
              "      <td>190</td>\n",
              "      <td>188</td>\n",
              "      <td>184</td>\n",
              "      <td>220</td>\n",
              "      <td>219</td>\n",
              "      <td>214</td>\n",
              "      <td>195</td>\n",
              "      <td>193</td>\n",
              "      <td>187</td>\n",
              "      <td>176</td>\n",
              "      <td>174</td>\n",
              "      <td>169</td>\n",
              "      <td>175</td>\n",
              "      <td>173</td>\n",
              "      <td>168</td>\n",
              "      <td>192</td>\n",
              "      <td>189</td>\n",
              "      <td>185</td>\n",
              "      <td>185</td>\n",
              "      <td>182</td>\n",
              "      <td>178</td>\n",
              "      <td>168</td>\n",
              "      <td>165</td>\n",
              "      <td>161</td>\n",
              "      <td>178</td>\n",
              "      <td>176</td>\n",
              "      <td>171</td>\n",
              "      <td>183</td>\n",
              "      <td>182</td>\n",
              "      <td>176</td>\n",
              "      <td>175</td>\n",
              "      <td>175</td>\n",
              "      <td>168</td>\n",
              "      <td>181</td>\n",
              "      <td>181</td>\n",
              "      <td>175</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>101</td>\n",
              "      <td>108</td>\n",
              "      <td>101</td>\n",
              "      <td>101</td>\n",
              "      <td>108</td>\n",
              "      <td>101</td>\n",
              "      <td>102</td>\n",
              "      <td>109</td>\n",
              "      <td>102</td>\n",
              "      <td>103</td>\n",
              "      <td>110</td>\n",
              "      <td>103</td>\n",
              "      <td>105</td>\n",
              "      <td>112</td>\n",
              "      <td>104</td>\n",
              "      <td>105</td>\n",
              "      <td>114</td>\n",
              "      <td>109</td>\n",
              "      <td>105</td>\n",
              "      <td>116</td>\n",
              "      <td>112</td>\n",
              "      <td>106</td>\n",
              "      <td>115</td>\n",
              "      <td>110</td>\n",
              "      <td>108</td>\n",
              "      <td>115</td>\n",
              "      <td>108</td>\n",
              "      <td>107</td>\n",
              "      <td>115</td>\n",
              "      <td>107</td>\n",
              "      <td>104</td>\n",
              "      <td>115</td>\n",
              "      <td>106</td>\n",
              "      <td>104</td>\n",
              "      <td>116</td>\n",
              "      <td>108</td>\n",
              "      <td>108</td>\n",
              "      <td>119</td>\n",
              "      <td>115</td>\n",
              "      <td>108</td>\n",
              "      <td>...</td>\n",
              "      <td>149</td>\n",
              "      <td>153</td>\n",
              "      <td>159</td>\n",
              "      <td>164</td>\n",
              "      <td>132</td>\n",
              "      <td>140</td>\n",
              "      <td>142</td>\n",
              "      <td>134</td>\n",
              "      <td>144</td>\n",
              "      <td>145</td>\n",
              "      <td>112</td>\n",
              "      <td>119</td>\n",
              "      <td>118</td>\n",
              "      <td>108</td>\n",
              "      <td>115</td>\n",
              "      <td>111</td>\n",
              "      <td>114</td>\n",
              "      <td>124</td>\n",
              "      <td>122</td>\n",
              "      <td>106</td>\n",
              "      <td>118</td>\n",
              "      <td>116</td>\n",
              "      <td>101</td>\n",
              "      <td>113</td>\n",
              "      <td>108</td>\n",
              "      <td>100</td>\n",
              "      <td>111</td>\n",
              "      <td>102</td>\n",
              "      <td>99</td>\n",
              "      <td>110</td>\n",
              "      <td>103</td>\n",
              "      <td>100</td>\n",
              "      <td>109</td>\n",
              "      <td>104</td>\n",
              "      <td>100</td>\n",
              "      <td>109</td>\n",
              "      <td>103</td>\n",
              "      <td>100</td>\n",
              "      <td>109</td>\n",
              "      <td>102</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>85</td>\n",
              "      <td>115</td>\n",
              "      <td>27</td>\n",
              "      <td>63</td>\n",
              "      <td>90</td>\n",
              "      <td>25</td>\n",
              "      <td>37</td>\n",
              "      <td>66</td>\n",
              "      <td>15</td>\n",
              "      <td>69</td>\n",
              "      <td>93</td>\n",
              "      <td>49</td>\n",
              "      <td>40</td>\n",
              "      <td>52</td>\n",
              "      <td>36</td>\n",
              "      <td>18</td>\n",
              "      <td>24</td>\n",
              "      <td>18</td>\n",
              "      <td>31</td>\n",
              "      <td>38</td>\n",
              "      <td>27</td>\n",
              "      <td>79</td>\n",
