MNIST_Classification.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aISVhqcrnanH"
      },
      "source": [
        "# Multi- class Classification with MNIST Dataset of Handwritten Digits."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "9n9_cTveKmse"
      },
      "outputs": [],
      "source": [
        "# Import relevant modules\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras import layers\n",
        "from matplotlib import pyplot as plt\n",
        "\n",
        "# The following lines adjust the granularity of reporting.\n",
        "pd.options.display.max_rows = 10\n",
        "pd.options.display.float_format = \"{:.1f}\".format\n",
        "\n",
        "# The following line improves formatting when ouputting NumPy arrays.\n",
        "np.set_printoptions(linewidth = 200)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I7BTXA9ammkR"
      },
      "source": [
        "## **Using Deep Neural Network(DNN)**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JZlvdpyYKx7V",
        "outputId": "b78e2e2c-1261-4246-f35f-05f8fab05c79"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "11490434/11490434 [==============================] - 0s 0us/step\n"
          ]
        }
      ],
      "source": [
        "# Load Dataset\n",
        "(x_train, y_train),(x_test, y_test) = tf.keras.datasets.mnist.load_data()\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NKBQTUJU-LKx"
      },
      "source": [
        "### View Dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "id": "IoOhpjkeCL8Q",
        "outputId": "13987dd4-8f77-480c-9837-a101a555c91d"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<style>\n",
              "      .ndarray_repr .ndarray_raw_data {\n",
              "        display: none;\n",
              "      }\n",
              "      .ndarray_repr.show_array .ndarray_raw_data {\n",
              "        display: block;\n",
              "      }\n",
              "      .ndarray_repr.show_array .ndarray_image_preview {\n",
              "        display: none;\n",
              "      }\n",
              "      </style>\n",
              "      <div id=\"id-da3e8439-e67a-4ff0-9093-6e1f72f70809\" class=\"ndarray_repr\"><pre>ndarray (28, 28) <button style=\"padding: 0 2px;\">show data</button></pre><img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABwAAAAcCAAAAABXZoBIAAAA1ElEQVR4nGNgGEqAEcaof2oR/PGiyqnlxz9jKMr79Q8KdnBhSN6Ayf37bQYXZILS2q3H3brTQ7U97jObYeiEA6t/N7jRdcKBCYMaLw59LMHf/x3AdBEDAwMDg2rNv3+/9LFKFW769e/fvz+bGoww5Vz/wHwzE9NBHgwMnz8z/PyH3UaXIFbWIFXH5f/24XAuAwPDImzGQgEj1/9NODWm/ZuBU07t2Q0RHFKM7vf+9eHS5/Tv31wswixKSn4LH/38N4sNUy5hx79///79+xnLitM1pAAACDhcuN3dt1kAAAAASUVORK5CYII=\" class=\"ndarray_image_preview\" /><pre class=\"ndarray_raw_data\">array([[  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0, 127, 100, 156, 239, 224, 177, 213, 159,  70,  13,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0, 110, 250, 254, 254, 254, 254, 254, 254, 254, 254, 184,  10,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0, 216, 254, 254, 254, 254, 254, 254, 254, 254, 254, 251,  54,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   3, 131, 197,  68, 137, 101,  83,  41,  70, 221, 254, 108,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  58, 254, 216,  11,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0, 110, 254, 254,  24,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   4,  83, 247, 254, 192,  10,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  37, 124, 254, 254, 250,  47,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0, 113, 178, 250, 254, 254, 254, 252, 178, 128,  50,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  69, 252, 254, 254, 254, 254, 254, 254, 254, 254, 153,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0, 141, 252, 254, 241, 241, 254, 247, 252, 254, 254, 153,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  68,  82,   5,   5,  82,  37,  65, 167, 254, 190,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0, 162, 254, 153,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   1, 172, 253,  75,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0, 102, 254, 152,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  38, 230, 216,  20,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   1,  71, 222, 254, 142,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  66, 254, 254, 157,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   4,  34,  34,  78, 161, 226, 249, 254, 154,   6,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,  96, 184, 254, 254, 254, 254, 254, 249,  93,   5,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0]], dtype=uint8)</pre></div><script>\n",
              "      (() => {\n",
              "      const titles = ['show data', 'hide data'];\n",
              "      let index = 0\n",
              "      document.querySelector('#id-da3e8439-e67a-4ff0-9093-6e1f72f70809 button').onclick = (e) => {\n",
              "        document.querySelector('#id-da3e8439-e67a-4ff0-9093-6e1f72f70809').classList.toggle('show_array');\n",
              "        index = (++index) % 2;\n",
              "        document.querySelector('#id-da3e8439-e67a-4ff0-9093-6e1f72f70809 button').textContent = titles[index];\n",
              "        e.preventDefault();\n",
              "        e.stopPropagation();\n",
              "      }\n",
              "      })();\n",
              "    </script>"
            ],
            "text/plain": [
              "array([[  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0, 127, 100, 156, 239, 224, 177, 213, 159,  70,  13,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0, 110, 250, 254, 254, 254, 254, 254, 254, 254, 254, 184,  10,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0, 216, 254, 254, 254, 254, 254, 254, 254, 254, 254, 251,  54,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   3, 131, 197,  68, 137, 101,  83,  41,  70, 221, 254, 108,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  58, 254, 216,  11,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0, 110, 254, 254,  24,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   4,  83, 247, 254, 192,  10,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  37, 124, 254, 254, 250,  47,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0, 113, 178, 250, 254, 254, 254, 252, 178, 128,  50,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  69, 252, 254, 254, 254, 254, 254, 254, 254, 254, 153,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0, 141, 252, 254, 241, 241, 254, 247, 252, 254, 254, 153,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  68,  82,   5,   5,  82,  37,  65, 167, 254, 190,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0, 162, 254, 153,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   1, 172, 253,  75,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0, 102, 254, 152,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  38, 230, 216,  20,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   1,  71, 222, 254, 142,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  66, 254, 254, 157,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   4,  34,  34,  78, 161, 226, 249, 254, 154,   6,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,  96, 184, 254, 254, 254, 254, 254, 249,  93,   5,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0],\n",
              "       [  0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0]], dtype=uint8)"
            ]
          },
          "execution_count": 6,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Output example #2917 of the training set.\n",
        "x_train[2917]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "znoqcdl52vSn",
        "outputId": "85f05d0c-ce95-48d0-9f1d-86ecb35ba37a"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(60000, 28, 28)"
            ]
          },
          "execution_count": 7,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "x_train.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ji1Fmf0qnHO_",
        "outputId": "dca892ad-84e7-4041-b994-ea58a1050ff4"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "7\n"
          ]
        }
      ],
      "source": [
        "print(y_train[5000])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 448
        },
        "id": "siRC8a1hJvmq",
        "outputId": "e4bde85a-ac95-460c-e578-3d9601640d3f"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x78608a4f3040>"
            ]
          },
          "execution_count": 13,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.imshow(x_train[5000])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "V-he9IcihDxb",
        "outputId": "bdc44d96-3173-4ce8-c36d-d20b6dd6ce38"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "array([  0,   0, 226, 254, 208, 199, 199, 199, 199, 139,  61,  61,  61,  61,  61, 128, 222, 254, 254, 189,  21,   0,   0,   0,   0,   0,   0,   0], dtype=uint8)"
            ]
          },
          "execution_count": 12,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Output row #10 of example #2000.\n",
        "x_train[5000][10]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DUEWipalhQ8J",
        "outputId": "ba093eec-6086-4183-d2e3-e2c53603ad25"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "222"
            ]
          },
          "execution_count": 11,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Output pixel #16 of row #10 of example #5000.\n",
        "x_train[5000][10][16]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0NHtGehlrkTu"
      },
      "source": [
        "**Indicates almost Balanced data**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zntsGKWo2h5d",
        "outputId": "78cf2efc-f186-4cad-fc1f-78b04a04515b"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Class 0: 5923 samples\n",
            "Class 1: 6742 samples\n",
            "Class 2: 5958 samples\n",
            "Class 3: 6131 samples\n",
            "Class 4: 5842 samples\n",
            "Class 5: 5421 samples\n",
            "Class 6: 5918 samples\n",
            "Class 7: 6265 samples\n",
            "Class 8: 5851 samples\n",
            "Class 9: 5949 samples\n"
          ]
        }
      ],
      "source": [
        "unique_labels, counts = np.unique(y_train, return_counts=True)\n",
        "\n",
        "# Print unique labels and their counts\n",
        "for label, count in zip(unique_labels, counts):\n",
        "    print(f\"Class {label}: {count} samples\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ghM-BJY7EeSF",
        "outputId": "10d342d0-8c73-44b1-b909-6305a7cdbdfe"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " rescaling (Rescaling)       (None, 28, 28, 1)         0         \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 784)               0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 256)               200960    \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 128)               32896     \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 128)               0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 10)                1290      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 235146 (918.54 KB)\n",