              "      <td>91</td>\n",
              "      <td>61</td>\n",
              "      <td>104</td>\n",
              "      <td>117</td>\n",
              "      <td>78</td>\n",
              "      <td>72</td>\n",
              "      <td>84</td>\n",
              "      <td>52</td>\n",
              "      <td>68</td>\n",
              "      <td>79</td>\n",
              "      <td>52</td>\n",
              "      <td>122</td>\n",
              "      <td>131</td>\n",
              "      <td>85</td>\n",
              "      <td>104</td>\n",
              "      <td>113</td>\n",
              "      <td>67</td>\n",
              "      <td>77</td>\n",
              "      <td>...</td>\n",
              "      <td>139</td>\n",
              "      <td>174</td>\n",
              "      <td>188</td>\n",
              "      <td>139</td>\n",
              "      <td>182</td>\n",
              "      <td>198</td>\n",
              "      <td>150</td>\n",
              "      <td>179</td>\n",
              "      <td>194</td>\n",
              "      <td>147</td>\n",
              "      <td>175</td>\n",
              "      <td>186</td>\n",
              "      <td>142</td>\n",
              "      <td>187</td>\n",
              "      <td>195</td>\n",
              "      <td>158</td>\n",
              "      <td>176</td>\n",
              "      <td>191</td>\n",
              "      <td>142</td>\n",
              "      <td>169</td>\n",
              "      <td>190</td>\n",
              "      <td>132</td>\n",
              "      <td>177</td>\n",
              "      <td>193</td>\n",
              "      <td>144</td>\n",
              "      <td>180</td>\n",
              "      <td>189</td>\n",
              "      <td>147</td>\n",
              "      <td>175</td>\n",
              "      <td>190</td>\n",
              "      <td>141</td>\n",
              "      <td>172</td>\n",
              "      <td>193</td>\n",
              "      <td>136</td>\n",
              "      <td>173</td>\n",
              "      <td>192</td>\n",
              "      <td>138</td>\n",
              "      <td>179</td>\n",
              "      <td>192</td>\n",
              "      <td>149</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>213</td>\n",
              "      <td>213</td>\n",
              "      <td>214</td>\n",
              "      <td>215</td>\n",
              "      <td>214</td>\n",
              "      <td>218</td>\n",
              "      <td>220</td>\n",
              "      <td>218</td>\n",
              "      <td>226</td>\n",
              "      <td>223</td>\n",
              "      <td>221</td>\n",
              "      <td>233</td>\n",
              "      <td>226</td>\n",
              "      <td>225</td>\n",
              "      <td>231</td>\n",
              "      <td>225</td>\n",
              "      <td>226</td>\n",
              "      <td>223</td>\n",
              "      <td>223</td>\n",
              "      <td>223</td>\n",
              "      <td>223</td>\n",
              "      <td>227</td>\n",
              "      <td>226</td>\n",
              "      <td>231</td>\n",
              "      <td>229</td>\n",
              "      <td>228</td>\n",
              "      <td>236</td>\n",
              "      <td>229</td>\n",
              "      <td>228</td>\n",
              "      <td>236</td>\n",
              "      <td>230</td>\n",
              "      <td>229</td>\n",
              "      <td>236</td>\n",
              "      <td>228</td>\n",
              "      <td>227</td>\n",
              "      <td>235</td>\n",
              "      <td>227</td>\n",
              "      <td>226</td>\n",
              "      <td>234</td>\n",
              "      <td>225</td>\n",
              "      <td>...</td>\n",
              "      <td>20</td>\n",
              "      <td>111</td>\n",
              "      <td>41</td>\n",
              "      <td>28</td>\n",
              "      <td>111</td>\n",
              "      <td>63</td>\n",
              "      <td>60</td>\n",
              "      <td>120</td>\n",
              "      <td>92</td>\n",
              "      <td>98</td>\n",
              "      <td>138</td>\n",
              "      <td>117</td>\n",
              "      <td>129</td>\n",
              "      <td>153</td>\n",
              "      <td>140</td>\n",
              "      <td>158</td>\n",
              "      <td>172</td>\n",
              "      <td>161</td>\n",
              "      <td>187</td>\n",
              "      <td>188</td>\n",
              "      <td>174</td>\n",
              "      <td>205</td>\n",
              "      <td>195</td>\n",
              "      <td>182</td>\n",
              "      <td>212</td>\n",
              "      <td>199</td>\n",
              "      <td>185</td>\n",
              "      <td>216</td>\n",
              "      <td>197</td>\n",
              "      <td>186</td>\n",
              "      <td>216</td>\n",
              "      <td>193</td>\n",
              "      <td>194</td>\n",
              "      <td>209</td>\n",
              "      <td>201</td>\n",
              "      <td>204</td>\n",
              "      <td>216</td>\n",
              "      <td>203</td>\n",
              "      <td>201</td>\n",
              "      <td>237</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>41</td>\n",
              "      <td>74</td>\n",
              "      <td>144</td>\n",
              "      <td>41</td>\n",
              "      <td>75</td>\n",
              "      <td>139</td>\n",
              "      <td>41</td>\n",
              "      <td>75</td>\n",
              "      <td>139</td>\n",
              "      <td>41</td>\n",
              "      <td>74</td>\n",
              "      <td>144</td>\n",
              "      <td>39</td>\n",
              "      <td>76</td>\n",
              "      <td>140</td>\n",
              "      <td>37</td>\n",
              "      <td>79</td>\n",
              "      <td>130</td>\n",
              "      <td>38</td>\n",
              "      <td>77</td>\n",
              "      <td>135</td>\n",
              "      <td>40</td>\n",
              "      <td>75</td>\n",
              "      <td>142</td>\n",
              "      <td>43</td>\n",
              "      <td>72</td>\n",
              "      <td>145</td>\n",
              "      <td>47</td>\n",
              "      <td>71</td>\n",
              "      <td>146</td>\n",
              "      <td>47</td>\n",
              "      <td>73</td>\n",
              "      <td>140</td>\n",
              "      <td>43</td>\n",
              "      <td>75</td>\n",
              "      <td>140</td>\n",
              "      <td>39</td>\n",
              "      <td>77</td>\n",
              "      <td>143</td>\n",
              "      <td>37</td>\n",
              "      <td>...</td>\n",
              "      <td>149</td>\n",
              "      <td>42</td>\n",
              "      <td>71</td>\n",
              "      <td>135</td>\n",
              "      <td>43</td>\n",
              "      <td>72</td>\n",
              "      <td>125</td>\n",
              "      <td>42</td>\n",
              "      <td>72</td>\n",
              "      <td>122</td>\n",
              "      <td>51</td>\n",
              "      <td>72</td>\n",
              "      <td>117</td>\n",
              "      <td>64</td>\n",
              "      <td>70</td>\n",
              "      <td>109</td>\n",
              "      <td>65</td>\n",
              "      <td>72</td>\n",
              "      <td>99</td>\n",
              "      <td>53</td>\n",
              "      <td>76</td>\n",
              "      <td>113</td>\n",
              "      <td>35</td>\n",
              "      <td>75</td>\n",
              "      <td>136</td>\n",
              "      <td>30</td>\n",
              "      <td>81</td>\n",
              "      <td>140</td>\n",
              "      <td>34</td>\n",
              "      <td>79</td>\n",
              "      <td>133</td>\n",
              "      <td>41</td>\n",
              "      <td>77</td>\n",
              "      <td>130</td>\n",
              "      <td>44</td>\n",
              "      <td>75</td>\n",
              "      <td>133</td>\n",
              "      <td>42</td>\n",
              "      <td>73</td>\n",
              "      <td>144</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 3072 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     0    1    2    3    4    5    6  ...  3065  3066  3067  3068  3069  3070  3071\n",
              "0   98  105  108   92  101  106   91  ...   176   175   175   168   181   181   175\n",
              "1  101  108  101  101  108  101  102  ...   104   100   109   103   100   109   102\n",
              "2   85  115   27   63   90   25   37  ...   136   173   192   138   179   192   149\n",
              "3  213  213  214  215  214  218  220  ...   209   201   204   216   203   201   237\n",
              "4   41   74  144   41   75  139   41  ...   130    44    75   133    42    73   144\n",
              "\n",
              "[5 rows x 3072 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 94
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 256
        },
        "id": "6F9Tycq5XdiM",
        "outputId": "56373234-381a-4f82-bfee-99f2ee45129b"
      },
      "source": [
        "#normalize the test_data\n",
        "test_data = test_data/255.0\n",
        "test_data.head()"
      ],
      "execution_count": 95,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "      <th>12</th>\n",
              "      <th>13</th>\n",
              "      <th>14</th>\n",
              "      <th>15</th>\n",
              "      <th>16</th>\n",
              "      <th>17</th>\n",
              "      <th>18</th>\n",
              "      <th>19</th>\n",
              "      <th>20</th>\n",
              "      <th>21</th>\n",
              "      <th>22</th>\n",
              "      <th>23</th>\n",
              "      <th>24</th>\n",
              "      <th>25</th>\n",
              "      <th>26</th>\n",
              "      <th>27</th>\n",
              "      <th>28</th>\n",
              "      <th>29</th>\n",
              "      <th>30</th>\n",
              "      <th>31</th>\n",
              "      <th>32</th>\n",
              "      <th>33</th>\n",
              "      <th>34</th>\n",
              "      <th>35</th>\n",
              "      <th>36</th>\n",
              "      <th>37</th>\n",
              "      <th>38</th>\n",
              "      <th>39</th>\n",