            "Trainable params: 235146 (918.54 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Flatten\n",
        "from keras.optimizers import Adam\n",
        "\n",
        "# Define your model\n",
        "model = Sequential()\n",
        "model.add(tf.keras.layers.experimental.preprocessing.Rescaling(1./255., input_shape=(28, 28, 1)))\n",
        "model.add(Flatten(input_shape=(28, 28, 1)))  # Example input shape for MNIST dataset\n",
        "model.add(Dense(256, activation='relu'))\n",
        "model.add(Dense(128, activation='relu'))\n",
        "model.add(tf.keras.layers.Dropout(rate=0.2))\n",
        "model.add(Dense(10, activation='softmax'))  # Output layer with 10 classes\n",
        "\n",
        "model.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "cellView": "both",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "nj3v5EKQFY8s",
        "outputId": "79128258-fc9e-4844-9457-6639bb01dc6c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "12/12 [==============================] - 2s 119ms/step - loss: 1.0590 - accuracy: 0.6911 - val_loss: 0.3465 - val_accuracy: 0.8985\n",
            "Epoch 2/100\n",
            "12/12 [==============================] - 1s 101ms/step - loss: 0.3564 - accuracy: 0.8946 - val_loss: 0.2374 - val_accuracy: 0.9314\n",
            "Epoch 3/100\n",
            "12/12 [==============================] - 1s 108ms/step - loss: 0.2457 - accuracy: 0.9284 - val_loss: 0.1863 - val_accuracy: 0.9457\n",
            "Epoch 4/100\n",
            "12/12 [==============================] - 1s 103ms/step - loss: 0.1899 - accuracy: 0.9437 - val_loss: 0.1535 - val_accuracy: 0.9558\n",
            "Epoch 5/100\n",
            "12/12 [==============================] - 1s 104ms/step - loss: 0.1551 - accuracy: 0.9544 - val_loss: 0.1347 - val_accuracy: 0.9599\n",
            "Epoch 6/100\n",
            "12/12 [==============================] - 3s 220ms/step - loss: 0.1305 - accuracy: 0.9612 - val_loss: 0.1218 - val_accuracy: 0.9631\n",
            "Epoch 7/100\n",
            "12/12 [==============================] - 2s 194ms/step - loss: 0.1124 - accuracy: 0.9662 - val_loss: 0.1169 - val_accuracy: 0.9659\n",
            "Epoch 8/100\n",
            "12/12 [==============================] - 2s 170ms/step - loss: 0.0997 - accuracy: 0.9711 - val_loss: 0.1042 - val_accuracy: 0.9695\n",
            "Epoch 9/100\n",
            "12/12 [==============================] - 3s 238ms/step - loss: 0.0839 - accuracy: 0.9756 - val_loss: 0.0999 - val_accuracy: 0.9698\n",
            "Epoch 10/100\n",
            "12/12 [==============================] - 3s 264ms/step - loss: 0.0739 - accuracy: 0.9781 - val_loss: 0.0928 - val_accuracy: 0.9713\n",
            "Epoch 11/100\n",
            "12/12 [==============================] - 3s 246ms/step - loss: 0.0653 - accuracy: 0.9815 - val_loss: 0.0892 - val_accuracy: 0.9725\n",
            "Epoch 12/100\n",
            "12/12 [==============================] - 2s 180ms/step - loss: 0.0576 - accuracy: 0.9827 - val_loss: 0.0907 - val_accuracy: 0.9726\n",
            "Epoch 13/100\n",
            "12/12 [==============================] - 2s 180ms/step - loss: 0.0519 - accuracy: 0.9847 - val_loss: 0.0846 - val_accuracy: 0.9740\n",
            "Epoch 14/100\n",
            "12/12 [==============================] - 2s 137ms/step - loss: 0.0454 - accuracy: 0.9868 - val_loss: 0.0843 - val_accuracy: 0.9746\n",
            "Epoch 15/100\n",
            "12/12 [==============================] - 1s 98ms/step - loss: 0.0412 - accuracy: 0.9878 - val_loss: 0.0823 - val_accuracy: 0.9742\n",
            "Epoch 16/100\n",
            "12/12 [==============================] - 1s 96ms/step - loss: 0.0371 - accuracy: 0.9895 - val_loss: 0.0800 - val_accuracy: 0.9755\n",
            "Epoch 17/100\n",
            "12/12 [==============================] - 1s 96ms/step - loss: 0.0328 - accuracy: 0.9905 - val_loss: 0.0787 - val_accuracy: 0.9763\n",
            "Epoch 18/100\n",
            "12/12 [==============================] - 1s 105ms/step - loss: 0.0289 - accuracy: 0.9918 - val_loss: 0.0784 - val_accuracy: 0.9762\n",
            "Epoch 19/100\n",
            "12/12 [==============================] - 2s 184ms/step - loss: 0.0257 - accuracy: 0.9928 - val_loss: 0.0789 - val_accuracy: 0.9772\n",
            "Epoch 20/100\n",
            "12/12 [==============================] - 3s 210ms/step - loss: 0.0219 - accuracy: 0.9940 - val_loss: 0.0783 - val_accuracy: 0.9776\n",
            "Epoch 21/100\n",
            "12/12 [==============================] - 1s 102ms/step - loss: 0.0201 - accuracy: 0.9950 - val_loss: 0.0796 - val_accuracy: 0.9774\n",
            "Epoch 22/100\n",
            "12/12 [==============================] - 3s 219ms/step - loss: 0.0175 - accuracy: 0.9954 - val_loss: 0.0800 - val_accuracy: 0.9774\n",
            "Epoch 23/100\n",
            "12/12 [==============================] - 2s 203ms/step - loss: 0.0164 - accuracy: 0.9954 - val_loss: 0.0793 - val_accuracy: 0.9776\n",
            "Epoch 24/100\n",
            "12/12 [==============================] - 3s 246ms/step - loss: 0.0158 - accuracy: 0.9957 - val_loss: 0.0833 - val_accuracy: 0.9784\n",
            "Epoch 25/100\n",
            "12/12 [==============================] - 3s 246ms/step - loss: 0.0139 - accuracy: 0.9966 - val_loss: 0.0809 - val_accuracy: 0.9774\n",
            "Epoch 26/100\n",
            "12/12 [==============================] - 3s 224ms/step - loss: 0.0120 - accuracy: 0.9971 - val_loss: 0.0809 - val_accuracy: 0.9781\n",
            "Epoch 27/100\n",
            "12/12 [==============================] - 2s 131ms/step - loss: 0.0106 - accuracy: 0.9975 - val_loss: 0.0850 - val_accuracy: 0.9782\n",
            "Epoch 28/100\n",
            "12/12 [==============================] - 1s 104ms/step - loss: 0.0099 - accuracy: 0.9977 - val_loss: 0.0830 - val_accuracy: 0.9778\n",
            "Epoch 29/100\n",
            "12/12 [==============================] - 1s 98ms/step - loss: 0.0090 - accuracy: 0.9980 - val_loss: 0.0841 - val_accuracy: 0.9783\n",
            "Epoch 30/100\n",
            "12/12 [==============================] - 1s 99ms/step - loss: 0.0091 - accuracy: 0.9977 - val_loss: 0.0842 - val_accuracy: 0.9789\n",
            "Epoch 31/100\n",
            "12/12 [==============================] - 1s 102ms/step - loss: 0.0086 - accuracy: 0.9980 - val_loss: 0.0854 - val_accuracy: 0.9783\n",
            "Epoch 32/100\n",
            "12/12 [==============================] - 2s 158ms/step - loss: 0.0088 - accuracy: 0.9977 - val_loss: 0.0864 - val_accuracy: 0.9779\n",
            "Epoch 33/100\n",
            "12/12 [==============================] - 1s 121ms/step - loss: 0.0074 - accuracy: 0.9982 - val_loss: 0.0855 - val_accuracy: 0.9786\n",
            "Epoch 34/100\n",
            "12/12 [==============================] - 1s 100ms/step - loss: 0.0058 - accuracy: 0.9988 - val_loss: 0.0867 - val_accuracy: 0.9787\n",
            "Epoch 35/100\n",
            "12/12 [==============================] - 1s 103ms/step - loss: 0.0056 - accuracy: 0.9989 - val_loss: 0.0870 - val_accuracy: 0.9793\n",
            "Epoch 36/100\n",
            "12/12 [==============================] - 1s 103ms/step - loss: 0.0047 - accuracy: 0.9992 - val_loss: 0.0876 - val_accuracy: 0.9783\n",
            "Epoch 37/100\n",
            "12/12 [==============================] - 1s 105ms/step - loss: 0.0044 - accuracy: 0.9993 - val_loss: 0.0882 - val_accuracy: 0.9788\n",
            "Epoch 38/100\n",
            "12/12 [==============================] - 1s 103ms/step - loss: 0.0042 - accuracy: 0.9993 - val_loss: 0.0915 - val_accuracy: 0.9790\n",
            "Epoch 39/100\n",
            "12/12 [==============================] - 1s 101ms/step - loss: 0.0045 - accuracy: 0.9992 - val_loss: 0.0905 - val_accuracy: 0.9787\n",
            "Epoch 40/100\n",
            "12/12 [==============================] - 1s 97ms/step - loss: 0.0038 - accuracy: 0.9995 - val_loss: 0.0900 - val_accuracy: 0.9790\n",
            "Epoch 41/100\n",
            "12/12 [==============================] - 1s 128ms/step - loss: 0.0031 - accuracy: 0.9997 - val_loss: 0.0927 - val_accuracy: 0.9787\n",
            "Epoch 42/100\n",
            "12/12 [==============================] - 2s 171ms/step - loss: 0.0032 - accuracy: 0.9994 - val_loss: 0.0934 - val_accuracy: 0.9791\n",
            "Epoch 43/100\n",
            "12/12 [==============================] - 1s 105ms/step - loss: 0.0031 - accuracy: 0.9996 - val_loss: 0.0940 - val_accuracy: 0.9784\n",
            "Epoch 44/100\n",
            "12/12 [==============================] - 1s 105ms/step - loss: 0.0031 - accuracy: 0.9995 - val_loss: 0.0933 - val_accuracy: 0.9788\n",
            "Epoch 45/100\n",
            "12/12 [==============================] - 1s 104ms/step - loss: 0.0028 - accuracy: 0.9996 - val_loss: 0.0929 - val_accuracy: 0.9792\n",
            "Epoch 46/100\n",
            "12/12 [==============================] - 1s 100ms/step - loss: 0.0024 - accuracy: 0.9998 - val_loss: 0.0965 - val_accuracy: 0.9788\n",
            "Epoch 47/100\n",
            "12/12 [==============================] - 1s 99ms/step - loss: 0.0024 - accuracy: 0.9996 - val_loss: 0.0936 - val_accuracy: 0.9791\n",
            "Epoch 48/100\n",
            "12/12 [==============================] - 1s 103ms/step - loss: 0.0023 - accuracy: 0.9997 - val_loss: 0.0956 - val_accuracy: 0.9791\n",
            "Epoch 49/100\n",
            "12/12 [==============================] - 1s 101ms/step - loss: 0.0022 - accuracy: 0.9998 - val_loss: 0.0980 - val_accuracy: 0.9791\n",
            "Epoch 50/100\n",
            "12/12 [==============================] - 1s 105ms/step - loss: 0.0022 - accuracy: 0.9997 - val_loss: 0.0969 - val_accuracy: 0.9784\n",
            "Epoch 51/100\n",
            "12/12 [==============================] - 2s 160ms/step - loss: 0.0021 - accuracy: 0.9997 - val_loss: 0.0962 - val_accuracy: 0.9795\n",
            "Epoch 52/100\n",
            "12/12 [==============================] - 2s 134ms/step - loss: 0.0019 - accuracy: 0.9997 - val_loss: 0.0989 - val_accuracy: 0.9790\n",
            "Epoch 53/100\n",
            "12/12 [==============================] - 1s 106ms/step - loss: 0.0020 - accuracy: 0.9997 - val_loss: 0.1043 - val_accuracy: 0.9781\n",
            "Epoch 54/100\n",
            "12/12 [==============================] - 1s 105ms/step - loss: 0.0019 - accuracy: 0.9997 - val_loss: 0.0997 - val_accuracy: 0.9789\n",
            "Epoch 55/100\n",
            "12/12 [==============================] - 1s 101ms/step - loss: 0.0018 - accuracy: 0.9998 - val_loss: 0.0998 - val_accuracy: 0.9784\n",
            "Epoch 56/100\n",
            "12/12 [==============================] - 1s 99ms/step - loss: 0.0017 - accuracy: 0.9998 - val_loss: 0.1016 - val_accuracy: 0.9798\n",
            "Epoch 57/100\n",
            "12/12 [==============================] - 1s 103ms/step - loss: 0.0017 - accuracy: 0.9999 - val_loss: 0.0997 - val_accuracy: 0.9786\n",
            "Epoch 58/100\n",