              "      <th>...</th>\n",
              "      <th>3032</th>\n",
              "      <th>3033</th>\n",
              "      <th>3034</th>\n",
              "      <th>3035</th>\n",
              "      <th>3036</th>\n",
              "      <th>3037</th>\n",
              "      <th>3038</th>\n",
              "      <th>3039</th>\n",
              "      <th>3040</th>\n",
              "      <th>3041</th>\n",
              "      <th>3042</th>\n",
              "      <th>3043</th>\n",
              "      <th>3044</th>\n",
              "      <th>3045</th>\n",
              "      <th>3046</th>\n",
              "      <th>3047</th>\n",
              "      <th>3048</th>\n",
              "      <th>3049</th>\n",
              "      <th>3050</th>\n",
              "      <th>3051</th>\n",
              "      <th>3052</th>\n",
              "      <th>3053</th>\n",
              "      <th>3054</th>\n",
              "      <th>3055</th>\n",
              "      <th>3056</th>\n",
              "      <th>3057</th>\n",
              "      <th>3058</th>\n",
              "      <th>3059</th>\n",
              "      <th>3060</th>\n",
              "      <th>3061</th>\n",
              "      <th>3062</th>\n",
              "      <th>3063</th>\n",
              "      <th>3064</th>\n",
              "      <th>3065</th>\n",
              "      <th>3066</th>\n",
              "      <th>3067</th>\n",
              "      <th>3068</th>\n",
              "      <th>3069</th>\n",
              "      <th>3070</th>\n",
              "      <th>3071</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.384314</td>\n",
              "      <td>0.411765</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>0.360784</td>\n",
              "      <td>0.396078</td>\n",
              "      <td>0.415686</td>\n",
              "      <td>0.356863</td>\n",
              "      <td>0.396078</td>\n",
              "      <td>0.419608</td>\n",
              "      <td>0.364706</td>\n",
              "      <td>0.407843</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>0.349020</td>\n",
              "      <td>0.392157</td>\n",
              "      <td>0.407843</td>\n",
              "      <td>0.345098</td>\n",
              "      <td>0.388235</td>\n",
              "      <td>0.403922</td>\n",
              "      <td>0.352941</td>\n",
              "      <td>0.396078</td>\n",
              "      <td>0.415686</td>\n",
              "      <td>0.372549</td>\n",
              "      <td>0.411765</td>\n",
              "      <td>0.431373</td>\n",
              "      <td>0.388235</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>0.431373</td>\n",
              "      <td>0.427451</td>\n",
              "      <td>0.458824</td>\n",
              "      <td>0.454902</td>\n",
              "      <td>0.466667</td>\n",
              "      <td>0.490196</td>\n",
              "      <td>0.482353</td>\n",
              "      <td>0.490196</td>\n",
              "      <td>0.509804</td>\n",
              "      <td>0.501961</td>\n",
              "      <td>0.525490</td>\n",
              "      <td>0.549020</td>\n",
              "      <td>0.541176</td>\n",
              "      <td>0.549020</td>\n",
              "      <td>...</td>\n",
              "      <td>0.721569</td>\n",
              "      <td>0.698039</td>\n",
              "      <td>0.694118</td>\n",
              "      <td>0.674510</td>\n",
              "      <td>0.745098</td>\n",
              "      <td>0.737255</td>\n",
              "      <td>0.721569</td>\n",
              "      <td>0.862745</td>\n",
              "      <td>0.858824</td>\n",
              "      <td>0.839216</td>\n",
              "      <td>0.764706</td>\n",
              "      <td>0.756863</td>\n",
              "      <td>0.733333</td>\n",
              "      <td>0.690196</td>\n",
              "      <td>0.682353</td>\n",
              "      <td>0.662745</td>\n",
              "      <td>0.686275</td>\n",
              "      <td>0.678431</td>\n",
              "      <td>0.658824</td>\n",
              "      <td>0.752941</td>\n",
              "      <td>0.741176</td>\n",
              "      <td>0.725490</td>\n",
              "      <td>0.725490</td>\n",
              "      <td>0.713725</td>\n",
              "      <td>0.698039</td>\n",
              "      <td>0.658824</td>\n",
              "      <td>0.647059</td>\n",
              "      <td>0.631373</td>\n",
              "      <td>0.698039</td>\n",
              "      <td>0.690196</td>\n",
              "      <td>0.670588</td>\n",
              "      <td>0.717647</td>\n",
              "      <td>0.713725</td>\n",
              "      <td>0.690196</td>\n",
              "      <td>0.686275</td>\n",
              "      <td>0.686275</td>\n",
              "      <td>0.658824</td>\n",
              "      <td>0.709804</td>\n",
              "      <td>0.709804</td>\n",
              "      <td>0.686275</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.396078</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>0.396078</td>\n",
              "      <td>0.396078</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>0.396078</td>\n",