            "12/12 [==============================] - 1s 108ms/step - loss: 0.0014 - accuracy: 0.9999 - val_loss: 0.1027 - val_accuracy: 0.9791\n",
            "Epoch 59/100\n",
            "12/12 [==============================] - 1s 104ms/step - loss: 0.0013 - accuracy: 0.9999 - val_loss: 0.1021 - val_accuracy: 0.9788\n",
            "Epoch 60/100\n",
            "12/12 [==============================] - 1s 123ms/step - loss: 0.0014 - accuracy: 0.9998 - val_loss: 0.1018 - val_accuracy: 0.9789\n",
            "Epoch 61/100\n",
            "12/12 [==============================] - 2s 175ms/step - loss: 0.0013 - accuracy: 0.9998 - val_loss: 0.1002 - val_accuracy: 0.9792\n",
            "Epoch 62/100\n",
            "12/12 [==============================] - 1s 98ms/step - loss: 0.0015 - accuracy: 0.9998 - val_loss: 0.1078 - val_accuracy: 0.9781\n",
            "Epoch 63/100\n",
            "12/12 [==============================] - 1s 103ms/step - loss: 0.0017 - accuracy: 0.9997 - val_loss: 0.1088 - val_accuracy: 0.9782\n",
            "Epoch 64/100\n",
            "12/12 [==============================] - 1s 107ms/step - loss: 0.0014 - accuracy: 0.9998 - val_loss: 0.1047 - val_accuracy: 0.9782\n",
            "Epoch 65/100\n",
            "12/12 [==============================] - 1s 107ms/step - loss: 0.0013 - accuracy: 0.9999 - val_loss: 0.1074 - val_accuracy: 0.9793\n",
            "Epoch 66/100\n",
            "12/12 [==============================] - 1s 104ms/step - loss: 0.0011 - accuracy: 0.9999 - val_loss: 0.1045 - val_accuracy: 0.9793\n",
            "Epoch 67/100\n",
            "12/12 [==============================] - 1s 109ms/step - loss: 0.0013 - accuracy: 0.9998 - val_loss: 0.1091 - val_accuracy: 0.9795\n",
            "Epoch 68/100\n",
            "12/12 [==============================] - 1s 103ms/step - loss: 0.0012 - accuracy: 0.9999 - val_loss: 0.1074 - val_accuracy: 0.9786\n",
            "Epoch 69/100\n",
            "12/12 [==============================] - 1s 107ms/step - loss: 0.0016 - accuracy: 0.9997 - val_loss: 0.1066 - val_accuracy: 0.9784\n",
            "Epoch 70/100\n",
            "12/12 [==============================] - 2s 179ms/step - loss: 0.0016 - accuracy: 0.9998 - val_loss: 0.1088 - val_accuracy: 0.9784\n",
            "Epoch 71/100\n",
            "12/12 [==============================] - 1s 120ms/step - loss: 0.0017 - accuracy: 0.9996 - val_loss: 0.1066 - val_accuracy: 0.9801\n",
            "Epoch 72/100\n",
            "12/12 [==============================] - 1s 100ms/step - loss: 0.0012 - accuracy: 0.9999 - val_loss: 0.1096 - val_accuracy: 0.9788\n",
            "Epoch 73/100\n",
            "12/12 [==============================] - 1s 101ms/step - loss: 0.0011 - accuracy: 0.9999 - val_loss: 0.1049 - val_accuracy: 0.9795\n",
            "Epoch 74/100\n",
            "12/12 [==============================] - 1s 102ms/step - loss: 0.0013 - accuracy: 0.9998 - val_loss: 0.1076 - val_accuracy: 0.9791\n",
            "Epoch 75/100\n",
            "12/12 [==============================] - 1s 98ms/step - loss: 0.0012 - accuracy: 0.9999 - val_loss: 0.1102 - val_accuracy: 0.9793\n",
            "Epoch 76/100\n",
            "12/12 [==============================] - 1s 104ms/step - loss: 0.0015 - accuracy: 0.9996 - val_loss: 0.1086 - val_accuracy: 0.9775\n",
            "Epoch 77/100\n",
            "12/12 [==============================] - 1s 109ms/step - loss: 0.0019 - accuracy: 0.9996 - val_loss: 0.1125 - val_accuracy: 0.9782\n",
            "Epoch 78/100\n",
            "12/12 [==============================] - 1s 107ms/step - loss: 0.0018 - accuracy: 0.9996 - val_loss: 0.1102 - val_accuracy: 0.9794\n",
            "Epoch 79/100\n",
            "12/12 [==============================] - 2s 145ms/step - loss: 0.0021 - accuracy: 0.9994 - val_loss: 0.1169 - val_accuracy: 0.9769\n",
            "Epoch 80/100\n",
            "12/12 [==============================] - 2s 160ms/step - loss: 0.0024 - accuracy: 0.9994 - val_loss: 0.1123 - val_accuracy: 0.9773\n",
            "Epoch 81/100\n",
            "12/12 [==============================] - 1s 112ms/step - loss: 0.0018 - accuracy: 0.9997 - val_loss: 0.1065 - val_accuracy: 0.9794\n",
            "Epoch 82/100\n",
            "12/12 [==============================] - 2s 146ms/step - loss: 0.0016 - accuracy: 0.9997 - val_loss: 0.1097 - val_accuracy: 0.9793\n",
            "Epoch 83/100\n",
            "12/12 [==============================] - 1s 99ms/step - loss: 0.0012 - accuracy: 0.9998 - val_loss: 0.1118 - val_accuracy: 0.9793\n",
            "Epoch 84/100\n",
            "12/12 [==============================] - 1s 101ms/step - loss: 9.3054e-04 - accuracy: 0.9999 - val_loss: 0.1154 - val_accuracy: 0.9785\n",
            "Epoch 85/100\n",
            "12/12 [==============================] - 1s 98ms/step - loss: 9.9046e-04 - accuracy: 0.9998 - val_loss: 0.1165 - val_accuracy: 0.9788\n",
            "Epoch 86/100\n",
            "12/12 [==============================] - 1s 95ms/step - loss: 0.0010 - accuracy: 0.9998 - val_loss: 0.1156 - val_accuracy: 0.9768\n",
            "Epoch 87/100\n",
            "12/12 [==============================] - 1s 109ms/step - loss: 0.0012 - accuracy: 0.9998 - val_loss: 0.1209 - val_accuracy: 0.9778\n",
            "Epoch 88/100\n",
            "12/12 [==============================] - 2s 137ms/step - loss: 9.3191e-04 - accuracy: 0.9998 - val_loss: 0.1138 - val_accuracy: 0.9787\n",
            "Epoch 89/100\n",
            "12/12 [==============================] - 2s 167ms/step - loss: 9.3108e-04 - accuracy: 0.9999 - val_loss: 0.1196 - val_accuracy: 0.9783\n",
            "Epoch 90/100\n",
            "12/12 [==============================] - 1s 101ms/step - loss: 0.0011 - accuracy: 0.9998 - val_loss: 0.1128 - val_accuracy: 0.9793\n",
            "Epoch 91/100\n",
            "12/12 [==============================] - 1s 106ms/step - loss: 0.0010 - accuracy: 0.9998 - val_loss: 0.1127 - val_accuracy: 0.9787\n",
            "Epoch 92/100\n",
            "12/12 [==============================] - 1s 101ms/step - loss: 0.0011 - accuracy: 0.9999 - val_loss: 0.1138 - val_accuracy: 0.9794\n",
            "Epoch 93/100\n",
            "12/12 [==============================] - 1s 98ms/step - loss: 0.0010 - accuracy: 0.9998 - val_loss: 0.1246 - val_accuracy: 0.9784\n",
            "Epoch 94/100\n",
            "12/12 [==============================] - 1s 100ms/step - loss: 0.0015 - accuracy: 0.9996 - val_loss: 0.1171 - val_accuracy: 0.9778\n",
            "Epoch 95/100\n",
            "12/12 [==============================] - 1s 105ms/step - loss: 0.0015 - accuracy: 0.9996 - val_loss: 0.1164 - val_accuracy: 0.9774\n",
            "Epoch 96/100\n",
            "12/12 [==============================] - 1s 104ms/step - loss: 0.0015 - accuracy: 0.9996 - val_loss: 0.1173 - val_accuracy: 0.9783\n",
            "Epoch 97/100\n",
            "12/12 [==============================] - 1s 102ms/step - loss: 0.0013 - accuracy: 0.9997 - val_loss: 0.1218 - val_accuracy: 0.9778\n",
            "Epoch 98/100\n",
            "12/12 [==============================] - 2s 161ms/step - loss: 0.0011 - accuracy: 0.9998 - val_loss: 0.1185 - val_accuracy: 0.9795\n",
            "Epoch 99/100\n",
            "12/12 [==============================] - 2s 129ms/step - loss: 8.4951e-04 - accuracy: 0.9999 - val_loss: 0.1181 - val_accuracy: 0.9795\n",
            "Epoch 100/100\n",
            "12/12 [==============================] - 1s 99ms/step - loss: 6.3584e-04 - accuracy: 0.9999 - val_loss: 0.1192 - val_accuracy: 0.9791\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1500x300 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# The following variables are the hyperparameters.\n",
        "learning_rate = 0.003\n",
        "epochs = 100\n",
        "batch_size = 4000\n",
        "validation_split = 0.2\n",
        "\n",
        "model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=learning_rate),\n",
        "            loss=\"sparse_categorical_crossentropy\",\n",
        "            metrics=['accuracy'])\n",
        "\n",
        "# Train the model.\n",
        "history = model.fit(x=x_train, y=y_train, batch_size=batch_size,\n",
        "                    epochs=epochs, shuffle=True,\n",
        "                    validation_split=validation_split)\n",
        "\n",
        "# To track the progression of training, gather a snapshot\n",
        "# of the model's metrics at each epoch.\n",
        "epochs = history.epoch\n",
        "# hist = pd.DataFrame(history.history)\n",
        "\n",
        "# Plot a graph of the metric vs. epochs.\n",
        "\n",
        "plt.figure(figsize=(15, 3))\n",
        "\n",
        "# Train Loss vs Validation Loss\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(history.history[model.metrics_names[0]], label='Train Loss')\n",
        "plt.plot(history.history[f'val_{model.metrics_names[0]}'], label='Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "\n",
        "# Train Accuracy vs Validation Accuracy\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(history.history[model.metrics_names[1]], label='Train Accuracy')\n",
        "plt.plot(history.history[f'val_{model.metrics_names[1]}'], label='Validation Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.legend()\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rYqBuRdvscoV"
      },
      "source": [
        "Somewhat overfitting with 100 epochs"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Zjf71cKYsyej",
        "outputId": "09600486-0ddf-4b7a-c1a5-ab37ef29342b"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "model.save('/content/drive/MyDrive/MLMarch/100_model.h5')\n",
        "\n",
        "# model.save('./models/100_model.h5') # For saving it locally"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 28,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7CaqP5YTEQaG",
        "outputId": "d71a4371-9400-4162-d85f-6daf36ea19e6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n",
            "\n",
            " Evaluate the new model against the test set:\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.1070 - accuracy: 0.9797\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "[0.10700371116399765, 0.9797000288963318]"
            ]
          },
          "execution_count": 28,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# from tensorflow.keras.models import load_model\n",
        "# from google.colab import drive\n",
        "# drive.mount('/content/drive')\n",
        "# model = load_model('/content/drive/MyDrive/MLMarch/100_model.h5')\n",
        "\n",
        "# Evaluate against the test set.\n",
        "\n",
        "print(\"\\n Evaluate the new model against the test set:\")\n",
        "model.evaluate(x=x_test, y=y_test, batch_size=batch_size)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 655
        },
        "id": "EeNh8yb18NS7",
        "outputId": "418d1e6a-9170-4a5e-b61a-cbe434516799"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 1s 2ms/step\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "Text(95.72222222222221, 0.5, 'Truth')"
            ]
          },
          "execution_count": 20,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1000x700 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import seaborn as sn\n",
        "y_predicted = model.predict(x_test)\n",
        "y_predicted_labels = [np.argmax(i) for i in y_predicted]\n",
        "cm = tf.math.confusion_matrix(labels=y_test,predictions=y_predicted_labels)\n",
        "\n",
        "plt.figure(figsize = (10,7))\n",
        "sn.heatmap(cm, annot=True, fmt='d')\n",
        "plt.xlabel('Predicted')\n",
        "plt.ylabel('Truth')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mC5Vl9ys_3w6",
        "outputId": "1bca47d7-3efb-411c-8d34-70251bf769c6"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "array([[2.3779655e-11, 4.9085274e-09, 3.7788103e-10, ..., 9.9999994e-01, 2.2014523e-12, 4.5727269e-10],\n",
              "       [4.0631430e-18, 1.6763364e-12, 9.9999994e-01, ..., 2.1872372e-21, 3.8732015e-20, 5.9075538e-26],\n",
              "       [2.2156001e-10, 9.9999994e-01, 6.4413014e-10, ..., 1.9172968e-09, 1.5348697e-08, 5.8842501e-12],\n",
              "       ...,\n",
              "       [2.5596686e-20, 3.5313913e-15, 3.0741720e-22, ..., 7.0845253e-14, 8.2694040e-18, 2.0392783e-13],\n",
              "       [1.5772588e-15, 3.4802479e-17, 5.2663653e-19, ..., 1.5340644e-16, 3.8038000e-11, 6.1281167e-18],\n",
              "       [9.9389215e-16, 1.2345499e-19, 1.1429231e-18, ..., 5.0552514e-27, 4.1235719e-18, 9.6526420e-21]], dtype=float32)"
            ]
          },
          "execution_count": 21,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "y_predicted"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "f0yVbdz3_juu",
        "outputId": "a9d6c05a-02ef-4086-a161-07dfa1825444"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "__Classification Report__ [For Normal Deep Neural Net without any augmentation]:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.98      0.99      0.99       980\n",
            "           1       0.99      0.99      0.99      1135\n",
            "           2       0.98      0.98      0.98      1032\n",
            "           3       0.98      0.98      0.98      1010\n",
            "           4       0.98      0.98      0.98       982\n",
            "           5       0.97      0.98      0.98       892\n",
            "           6       0.99      0.98      0.98       958\n",
            "           7       0.99      0.97      0.98      1028\n",
            "           8       0.98      0.96      0.97       974\n",
            "           9       0.97      0.98      0.98      1009\n",
            "\n",
            "    accuracy                           0.98     10000\n",
            "   macro avg       0.98      0.98      0.98     10000\n",
            "weighted avg       0.98      0.98      0.98     10000\n",
            "\n"
          ]
        }
      ],
      "source": [
        "from sklearn.metrics import classification_report\n",
        "# Generate classification report\n",
        "report = classification_report(y_test, y_predicted_labels)\n",
        "\n",
        "print(\"__Classification Report__ [For Normal Deep Neural Net without any augmentation]:\")\n",
        "print(report)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "W4rf_9OSHcAP"
      },
      "source": [
        "## **Using Data Augmentation with DNN**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DXDHT4j0HoM5",
        "outputId": "86277397-676a-4df9-a0a5-6eb0c92882c9"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "11490434/11490434 [==============================] - 0s 0us/step\n"
          ]
        }
      ],
      "source": [
        "# Importing again otherwise Normalization will happen twice\n",
        "(x_train, y_train),(x_test, y_test) = tf.keras.datasets.mnist.load_data()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7q9OhUcJGWGd",
        "outputId": "e4962803-9b15-4b80-ace3-e9b8bfbd5ec7"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(60000, 28, 28, 1)"
            ]
          },
          "execution_count": 3,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "x_train = np.expand_dims(x_train, axis=-1)\n",
        "x_train.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "IZkw7zieMYet"
      },
      "outputs": [],
      "source": [
        "# Train and validation split\n",
        "from sklearn.model_selection import train_test_split\n",
        "x_train, x_val, y_train, y_val = train_test_split(x_train, y_train, test_size=0.2)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "YSlEBCMZDndA"
      },
      "outputs": [],
      "source": [
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "batch_size  = 4000\n",
        "# Create an ImageDataGenerator instance with augmentation parameters\n",
        "datagen = ImageDataGenerator(\n",
        "    rotation_range=10,\n",
        "    width_shift_range=0.1,\n",
        "    height_shift_range=0.1,\n",
        "    shear_range=0.1,\n",
        "    zoom_range=0.2,\n",
        "    fill_mode='nearest')\n",
        "\n",
        "# Fit the ImageDataGenerator to your dataset\n",
        "datagen.fit(x_train)\n",
        "datagen.fit(x_val)\n",
        "# Generate augmented data\n",
        "train_generator = datagen.flow(x_train, y_train, batch_size=batch_size)\n",
        "val_generator = datagen.flow(x_val, y_val, batch_size=batch_size)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 595
        },
        "id": "iXqSHGVrBtqu",
        "outputId": "69576ffc-f16a-459d-dd5e-42dfc5fa87c9"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1500x300 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1500x300 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Plot some augmented images\n",
        "\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def plot_aug_images(generator, title):\n",
        "    fig, ax = plt.subplots(nrows=1, ncols=5, figsize=(15, 3))\n",
        "    fig.suptitle(title)\n",
        "\n",
        "    for i in range(5):\n",
        "        # Get a batch of images and labels from the generator\n",
        "        batch = next(generator)\n",
        "        images = batch[0]\n",
        "        labels_batch = batch[1]\n",
        "\n",
        "        # Pick one image from the batch\n",
        "        image = images[i]\n",
        "\n",
        "        # Rescale the image to the original range [0, 255]\n",
        "        # image = (image * 255).astype('uint8')\n",
        "\n",
        "        label = labels_batch[i]\n",
        "\n",
        "        # Plot raw pixel data\n",
        "        ax[i].imshow(image)\n",
        "        #ax[i].axis('off')\n",
        "        ax[i].set_title(label)\n",
        "\n",
        "    plt.show()\n",
        "\n",
        "# Plot 5 augmented train images\n",
        "plot_aug_images(train_generator, \"Augmented Training Images\")\n",
        "plot_aug_images(train_generator, \"Augmented Validation Images\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uOIE9u-iHUNd",
        "outputId": "ca45db29-a07a-4af7-db4e-cabb54ed6dcb"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " rescaling (Rescaling)       (None, 28, 28, 1)         0         \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 784)               0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 256)               200960    \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 128)               32896     \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 128)               0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 10)                1290      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 235146 (918.54 KB)\n",
            "Trainable params: 235146 (918.54 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Flatten\n",
        "from keras.optimizers import Adam\n",
        "\n",
        "# Define your model [Using SAME model architecture as above]\n",
        "model_aug = Sequential()\n",
        "model_aug.add(tf.keras.layers.experimental.preprocessing.Rescaling(1./255., input_shape=(28, 28, 1)))\n",
        "model_aug.add(Flatten(input_shape=(28, 28, 1)))  # Example input shape for MNIST dataset\n",
        "model_aug.add(Dense(256, activation='relu'))\n",
        "model_aug.add(Dense(128, activation='relu'))\n",
        "model_aug.add(tf.keras.layers.Dropout(rate=0.2))\n",
        "model_aug.add(Dense(10, activation='softmax'))  # Output layer with 10 classes\n",
        "\n",
        "model_aug.summary()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "S3gDoC_DRbkW",
        "outputId": "83b03ad4-0171-4549-af7e-641d1bfc2b45"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "12/12 [==============================] - 19s 1s/step - loss: 1.8807 - accuracy: 0.3821 - val_loss: 1.3992 - val_accuracy: 0.5741\n",
            "Epoch 2/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 1.2852 - accuracy: 0.5873 - val_loss: 1.0403 - val_accuracy: 0.6856\n",
            "Epoch 3/100\n",
            "12/12 [==============================] - 39s 3s/step - loss: 1.0034 - accuracy: 0.6897 - val_loss: 0.8292 - val_accuracy: 0.7572\n",
            "Epoch 4/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.8075 - accuracy: 0.7595 - val_loss: 0.6620 - val_accuracy: 0.8150\n",
            "Epoch 5/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.6606 - accuracy: 0.8035 - val_loss: 0.5389 - val_accuracy: 0.8467\n",
            "Epoch 6/100\n",
            "12/12 [==============================] - 25s 2s/step - loss: 0.5489 - accuracy: 0.8403 - val_loss: 0.4457 - val_accuracy: 0.8783\n",
            "Epoch 7/100\n",
            "12/12 [==============================] - 18s 1s/step - loss: 0.4732 - accuracy: 0.8618 - val_loss: 0.3848 - val_accuracy: 0.8920\n",
            "Epoch 8/100\n",
            "12/12 [==============================] - 23s 2s/step - loss: 0.4144 - accuracy: 0.8792 - val_loss: 0.3428 - val_accuracy: 0.9053\n",
            "Epoch 9/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.3670 - accuracy: 0.8925 - val_loss: 0.3045 - val_accuracy: 0.9122\n",
            "Epoch 10/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.3363 - accuracy: 0.9024 - val_loss: 0.2812 - val_accuracy: 0.9212\n",
            "Epoch 11/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.3129 - accuracy: 0.9090 - val_loss: 0.2627 - val_accuracy: 0.9251\n",
            "Epoch 12/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 0.2901 - accuracy: 0.9150 - val_loss: 0.2464 - val_accuracy: 0.9283\n",
            "Epoch 13/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 0.2728 - accuracy: 0.9220 - val_loss: 0.2419 - val_accuracy: 0.9293\n",
            "Epoch 14/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.2612 - accuracy: 0.9235 - val_loss: 0.2306 - val_accuracy: 0.9325\n",
            "Epoch 15/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.2462 - accuracy: 0.9282 - val_loss: 0.2122 - val_accuracy: 0.9390\n",
            "Epoch 16/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.2365 - accuracy: 0.9315 - val_loss: 0.1995 - val_accuracy: 0.9433\n",
            "Epoch 17/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.2284 - accuracy: 0.9330 - val_loss: 0.2007 - val_accuracy: 0.9433\n",