              "      <td>0.400000</td>\n",
              "      <td>0.427451</td>\n",
              "      <td>0.400000</td>\n",
              "      <td>0.403922</td>\n",
              "      <td>0.431373</td>\n",
              "      <td>0.403922</td>\n",
              "      <td>0.411765</td>\n",
              "      <td>0.439216</td>\n",
              "      <td>0.407843</td>\n",
              "      <td>0.411765</td>\n",
              "      <td>0.447059</td>\n",
              "      <td>0.427451</td>\n",
              "      <td>0.411765</td>\n",
              "      <td>0.454902</td>\n",
              "      <td>0.439216</td>\n",
              "      <td>0.415686</td>\n",
              "      <td>0.450980</td>\n",
              "      <td>0.431373</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>0.450980</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>0.419608</td>\n",
              "      <td>0.450980</td>\n",
              "      <td>0.419608</td>\n",
              "      <td>0.407843</td>\n",
              "      <td>0.450980</td>\n",
              "      <td>0.415686</td>\n",
              "      <td>0.407843</td>\n",
              "      <td>0.454902</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>0.466667</td>\n",
              "      <td>0.450980</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>...</td>\n",
              "      <td>0.584314</td>\n",
              "      <td>0.600000</td>\n",
              "      <td>0.623529</td>\n",
              "      <td>0.643137</td>\n",
              "      <td>0.517647</td>\n",
              "      <td>0.549020</td>\n",
              "      <td>0.556863</td>\n",
              "      <td>0.525490</td>\n",
              "      <td>0.564706</td>\n",
              "      <td>0.568627</td>\n",
              "      <td>0.439216</td>\n",
              "      <td>0.466667</td>\n",
              "      <td>0.462745</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>0.450980</td>\n",
              "      <td>0.435294</td>\n",
              "      <td>0.447059</td>\n",
              "      <td>0.486275</td>\n",
              "      <td>0.478431</td>\n",
              "      <td>0.415686</td>\n",
              "      <td>0.462745</td>\n",
              "      <td>0.454902</td>\n",
              "      <td>0.396078</td>\n",
              "      <td>0.443137</td>\n",
              "      <td>0.423529</td>\n",
              "      <td>0.392157</td>\n",
              "      <td>0.435294</td>\n",
              "      <td>0.400000</td>\n",
              "      <td>0.388235</td>\n",
              "      <td>0.431373</td>\n",
              "      <td>0.403922</td>\n",
              "      <td>0.392157</td>\n",
              "      <td>0.427451</td>\n",
              "      <td>0.407843</td>\n",
              "      <td>0.392157</td>\n",
              "      <td>0.427451</td>\n",
              "      <td>0.403922</td>\n",
              "      <td>0.392157</td>\n",
              "      <td>0.427451</td>\n",
              "      <td>0.400000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.333333</td>\n",
              "      <td>0.450980</td>\n",
              "      <td>0.105882</td>\n",
              "      <td>0.247059</td>\n",
              "      <td>0.352941</td>\n",
              "      <td>0.098039</td>\n",
              "      <td>0.145098</td>\n",
              "      <td>0.258824</td>\n",
              "      <td>0.058824</td>\n",
              "      <td>0.270588</td>\n",
              "      <td>0.364706</td>\n",
              "      <td>0.192157</td>\n",
              "      <td>0.156863</td>\n",
              "      <td>0.203922</td>\n",
              "      <td>0.141176</td>\n",
              "      <td>0.070588</td>\n",
              "      <td>0.094118</td>\n",
              "      <td>0.070588</td>\n",
              "      <td>0.121569</td>\n",
              "      <td>0.149020</td>\n",
              "      <td>0.105882</td>\n",
              "      <td>0.309804</td>\n",
              "      <td>0.356863</td>\n",
              "      <td>0.239216</td>\n",
              "      <td>0.407843</td>\n",
              "      <td>0.458824</td>\n",
              "      <td>0.305882</td>\n",
              "      <td>0.282353</td>\n",
              "      <td>0.329412</td>\n",
              "      <td>0.203922</td>\n",
              "      <td>0.266667</td>\n",
              "      <td>0.309804</td>\n",
              "      <td>0.203922</td>\n",
              "      <td>0.478431</td>\n",
              "      <td>0.513725</td>\n",
              "      <td>0.333333</td>\n",
              "      <td>0.407843</td>\n",
              "      <td>0.443137</td>\n",
              "      <td>0.262745</td>\n",
              "      <td>0.301961</td>\n",
              "      <td>...</td>\n",
              "      <td>0.545098</td>\n",
              "      <td>0.682353</td>\n",
              "      <td>0.737255</td>\n",
              "      <td>0.545098</td>\n",
              "      <td>0.713725</td>\n",
              "      <td>0.776471</td>\n",
              "      <td>0.588235</td>\n",
              "      <td>0.701961</td>\n",
              "      <td>0.760784</td>\n",
              "      <td>0.576471</td>\n",
              "      <td>0.686275</td>\n",