            "Epoch 18/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 0.2180 - accuracy: 0.9364 - val_loss: 0.1846 - val_accuracy: 0.9462\n",
            "Epoch 19/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.2098 - accuracy: 0.9386 - val_loss: 0.1832 - val_accuracy: 0.9458\n",
            "Epoch 20/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.2043 - accuracy: 0.9406 - val_loss: 0.1822 - val_accuracy: 0.9467\n",
            "Epoch 21/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1973 - accuracy: 0.9424 - val_loss: 0.1729 - val_accuracy: 0.9472\n",
            "Epoch 22/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1923 - accuracy: 0.9434 - val_loss: 0.1694 - val_accuracy: 0.9481\n",
            "Epoch 23/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 0.1866 - accuracy: 0.9453 - val_loss: 0.1652 - val_accuracy: 0.9500\n",
            "Epoch 24/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1825 - accuracy: 0.9466 - val_loss: 0.1651 - val_accuracy: 0.9523\n",
            "Epoch 25/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.1785 - accuracy: 0.9476 - val_loss: 0.1557 - val_accuracy: 0.9535\n",
            "Epoch 26/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1741 - accuracy: 0.9486 - val_loss: 0.1536 - val_accuracy: 0.9549\n",
            "Epoch 27/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1710 - accuracy: 0.9495 - val_loss: 0.1543 - val_accuracy: 0.9549\n",
            "Epoch 28/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 0.1664 - accuracy: 0.9507 - val_loss: 0.1464 - val_accuracy: 0.9574\n",
            "Epoch 29/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1617 - accuracy: 0.9535 - val_loss: 0.1489 - val_accuracy: 0.9585\n",
            "Epoch 30/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 0.1596 - accuracy: 0.9538 - val_loss: 0.1383 - val_accuracy: 0.9596\n",
            "Epoch 31/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1530 - accuracy: 0.9551 - val_loss: 0.1449 - val_accuracy: 0.9579\n",
            "Epoch 32/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 0.1514 - accuracy: 0.9550 - val_loss: 0.1391 - val_accuracy: 0.9575\n",
            "Epoch 33/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1471 - accuracy: 0.9560 - val_loss: 0.1346 - val_accuracy: 0.9613\n",
            "Epoch 34/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1506 - accuracy: 0.9551 - val_loss: 0.1382 - val_accuracy: 0.9585\n",
            "Epoch 35/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.1473 - accuracy: 0.9577 - val_loss: 0.1269 - val_accuracy: 0.9630\n",
            "Epoch 36/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1381 - accuracy: 0.9586 - val_loss: 0.1264 - val_accuracy: 0.9599\n",
            "Epoch 37/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1385 - accuracy: 0.9586 - val_loss: 0.1245 - val_accuracy: 0.9604\n",
            "Epoch 38/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1372 - accuracy: 0.9592 - val_loss: 0.1245 - val_accuracy: 0.9611\n",
            "Epoch 39/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.1355 - accuracy: 0.9597 - val_loss: 0.1221 - val_accuracy: 0.9628\n",
            "Epoch 40/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 0.1358 - accuracy: 0.9595 - val_loss: 0.1271 - val_accuracy: 0.9606\n",
            "Epoch 41/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1324 - accuracy: 0.9601 - val_loss: 0.1232 - val_accuracy: 0.9638\n",
            "Epoch 42/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1286 - accuracy: 0.9615 - val_loss: 0.1215 - val_accuracy: 0.9622\n",
            "Epoch 43/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1273 - accuracy: 0.9620 - val_loss: 0.1140 - val_accuracy: 0.9657\n",
            "Epoch 44/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1261 - accuracy: 0.9629 - val_loss: 0.1170 - val_accuracy: 0.9657\n",
            "Epoch 45/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1246 - accuracy: 0.9629 - val_loss: 0.1139 - val_accuracy: 0.9655\n",
            "Epoch 46/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.1199 - accuracy: 0.9640 - val_loss: 0.1148 - val_accuracy: 0.9658\n",
            "Epoch 47/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1230 - accuracy: 0.9637 - val_loss: 0.1075 - val_accuracy: 0.9663\n",
            "Epoch 48/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1181 - accuracy: 0.9649 - val_loss: 0.1131 - val_accuracy: 0.9652\n",
            "Epoch 49/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1162 - accuracy: 0.9648 - val_loss: 0.1086 - val_accuracy: 0.9664\n",
            "Epoch 50/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1149 - accuracy: 0.9659 - val_loss: 0.1092 - val_accuracy: 0.9663\n",
            "Epoch 51/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1176 - accuracy: 0.9647 - val_loss: 0.1048 - val_accuracy: 0.9669\n",
            "Epoch 52/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1149 - accuracy: 0.9653 - val_loss: 0.1028 - val_accuracy: 0.9686\n",
            "Epoch 53/100\n",
            "12/12 [==============================] - 28s 2s/step - loss: 0.1156 - accuracy: 0.9652 - val_loss: 0.0997 - val_accuracy: 0.9697\n",
            "Epoch 54/100\n",
            "12/12 [==============================] - 26s 2s/step - loss: 0.1109 - accuracy: 0.9667 - val_loss: 0.1037 - val_accuracy: 0.9688\n",
            "Epoch 55/100\n",
            "12/12 [==============================] - 19s 2s/step - loss: 0.1104 - accuracy: 0.9662 - val_loss: 0.0980 - val_accuracy: 0.9698\n",
            "Epoch 56/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1099 - accuracy: 0.9671 - val_loss: 0.1030 - val_accuracy: 0.9694\n",
            "Epoch 57/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1086 - accuracy: 0.9672 - val_loss: 0.1003 - val_accuracy: 0.9703\n",
            "Epoch 58/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1052 - accuracy: 0.9683 - val_loss: 0.0986 - val_accuracy: 0.9695\n",
            "Epoch 59/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1066 - accuracy: 0.9673 - val_loss: 0.0996 - val_accuracy: 0.9691\n",
            "Epoch 60/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1082 - accuracy: 0.9671 - val_loss: 0.0954 - val_accuracy: 0.9693\n",
            "Epoch 61/100\n",
            "12/12 [==============================] - 19s 2s/step - loss: 0.1050 - accuracy: 0.9682 - val_loss: 0.0992 - val_accuracy: 0.9687\n",
            "Epoch 62/100\n",
            "12/12 [==============================] - 30s 3s/step - loss: 0.1030 - accuracy: 0.9683 - val_loss: 0.0942 - val_accuracy: 0.9701\n",
            "Epoch 63/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1054 - accuracy: 0.9680 - val_loss: 0.0972 - val_accuracy: 0.9691\n",
            "Epoch 64/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1011 - accuracy: 0.9694 - val_loss: 0.0997 - val_accuracy: 0.9696\n",
            "Epoch 65/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1024 - accuracy: 0.9693 - val_loss: 0.0959 - val_accuracy: 0.9688\n",
            "Epoch 66/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0976 - accuracy: 0.9701 - val_loss: 0.0947 - val_accuracy: 0.9704\n",
            "Epoch 67/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0980 - accuracy: 0.9703 - val_loss: 0.0927 - val_accuracy: 0.9713\n",
            "Epoch 68/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0948 - accuracy: 0.9707 - val_loss: 0.0887 - val_accuracy: 0.9712\n",
            "Epoch 69/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0986 - accuracy: 0.9706 - val_loss: 0.0898 - val_accuracy: 0.9718\n",
            "Epoch 70/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0978 - accuracy: 0.9701 - val_loss: 0.0944 - val_accuracy: 0.9702\n",
            "Epoch 71/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0967 - accuracy: 0.9702 - val_loss: 0.0889 - val_accuracy: 0.9722\n",
            "Epoch 72/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0952 - accuracy: 0.9717 - val_loss: 0.0907 - val_accuracy: 0.9714\n",
            "Epoch 73/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0942 - accuracy: 0.9713 - val_loss: 0.0882 - val_accuracy: 0.9743\n",
            "Epoch 74/100\n",
            "12/12 [==============================] - 24s 2s/step - loss: 0.0909 - accuracy: 0.9719 - val_loss: 0.0876 - val_accuracy: 0.9728\n",
            "Epoch 75/100\n",
            "12/12 [==============================] - 22s 2s/step - loss: 0.0926 - accuracy: 0.9718 - val_loss: 0.0878 - val_accuracy: 0.9730\n",
            "Epoch 76/100\n",
            "12/12 [==============================] - 20s 2s/step - loss: 0.0925 - accuracy: 0.9716 - val_loss: 0.0812 - val_accuracy: 0.9753\n",
            "Epoch 77/100\n",
            "12/12 [==============================] - 25s 2s/step - loss: 0.0898 - accuracy: 0.9728 - val_loss: 0.0898 - val_accuracy: 0.9719\n",
            "Epoch 78/100\n",
            "12/12 [==============================] - 20s 2s/step - loss: 0.0907 - accuracy: 0.9717 - val_loss: 0.0932 - val_accuracy: 0.9692\n",
            "Epoch 79/100\n",
            "12/12 [==============================] - 19s 1s/step - loss: 0.0922 - accuracy: 0.9722 - val_loss: 0.0877 - val_accuracy: 0.9733\n",
            "Epoch 80/100\n",
            "12/12 [==============================] - 18s 2s/step - loss: 0.0876 - accuracy: 0.9736 - val_loss: 0.0860 - val_accuracy: 0.9732\n",
            "Epoch 81/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 0.0896 - accuracy: 0.9725 - val_loss: 0.0808 - val_accuracy: 0.9736\n",
            "Epoch 82/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0878 - accuracy: 0.9734 - val_loss: 0.0855 - val_accuracy: 0.9724\n",
            "Epoch 83/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0874 - accuracy: 0.9738 - val_loss: 0.0785 - val_accuracy: 0.9760\n",
            "Epoch 84/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.0845 - accuracy: 0.9735 - val_loss: 0.0813 - val_accuracy: 0.9741\n",
            "Epoch 85/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0868 - accuracy: 0.9738 - val_loss: 0.0848 - val_accuracy: 0.9728\n",
            "Epoch 86/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0862 - accuracy: 0.9736 - val_loss: 0.0782 - val_accuracy: 0.9761\n",
            "Epoch 87/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0872 - accuracy: 0.9731 - val_loss: 0.0859 - val_accuracy: 0.9739\n",
            "Epoch 88/100\n",
            "12/12 [==============================] - 22s 2s/step - loss: 0.0849 - accuracy: 0.9745 - val_loss: 0.0791 - val_accuracy: 0.9764\n",
            "Epoch 89/100\n",
            "12/12 [==============================] - 25s 2s/step - loss: 0.0833 - accuracy: 0.9749 - val_loss: 0.0805 - val_accuracy: 0.9744\n",
            "Epoch 90/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.0823 - accuracy: 0.9747 - val_loss: 0.0792 - val_accuracy: 0.9774\n",
            "Epoch 91/100\n",
            "12/12 [==============================] - 19s 2s/step - loss: 0.0803 - accuracy: 0.9753 - val_loss: 0.0794 - val_accuracy: 0.9758\n",
            "Epoch 92/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0808 - accuracy: 0.9753 - val_loss: 0.0825 - val_accuracy: 0.9754\n",
            "Epoch 93/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.0822 - accuracy: 0.9745 - val_loss: 0.0821 - val_accuracy: 0.9733\n",
            "Epoch 94/100\n",
            "12/12 [==============================] - 22s 2s/step - loss: 0.0816 - accuracy: 0.9744 - val_loss: 0.0790 - val_accuracy: 0.9754\n",
            "Epoch 95/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0799 - accuracy: 0.9754 - val_loss: 0.0784 - val_accuracy: 0.9745\n",
            "Epoch 96/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.0810 - accuracy: 0.9760 - val_loss: 0.0746 - val_accuracy: 0.9768\n",
            "Epoch 97/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 0.0805 - accuracy: 0.9754 - val_loss: 0.0789 - val_accuracy: 0.9744\n",
            "Epoch 98/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0788 - accuracy: 0.9759 - val_loss: 0.0730 - val_accuracy: 0.9763\n",
            "Epoch 99/100\n",
            "12/12 [==============================] - 14s 1s/step - loss: 0.0800 - accuracy: 0.9759 - val_loss: 0.0740 - val_accuracy: 0.9768\n",
            "Epoch 100/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0777 - accuracy: 0.9770 - val_loss: 0.0759 - val_accuracy: 0.9758\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1500x300 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Compile the model\n",
        "model_aug.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
        "'''sparse_categorical_crossentropy is for integer targets;\n",
        "    categorical_crossentropy is for one hot encoded targets'''\n",
        "\n",
        "learning_rate = 0.003\n",
        "epochs = 100\n",
        "batch_size = 4000\n",
        "\n",
        "# Train model\n",
        "history = model_aug.fit(\n",
        "    x=train_generator,\n",
        "    batch_size=batch_size,\n",
        "    epochs=epochs,\n",
        "    validation_data=val_generator,\n",
        ")\n",
        "\n",
        "# Plot metrics during training\n",
        "plt.figure(figsize=(15, 3))\n",
        "\n",
        "# Train Loss vs Validation Loss\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(history.history[model_aug.metrics_names[0]], label='Train Loss')\n",
        "plt.plot(history.history[f'val_{model_aug.metrics_names[0]}'], label='Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "\n",
        "# Train Accuracy vs Validation Accuracy\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(history.history[model_aug.metrics_names[1]], label='Train Accuracy')\n",
        "plt.plot(history.history[f'val_{model_aug.metrics_names[1]}'], label='Validation Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.legend()\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AT4dOSrquwR8"
      },
      "source": [
        "Issue of overfitting is resolved to some extent\n",
        "\n",
        "Trained with T4 GPU, Colab\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nOwbiD3HhKD_",
        "outputId": "405677f1-5573-41b3-98cb-861c3c53b588"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "model_aug.save('/content/drive/MyDrive/MLMarch/Augmented_100_model.h5')\n",
        "\n",
        "# model_aug.save('./models/Augmented_100_model.h5') # For saving it locally"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 29,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "K-Y1W0_ubsyh",
        "outputId": "c78aee21-53cd-4a51-cb09-64cef831ea25"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n",
            "\n",
            " Evaluate the model against the test set:\n",
            "3/3 [==============================] - 0s 6ms/step - loss: 0.0365 - accuracy: 0.9872\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "[0.036526937037706375, 0.9872000217437744]"
            ]
          },
          "execution_count": 29,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# from tensorflow.keras.models import load_model\n",
        "# from google.colab import drive\n",
        "# drive.mount('/content/drive')\n",
        "# model_aug = load_model('/content/drive/MyDrive/MLMarch/Augmented_100_model.h5')\n",
        "\n",
        "print(\"\\n Evaluate the model against the test set:\")\n",
        "model_aug.evaluate(x=x_test, y=y_test, batch_size=batch_size)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 655
        },
        "id": "45vRqgphdb_P",
        "outputId": "4ea20727-92bd-4528-8900-c1280c5b5846"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 1s 2ms/step\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "Text(95.72222222222221, 0.5, 'Truth')"
            ]
          },
          "execution_count": 14,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1000x700 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import seaborn as sn\n",
        "y_predicted = model_aug.predict(x_test)\n",
        "y_predicted_labels = [np.argmax(i) for i in y_predicted]\n",
        "cm = tf.math.confusion_matrix(labels=y_test,predictions=y_predicted_labels)\n",
        "\n",
        "plt.figure(figsize = (10,7))\n",
        "sn.heatmap(cm, annot=True, fmt='d')\n",
        "plt.xlabel('Predicted')\n",
        "plt.ylabel('Truth')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HJJaYV64dcme",
        "outputId": "9ad83c76-2d14-4e22-f68e-7adf865835d8"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "__Classification Report__ [Using Augmented data]:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.99      0.99      0.99       980\n",
            "           1       0.99      1.00      0.99      1135\n",
            "           2       0.99      0.99      0.99      1032\n",
            "           3       0.99      0.99      0.99      1010\n",
            "           4       0.98      0.99      0.99       982\n",
            "           5       0.99      0.99      0.99       892\n",
            "           6       0.99      0.99      0.99       958\n",
            "           7       0.98      0.99      0.99      1028\n",
            "           8       0.98      0.98      0.98       974\n",
            "           9       0.98      0.98      0.98      1009\n",
            "\n",
            "    accuracy                           0.99     10000\n",
            "   macro avg       0.99      0.99      0.99     10000\n",
            "weighted avg       0.99      0.99      0.99     10000\n",
            "\n"
          ]
        }
      ],
      "source": [
        "from sklearn.metrics import classification_report\n",
        "# Generate classification report\n",
        "report = classification_report(y_test, y_predicted_labels)\n",
        "\n",
        "print(\"__Classification Report__ [Using Augmented data]:\")\n",
        "print(report)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NYDIHzcQshLu"
      },
      "source": [
        "## **Using Data Augmentation + CNN**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "GuHMWOthmttj"
      },
      "outputs": [],
      "source": [
        "# Importing again other Normalization will happen twicw\n",
        "(x_train, y_train),(x_test, y_test) = tf.keras.datasets.mnist.load_data()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yC1nDbtimqwo",
        "outputId": "df9e5f04-3d7f-471c-d189-7625511f3c77"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(60000, 28, 28, 1)"
            ]
          },
          "execution_count": 17,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "x_train = np.expand_dims(x_train, axis=-1)\n",
        "x_train.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "id": "HGv1OikPmqw2"
      },
      "outputs": [],
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "x_train, x_val, y_train, y_val = train_test_split(x_train, y_train, test_size=0.2)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "id": "y2m705qlmqw3"
      },
      "outputs": [],
      "source": [
        "from keras.preprocessing.image import ImageDataGenerator\n",
        "batch_size  = 4000\n",
        "# Create an ImageDataGenerator instance with augmentation parameters\n",
        "datagen = ImageDataGenerator(\n",
        "    rotation_range=10,\n",
        "    width_shift_range=0.1,\n",
        "    height_shift_range=0.1,\n",
        "    shear_range=0.1,\n",
        "    zoom_range=0.2,\n",
        "    fill_mode='nearest')\n",
        "\n",
        "# Fit the ImageDataGenerator to your dataset\n",
        "datagen.fit(x_train)\n",
        "datagen.fit(x_val)\n",
        "# Generate augmented data\n",
        "train_generator = datagen.flow(x_train, y_train, batch_size=batch_size)\n",
        "val_generator = datagen.flow(x_val, y_val, batch_size=batch_size)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 595
        },
        "id": "yoJUsXsFDM12",
        "outputId": "70e248b4-65d0-4b12-9f7c-db28ab3e9208"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1500x300 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1500x300 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Plot some augmented images\n",
        "\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def plot_aug_images(generator, title):\n",
        "    fig, ax = plt.subplots(nrows=1, ncols=5, figsize=(15, 3))\n",
        "    fig.suptitle(title)\n",
        "\n",
        "    for i in range(5):\n",
        "        # Get a batch of images and labels from the generator\n",
        "        batch = next(generator)\n",
        "        images = batch[0]\n",
        "        labels_batch = batch[1]\n",
        "\n",
        "        # Pick one image from the batch\n",
        "        image = images[i]\n",
        "        label = labels_batch[i]\n",
        "\n",
        "        # Plot raw pixel data\n",
        "        ax[i].imshow(image)\n",
        "        #ax[i].axis('off')\n",
        "        ax[i].set_title(label)\n",
        "\n",
        "    plt.show()\n",
        "\n",
        "# Plot 5 augmented train images\n",
        "plot_aug_images(train_generator, \"Augmented Training Images\")\n",
        "plot_aug_images(train_generator, \"Augmented Validation Images\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PGuFKwtYgt30",
        "outputId": "401d5383-b765-4eb6-e8ce-133df5eb6080"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " rescaling_1 (Rescaling)     (None, 28, 28, 1)         0         \n",
            "                                                                 \n",
            " conv2d (Conv2D)             (None, 26, 26, 32)        320       \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2  (None, 13, 13, 32)        0         \n",
            " D)                                                              \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 11, 11, 64)        18496     \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPoolin  (None, 5, 5, 64)          0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 5, 5, 64)          0         \n",
            "                                                                 \n",