              "      <td>0.729412</td>\n",
              "      <td>0.556863</td>\n",
              "      <td>0.733333</td>\n",
              "      <td>0.764706</td>\n",
              "      <td>0.619608</td>\n",
              "      <td>0.690196</td>\n",
              "      <td>0.749020</td>\n",
              "      <td>0.556863</td>\n",
              "      <td>0.662745</td>\n",
              "      <td>0.745098</td>\n",
              "      <td>0.517647</td>\n",
              "      <td>0.694118</td>\n",
              "      <td>0.756863</td>\n",
              "      <td>0.564706</td>\n",
              "      <td>0.705882</td>\n",
              "      <td>0.741176</td>\n",
              "      <td>0.576471</td>\n",
              "      <td>0.686275</td>\n",
              "      <td>0.745098</td>\n",
              "      <td>0.552941</td>\n",
              "      <td>0.674510</td>\n",
              "      <td>0.756863</td>\n",
              "      <td>0.533333</td>\n",
              "      <td>0.678431</td>\n",
              "      <td>0.752941</td>\n",
              "      <td>0.541176</td>\n",
              "      <td>0.701961</td>\n",
              "      <td>0.752941</td>\n",
              "      <td>0.584314</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.835294</td>\n",
              "      <td>0.835294</td>\n",
              "      <td>0.839216</td>\n",
              "      <td>0.843137</td>\n",
              "      <td>0.839216</td>\n",
              "      <td>0.854902</td>\n",
              "      <td>0.862745</td>\n",
              "      <td>0.854902</td>\n",
              "      <td>0.886275</td>\n",
              "      <td>0.874510</td>\n",
              "      <td>0.866667</td>\n",
              "      <td>0.913725</td>\n",
              "      <td>0.886275</td>\n",
              "      <td>0.882353</td>\n",
              "      <td>0.905882</td>\n",
              "      <td>0.882353</td>\n",
              "      <td>0.886275</td>\n",
              "      <td>0.874510</td>\n",
              "      <td>0.874510</td>\n",
              "      <td>0.874510</td>\n",
              "      <td>0.874510</td>\n",
              "      <td>0.890196</td>\n",
              "      <td>0.886275</td>\n",
              "      <td>0.905882</td>\n",
              "      <td>0.898039</td>\n",
              "      <td>0.894118</td>\n",
              "      <td>0.925490</td>\n",
              "      <td>0.898039</td>\n",
              "      <td>0.894118</td>\n",
              "      <td>0.925490</td>\n",
              "      <td>0.901961</td>\n",
              "      <td>0.898039</td>\n",
              "      <td>0.925490</td>\n",
              "      <td>0.894118</td>\n",
              "      <td>0.890196</td>\n",
              "      <td>0.921569</td>\n",
              "      <td>0.890196</td>\n",
              "      <td>0.886275</td>\n",
              "      <td>0.917647</td>\n",
              "      <td>0.882353</td>\n",
              "      <td>...</td>\n",
              "      <td>0.078431</td>\n",
              "      <td>0.435294</td>\n",
              "      <td>0.160784</td>\n",
              "      <td>0.109804</td>\n",
              "      <td>0.435294</td>\n",
              "      <td>0.247059</td>\n",
              "      <td>0.235294</td>\n",
              "      <td>0.470588</td>\n",
              "      <td>0.360784</td>\n",
              "      <td>0.384314</td>\n",
              "      <td>0.541176</td>\n",
              "      <td>0.458824</td>\n",
              "      <td>0.505882</td>\n",
              "      <td>0.600000</td>\n",
              "      <td>0.549020</td>\n",
              "      <td>0.619608</td>\n",
              "      <td>0.674510</td>\n",
              "      <td>0.631373</td>\n",
              "      <td>0.733333</td>\n",
              "      <td>0.737255</td>\n",
              "      <td>0.682353</td>\n",
              "      <td>0.803922</td>\n",
              "      <td>0.764706</td>\n",
              "      <td>0.713725</td>\n",
              "      <td>0.831373</td>\n",
              "      <td>0.780392</td>\n",
              "      <td>0.725490</td>\n",
              "      <td>0.847059</td>\n",
              "      <td>0.772549</td>\n",
              "      <td>0.729412</td>\n",
              "      <td>0.847059</td>\n",
              "      <td>0.756863</td>\n",
              "      <td>0.760784</td>\n",
              "      <td>0.819608</td>\n",
              "      <td>0.788235</td>\n",
              "      <td>0.800000</td>\n",
              "      <td>0.847059</td>\n",
              "      <td>0.796078</td>\n",
              "      <td>0.788235</td>\n",
              "      <td>0.929412</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.160784</td>\n",
              "      <td>0.290196</td>\n",
              "      <td>0.564706</td>\n",
              "      <td>0.160784</td>\n",
              "      <td>0.294118</td>\n",
              "      <td>0.545098</td>\n",
              "      <td>0.160784</td>\n",
              "      <td>0.294118</td>\n",
              "      <td>0.545098</td>\n",
              "      <td>0.160784</td>\n",
              "      <td>0.290196</td>\n",
              "      <td>0.564706</td>\n",