            " flatten_1 (Flatten)         (None, 1600)              0         \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 128)               204928    \n",
            "                                                                 \n",
            " dropout_2 (Dropout)         (None, 128)               0         \n",
            "                                                                 \n",
            " dense_4 (Dense)             (None, 10)                1290      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 225034 (879.04 KB)\n",
            "Trainable params: 225034 (879.04 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ],
      "source": [
        "import tensorflow as tf\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout\n",
        "\n",
        "model_conv = Sequential()\n",
        "model_conv.add(tf.keras.layers.experimental.preprocessing.Rescaling(1./255., input_shape=(28, 28, 1)))\n",
        "model_conv.add(Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=(28, 28, 1)))\n",
        "model_conv.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model_conv.add(Conv2D(64, kernel_size=(3, 3), activation='relu'))\n",
        "model_conv.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model_conv.add(Dropout(0.2))\n",
        "# Flatten layer to transition from convolutional layers to dense layers\n",
        "model_conv.add(Flatten())\n",
        "\n",
        "# Add dense layers\n",
        "model_conv.add(Dense(128, activation='relu'))\n",
        "model_conv.add(Dropout(0.2))\n",
        "model_conv.add(Dense(10, activation='softmax'))  # Output layer with 10 classes\n",
        "model_conv.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "Os9zVDYhopmT",
        "outputId": "fad7b09c-3fbc-46a1-f773-c4790b0bdeda"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.4713 - accuracy: 0.8526 - val_loss: 0.3681 - val_accuracy: 0.8899\n",
            "Epoch 2/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.3976 - accuracy: 0.8802 - val_loss: 0.3107 - val_accuracy: 0.9047\n",
            "Epoch 3/100\n",
            "12/12 [==============================] - 18s 2s/step - loss: 0.3449 - accuracy: 0.8941 - val_loss: 0.2667 - val_accuracy: 0.9204\n",
            "Epoch 4/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.3089 - accuracy: 0.9048 - val_loss: 0.2413 - val_accuracy: 0.9262\n",
            "Epoch 5/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.2785 - accuracy: 0.9149 - val_loss: 0.2108 - val_accuracy: 0.9387\n",
            "Epoch 6/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.2614 - accuracy: 0.9200 - val_loss: 0.2010 - val_accuracy: 0.9384\n",
            "Epoch 7/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.2427 - accuracy: 0.9261 - val_loss: 0.1864 - val_accuracy: 0.9426\n",
            "Epoch 8/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.2197 - accuracy: 0.9323 - val_loss: 0.1635 - val_accuracy: 0.9492\n",
            "Epoch 9/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.2110 - accuracy: 0.9348 - val_loss: 0.1525 - val_accuracy: 0.9560\n",
            "Epoch 10/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1993 - accuracy: 0.9390 - val_loss: 0.1528 - val_accuracy: 0.9523\n",
            "Epoch 11/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.1874 - accuracy: 0.9424 - val_loss: 0.1438 - val_accuracy: 0.9548\n",
            "Epoch 12/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1780 - accuracy: 0.9448 - val_loss: 0.1285 - val_accuracy: 0.9597\n",
            "Epoch 13/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1724 - accuracy: 0.9465 - val_loss: 0.1283 - val_accuracy: 0.9611\n",
            "Epoch 14/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1626 - accuracy: 0.9503 - val_loss: 0.1171 - val_accuracy: 0.9643\n",
            "Epoch 15/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1565 - accuracy: 0.9520 - val_loss: 0.1158 - val_accuracy: 0.9660\n",
            "Epoch 16/100\n",
            "12/12 [==============================] - 18s 2s/step - loss: 0.1493 - accuracy: 0.9545 - val_loss: 0.1136 - val_accuracy: 0.9649\n",
            "Epoch 17/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1446 - accuracy: 0.9564 - val_loss: 0.1081 - val_accuracy: 0.9686\n",
            "Epoch 18/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1408 - accuracy: 0.9569 - val_loss: 0.1014 - val_accuracy: 0.9696\n",
            "Epoch 19/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.1325 - accuracy: 0.9586 - val_loss: 0.0993 - val_accuracy: 0.9713\n",
            "Epoch 20/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.1285 - accuracy: 0.9613 - val_loss: 0.0936 - val_accuracy: 0.9712\n",
            "Epoch 21/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1269 - accuracy: 0.9616 - val_loss: 0.0913 - val_accuracy: 0.9722\n",
            "Epoch 22/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.1234 - accuracy: 0.9624 - val_loss: 0.0893 - val_accuracy: 0.9728\n",
            "Epoch 23/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1179 - accuracy: 0.9637 - val_loss: 0.0905 - val_accuracy: 0.9726\n",
            "Epoch 24/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1158 - accuracy: 0.9645 - val_loss: 0.0865 - val_accuracy: 0.9728\n",
            "Epoch 25/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1111 - accuracy: 0.9659 - val_loss: 0.0795 - val_accuracy: 0.9747\n",
            "Epoch 26/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1143 - accuracy: 0.9643 - val_loss: 0.0771 - val_accuracy: 0.9764\n",
            "Epoch 27/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1097 - accuracy: 0.9671 - val_loss: 0.0758 - val_accuracy: 0.9762\n",
            "Epoch 28/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1051 - accuracy: 0.9679 - val_loss: 0.0782 - val_accuracy: 0.9757\n",
            "Epoch 29/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.1027 - accuracy: 0.9695 - val_loss: 0.0765 - val_accuracy: 0.9773\n",
            "Epoch 30/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.1006 - accuracy: 0.9695 - val_loss: 0.0707 - val_accuracy: 0.9784\n",
            "Epoch 31/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0988 - accuracy: 0.9691 - val_loss: 0.0784 - val_accuracy: 0.9765\n",
            "Epoch 32/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0990 - accuracy: 0.9693 - val_loss: 0.0710 - val_accuracy: 0.9787\n",
            "Epoch 33/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0944 - accuracy: 0.9707 - val_loss: 0.0743 - val_accuracy: 0.9775\n",
            "Epoch 34/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0942 - accuracy: 0.9707 - val_loss: 0.0664 - val_accuracy: 0.9794\n",
            "Epoch 35/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0955 - accuracy: 0.9708 - val_loss: 0.0636 - val_accuracy: 0.9803\n",
            "Epoch 36/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0898 - accuracy: 0.9731 - val_loss: 0.0657 - val_accuracy: 0.9811\n",
            "Epoch 37/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0888 - accuracy: 0.9727 - val_loss: 0.0613 - val_accuracy: 0.9816\n",
            "Epoch 38/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0914 - accuracy: 0.9723 - val_loss: 0.0652 - val_accuracy: 0.9809\n",
            "Epoch 39/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.0909 - accuracy: 0.9721 - val_loss: 0.0631 - val_accuracy: 0.9807\n",
            "Epoch 40/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0850 - accuracy: 0.9741 - val_loss: 0.0611 - val_accuracy: 0.9819\n",
            "Epoch 41/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0840 - accuracy: 0.9741 - val_loss: 0.0623 - val_accuracy: 0.9801\n",
            "Epoch 42/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0812 - accuracy: 0.9748 - val_loss: 0.0578 - val_accuracy: 0.9822\n",
            "Epoch 43/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0811 - accuracy: 0.9748 - val_loss: 0.0586 - val_accuracy: 0.9826\n",
            "Epoch 44/100\n",
            "12/12 [==============================] - 18s 2s/step - loss: 0.0794 - accuracy: 0.9765 - val_loss: 0.0585 - val_accuracy: 0.9817\n",
            "Epoch 45/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0782 - accuracy: 0.9762 - val_loss: 0.0560 - val_accuracy: 0.9831\n",
            "Epoch 46/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0773 - accuracy: 0.9765 - val_loss: 0.0556 - val_accuracy: 0.9831\n",
            "Epoch 47/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0764 - accuracy: 0.9766 - val_loss: 0.0559 - val_accuracy: 0.9835\n",
            "Epoch 48/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0748 - accuracy: 0.9777 - val_loss: 0.0551 - val_accuracy: 0.9829\n",
            "Epoch 49/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0761 - accuracy: 0.9768 - val_loss: 0.0566 - val_accuracy: 0.9821\n",
            "Epoch 50/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.0750 - accuracy: 0.9767 - val_loss: 0.0550 - val_accuracy: 0.9837\n",
            "Epoch 51/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0753 - accuracy: 0.9763 - val_loss: 0.0536 - val_accuracy: 0.9836\n",
            "Epoch 52/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0718 - accuracy: 0.9780 - val_loss: 0.0531 - val_accuracy: 0.9837\n",
            "Epoch 53/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0689 - accuracy: 0.9791 - val_loss: 0.0503 - val_accuracy: 0.9852\n",
            "Epoch 54/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0728 - accuracy: 0.9773 - val_loss: 0.0528 - val_accuracy: 0.9845\n",
            "Epoch 55/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0715 - accuracy: 0.9782 - val_loss: 0.0511 - val_accuracy: 0.9833\n",
            "Epoch 56/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0697 - accuracy: 0.9785 - val_loss: 0.0533 - val_accuracy: 0.9837\n",
            "Epoch 57/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0674 - accuracy: 0.9797 - val_loss: 0.0495 - val_accuracy: 0.9848\n",
            "Epoch 58/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0681 - accuracy: 0.9796 - val_loss: 0.0478 - val_accuracy: 0.9860\n",
            "Epoch 59/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0689 - accuracy: 0.9791 - val_loss: 0.0459 - val_accuracy: 0.9861\n",
            "Epoch 60/100\n",
            "12/12 [==============================] - 18s 2s/step - loss: 0.0673 - accuracy: 0.9794 - val_loss: 0.0499 - val_accuracy: 0.9847\n",
            "Epoch 61/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0645 - accuracy: 0.9803 - val_loss: 0.0467 - val_accuracy: 0.9858\n",
            "Epoch 62/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0651 - accuracy: 0.9799 - val_loss: 0.0495 - val_accuracy: 0.9848\n",
            "Epoch 63/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0632 - accuracy: 0.9814 - val_loss: 0.0473 - val_accuracy: 0.9862\n",
            "Epoch 64/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0623 - accuracy: 0.9812 - val_loss: 0.0471 - val_accuracy: 0.9866\n",