              "      <td>0.152941</td>\n",
              "      <td>0.298039</td>\n",
              "      <td>0.549020</td>\n",
              "      <td>0.145098</td>\n",
              "      <td>0.309804</td>\n",
              "      <td>0.509804</td>\n",
              "      <td>0.149020</td>\n",
              "      <td>0.301961</td>\n",
              "      <td>0.529412</td>\n",
              "      <td>0.156863</td>\n",
              "      <td>0.294118</td>\n",
              "      <td>0.556863</td>\n",
              "      <td>0.168627</td>\n",
              "      <td>0.282353</td>\n",
              "      <td>0.568627</td>\n",
              "      <td>0.184314</td>\n",
              "      <td>0.278431</td>\n",
              "      <td>0.572549</td>\n",
              "      <td>0.184314</td>\n",
              "      <td>0.286275</td>\n",
              "      <td>0.549020</td>\n",
              "      <td>0.168627</td>\n",
              "      <td>0.294118</td>\n",
              "      <td>0.549020</td>\n",
              "      <td>0.152941</td>\n",
              "      <td>0.301961</td>\n",
              "      <td>0.560784</td>\n",
              "      <td>0.145098</td>\n",
              "      <td>...</td>\n",
              "      <td>0.584314</td>\n",
              "      <td>0.164706</td>\n",
              "      <td>0.278431</td>\n",
              "      <td>0.529412</td>\n",
              "      <td>0.168627</td>\n",
              "      <td>0.282353</td>\n",
              "      <td>0.490196</td>\n",
              "      <td>0.164706</td>\n",
              "      <td>0.282353</td>\n",
              "      <td>0.478431</td>\n",
              "      <td>0.200000</td>\n",
              "      <td>0.282353</td>\n",
              "      <td>0.458824</td>\n",
              "      <td>0.250980</td>\n",
              "      <td>0.274510</td>\n",
              "      <td>0.427451</td>\n",
              "      <td>0.254902</td>\n",
              "      <td>0.282353</td>\n",
              "      <td>0.388235</td>\n",
              "      <td>0.207843</td>\n",
              "      <td>0.298039</td>\n",
              "      <td>0.443137</td>\n",
              "      <td>0.137255</td>\n",
              "      <td>0.294118</td>\n",
              "      <td>0.533333</td>\n",
              "      <td>0.117647</td>\n",
              "      <td>0.317647</td>\n",
              "      <td>0.549020</td>\n",
              "      <td>0.133333</td>\n",
              "      <td>0.309804</td>\n",
              "      <td>0.521569</td>\n",
              "      <td>0.160784</td>\n",
              "      <td>0.301961</td>\n",
              "      <td>0.509804</td>\n",
              "      <td>0.172549</td>\n",
              "      <td>0.294118</td>\n",
              "      <td>0.521569</td>\n",
              "      <td>0.164706</td>\n",
              "      <td>0.286275</td>\n",
              "      <td>0.564706</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 3072 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "          0         1         2  ...      3069      3070      3071\n",
              "0  0.384314  0.411765  0.423529  ...  0.709804  0.709804  0.686275\n",
              "1  0.396078  0.423529  0.396078  ...  0.392157  0.427451  0.400000\n",
              "2  0.333333  0.450980  0.105882  ...  0.701961  0.752941  0.584314\n",
              "3  0.835294  0.835294  0.839216  ...  0.796078  0.788235  0.929412\n",
              "4  0.160784  0.290196  0.564706  ...  0.164706  0.286275  0.564706\n",
              "\n",
              "[5 rows x 3072 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 95
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LF6xp8iJXw9V",
        "outputId": "d5e6ada4-7bb3-4345-b4ae-43778227468a"
      },
      "source": [
        "#36 * 36 * 3 = 3072\n",
        "test_data.shape"
      ],
      "execution_count": 96,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2000, 3072)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 96
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BOb0-qFmkyrv",
        "outputId": "3a86f6b3-f11c-4fc1-87a9-6438886fdb91"
      },
      "source": [
        "#we have to reshape the test_data in order to make predictions\n",
        "x_test = test_data.values.reshape((2000, 32, 32, 3))\n",
        "x_test.shape"
      ],
      "execution_count": 105,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2000, 32, 32, 3)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 105
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xpnjBtolW-6C",
        "outputId": "72001409-6f0f-4627-fa94-4eb7e275e43b"
      },
      "source": [
        "prediction = model.predict_classes(x_test)\n",
        "prediction[:10]"
      ],
      "execution_count": 108,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([2, 3, 4, 8, 6, 1, 3, 7, 3, 5])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 108
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GoqEBLeZn4Vs"
      },
      "source": [
        "Map the predictions back to their respective classes"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LdJkIycnn86T",
        "outputId": "54bd6e55-e7f2-4c9e-c0b3-1dba4767bf4c"
      },
      "source": [
        "#create a dictionary to map them\n",
        "label_dict = {i: labels[i] for i in range(len(labels))}\n",
        "label_dict"
      ],
      "execution_count": 120,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{0: 'Airplane',\n",
              " 1: 'Automobile',\n",
              " 2: 'Bird',\n",
              " 3: 'Cat',\n",
              " 4: 'Deer',\n",
              " 5: 'Dog',\n",
              " 6: 'Frog',\n",
              " 7: 'Horse',\n",
              " 8: 'Ship',\n",
              " 9: 'Truck'}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 120
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vlS1kbQwuflE"
      },
      "source": [
        "#convert the predicted numbers back to labels\n",
        "pred = [label_dict.get(i) for i in prediction]"
      ],
      "execution_count": 127,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "TX9u_JDZzWW2",
        "outputId": "d40af48d-b55f-467f-afcd-609b3900f408"
      },
      "source": [
        "p = pd.DataFrame(pred, columns=['prediction'])\n",
        "p.head()"
      ],
      "execution_count": 130,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>prediction</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Bird</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Cat</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Deer</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Ship</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Frog</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  prediction\n",
              "0       Bird\n",
              "1        Cat\n",
              "2       Deer\n",
              "3       Ship\n",
              "4       Frog"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 130
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tyw0yQ561C-g"
      },
      "source": [
        "Save the predictions for upload"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "id": "u5GDWeuF1B3x",
        "outputId": "99584df6-855c-4f12-a76c-d4c31c3a611a"
      },
      "source": [
        "# To download the csv file locally\n",
        "from google.colab import files \n",
        "\n",
        "p.to_csv('predictions.csv')\n",
        "files.download('predictions.csv')"
      ],
      "execution_count": 131,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ],
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "application/javascript": [
              "download(\"download_2103fc73-7040-4c63-b09e-d5476563f6ad\", \"predictions.csv\", 21056)"
            ],
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "id": "6FRWmbej2MYD",
        "outputId": "2da17a4c-611a-4540-e10d-b0cadf20c45a"
      },
      "source": [
        "#incase we want to save the numbers\n",
        "p1 = pd.DataFrame(pred, columns=['prediction'])\n",
        "p1.to_csv('pred.csv')\n",
        "files.download('pred.csv')"
      ],
      "execution_count": 133,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/javascript": [
              "\n",
              "    async function download(id, filename, size) {\n",
              "      if (!google.colab.kernel.accessAllowed) {\n",
              "        return;\n",
              "      }\n",
              "      const div = document.createElement('div');\n",
              "      const label = document.createElement('label');\n",
              "      label.textContent = `Downloading \"${filename}\": `;\n",
              "      div.appendChild(label);\n",
              "      const progress = document.createElement('progress');\n",
              "      progress.max = size;\n",
              "      div.appendChild(progress);\n",
              "      document.body.appendChild(div);\n",
              "\n",
              "      const buffers = [];\n",
              "      let downloaded = 0;\n",
              "\n",
              "      const channel = await google.colab.kernel.comms.open(id);\n",
              "      // Send a message to notify the kernel that we're ready.\n",
              "      channel.send({})\n",
              "\n",
              "      for await (const message of channel.messages) {\n",
              "        // Send a message to notify the kernel that we're ready.\n",
              "        channel.send({})\n",
              "        if (message.buffers) {\n",
              "          for (const buffer of message.buffers) {\n",
              "            buffers.push(buffer);\n",
              "            downloaded += buffer.byteLength;\n",
              "            progress.value = downloaded;\n",
              "          }\n",
              "        }\n",
              "      }\n",
              "      const blob = new Blob(buffers, {type: 'application/binary'});\n",
              "      const a = document.createElement('a');\n",
              "      a.href = window.URL.createObjectURL(blob);\n",
              "      a.download = filename;\n",
              "      div.appendChild(a);\n",
              "      a.click();\n",
              "      div.remove();\n",
              "    }\n",
              "  "
            ],
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "application/javascript": [
              "download(\"download_0f7f695a-5049-4f32-b4e5-f4b39cd4f1eb\", \"pred.csv\", 21056)"
            ],
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}