            "Epoch 65/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0624 - accuracy: 0.9802 - val_loss: 0.0455 - val_accuracy: 0.9861\n",
            "Epoch 66/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.0635 - accuracy: 0.9799 - val_loss: 0.0461 - val_accuracy: 0.9875\n",
            "Epoch 67/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0625 - accuracy: 0.9810 - val_loss: 0.0461 - val_accuracy: 0.9855\n",
            "Epoch 68/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0619 - accuracy: 0.9807 - val_loss: 0.0469 - val_accuracy: 0.9862\n",
            "Epoch 69/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0638 - accuracy: 0.9795 - val_loss: 0.0428 - val_accuracy: 0.9868\n",
            "Epoch 70/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0601 - accuracy: 0.9818 - val_loss: 0.0417 - val_accuracy: 0.9868\n",
            "Epoch 71/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0590 - accuracy: 0.9817 - val_loss: 0.0417 - val_accuracy: 0.9881\n",
            "Epoch 72/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0612 - accuracy: 0.9814 - val_loss: 0.0463 - val_accuracy: 0.9864\n",
            "Epoch 73/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0594 - accuracy: 0.9821 - val_loss: 0.0433 - val_accuracy: 0.9855\n",
            "Epoch 74/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0584 - accuracy: 0.9820 - val_loss: 0.0427 - val_accuracy: 0.9868\n",
            "Epoch 75/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0619 - accuracy: 0.9809 - val_loss: 0.0431 - val_accuracy: 0.9871\n",
            "Epoch 76/100\n",
            "12/12 [==============================] - 18s 1s/step - loss: 0.0576 - accuracy: 0.9822 - val_loss: 0.0417 - val_accuracy: 0.9864\n",
            "Epoch 77/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0615 - accuracy: 0.9810 - val_loss: 0.0412 - val_accuracy: 0.9872\n",
            "Epoch 78/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0583 - accuracy: 0.9823 - val_loss: 0.0432 - val_accuracy: 0.9873\n",
            "Epoch 79/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0553 - accuracy: 0.9827 - val_loss: 0.0447 - val_accuracy: 0.9870\n",
            "Epoch 80/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0546 - accuracy: 0.9828 - val_loss: 0.0418 - val_accuracy: 0.9882\n",
            "Epoch 81/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0570 - accuracy: 0.9824 - val_loss: 0.0434 - val_accuracy: 0.9873\n",
            "Epoch 82/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0561 - accuracy: 0.9826 - val_loss: 0.0405 - val_accuracy: 0.9887\n",
            "Epoch 83/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0549 - accuracy: 0.9827 - val_loss: 0.0411 - val_accuracy: 0.9873\n",
            "Epoch 84/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0548 - accuracy: 0.9829 - val_loss: 0.0413 - val_accuracy: 0.9869\n",
            "Epoch 85/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0538 - accuracy: 0.9834 - val_loss: 0.0406 - val_accuracy: 0.9873\n",
            "Epoch 86/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0553 - accuracy: 0.9831 - val_loss: 0.0366 - val_accuracy: 0.9892\n",
            "Epoch 87/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0549 - accuracy: 0.9826 - val_loss: 0.0444 - val_accuracy: 0.9874\n",
            "Epoch 88/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0548 - accuracy: 0.9831 - val_loss: 0.0421 - val_accuracy: 0.9872\n",
            "Epoch 89/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0534 - accuracy: 0.9835 - val_loss: 0.0373 - val_accuracy: 0.9895\n",
            "Epoch 90/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0536 - accuracy: 0.9829 - val_loss: 0.0400 - val_accuracy: 0.9888\n",
            "Epoch 91/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0508 - accuracy: 0.9843 - val_loss: 0.0402 - val_accuracy: 0.9877\n",
            "Epoch 92/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0519 - accuracy: 0.9842 - val_loss: 0.0396 - val_accuracy: 0.9888\n",
            "Epoch 93/100\n",
            "12/12 [==============================] - 16s 1s/step - loss: 0.0539 - accuracy: 0.9835 - val_loss: 0.0392 - val_accuracy: 0.9877\n",
            "Epoch 94/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0525 - accuracy: 0.9846 - val_loss: 0.0382 - val_accuracy: 0.9881\n",
            "Epoch 95/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0533 - accuracy: 0.9840 - val_loss: 0.0388 - val_accuracy: 0.9883\n",
            "Epoch 96/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0499 - accuracy: 0.9846 - val_loss: 0.0365 - val_accuracy: 0.9886\n",
            "Epoch 97/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0508 - accuracy: 0.9840 - val_loss: 0.0381 - val_accuracy: 0.9887\n",
            "Epoch 98/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0498 - accuracy: 0.9849 - val_loss: 0.0375 - val_accuracy: 0.9887\n",
            "Epoch 99/100\n",
            "12/12 [==============================] - 17s 1s/step - loss: 0.0488 - accuracy: 0.9850 - val_loss: 0.0408 - val_accuracy: 0.9885\n",
            "Epoch 100/100\n",
            "12/12 [==============================] - 15s 1s/step - loss: 0.0475 - accuracy: 0.9854 - val_loss: 0.0375 - val_accuracy: 0.9883\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1500x300 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Compile the model\n",
        "model_conv.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
        "'''sparse_categorical_crossentropy is for integer targets;\n",
        "    categorical_crossentropy is for one hot encoded targets'''\n",
        "# Assuming you have already defined your ImageDataGenerator and augmented the data\n",
        "# augmented_data is the output of datagen.flow(x_train, y_train, batch_size=batch_size)\n",
        "learning_rate = 0.003\n",
        "epochs = 100\n",
        "batch_size = 4000\n",
        "\n",
        "history = model_conv.fit(\n",
        "    x=train_generator,\n",
        "    batch_size=batch_size,\n",
        "    epochs=epochs,\n",
        "    validation_data=val_generator,\n",
        ")\n",
        "\n",
        "# Plot metrics during training\n",
        "plt.figure(figsize=(15, 3))\n",
        "\n",
        "# Train Loss vs Validation Loss\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(history.history[model_conv.metrics_names[0]], label='Train Loss')\n",
        "plt.plot(history.history[f'val_{model_conv.metrics_names[0]}'], label='Validation Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "\n",
        "# Train Accuracy vs Validation Accuracy\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(history.history[model_conv.metrics_names[1]], label='Train Accuracy')\n",
        "plt.plot(history.history[f'val_{model_conv.metrics_names[1]}'], label='Validation Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.legend()\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AlmDVrYjyNgB"
      },
      "source": [
        "Trained with T4 GPU, Colab"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZDM1MnOKqOcQ",
        "outputId": "5ad665d8-4bc4-4c2a-e6d5-a0f4465012d4"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ],
      "source": [
        "\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "model_conv.save('/content/drive/MyDrive/MLMarch/CNN_Augmented_100_model.h5')\n",
        "\n",
        "# model_conv.save('./models/CNN_Augmented_100_model.h5') # For saving it locally"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "V4jt3qA2qOcO",
        "outputId": "77580d24-635f-41e5-b7b6-4e4605b88b50"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n",
            "\n",
            " Evaluate the new model against the test set:\n",
            "3/3 [==============================] - 0s 17ms/step - loss: 0.0153 - accuracy: 0.9945\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "[0.015324227511882782, 0.9944999814033508]"
            ]
          },
          "execution_count": 31,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# from tensorflow.keras.models import load_model\n",
        "# from google.colab import drive\n",
        "# drive.mount('/content/drive')\n",
        "# model_conv = load_model('/content/drive/MyDrive/MLMarch/CNN_Augmented_100_model.h5')\n",
        "\n",
        "print(\"\\n Evaluate the new model against the test set:\")\n",
        "model_conv.evaluate(x=x_test, y=y_test, batch_size=batch_size)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QcZjhezv_9wa"
      },
      "source": [
        "Best accuracy (99.45%) obtained till now."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 26,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 655
        },
        "id": "12aYvdswqOcQ",
        "outputId": "5d0f0506-ed04-4394-ce50-99e309cfdf30"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 1s 2ms/step\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "Text(95.72222222222221, 0.5, 'Truth')"
            ]
          },
          "execution_count": 26,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1000x700 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import seaborn as sn\n",
        "y_predicted = model_conv.predict(x_test)\n",
        "y_predicted_labels = [np.argmax(i) for i in y_predicted]\n",
        "cm = tf.math.confusion_matrix(labels=y_test,predictions=y_predicted_labels)\n",
        "\n",
        "plt.figure(figsize = (10,7))\n",
        "sn.heatmap(cm, annot=True, fmt='d')\n",
        "plt.xlabel('Predicted')\n",
        "plt.ylabel('Truth')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 27,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_ZAawHrxqOcR",
        "outputId": "078ae522-c843-41a9-a3b8-8a475bd2b9dc"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "__Classification Report__ [Using Augmented data with CNN]:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       1.00      1.00      1.00       980\n",
            "           1       1.00      1.00      1.00      1135\n",
            "           2       1.00      0.99      0.99      1032\n",
            "           3       1.00      1.00      1.00      1010\n",
            "           4       0.99      1.00      0.99       982\n",
            "           5       0.99      0.99      0.99       892\n",
            "           6       0.99      0.99      0.99       958\n",
            "           7       0.99      0.99      0.99      1028\n",
            "           8       0.99      0.99      0.99       974\n",
            "           9       1.00      0.99      0.99      1009\n",
            "\n",
            "    accuracy                           0.99     10000\n",
            "   macro avg       0.99      0.99      0.99     10000\n",
            "weighted avg       0.99      0.99      0.99     10000\n",
            "\n"
          ]
        }
      ],
      "source": [
        "from sklearn.metrics import classification_report\n",
        "# Generate classification report\n",
        "report = classification_report(y_test, y_predicted_labels)\n",
        "\n",
        "print(\"__Classification Report__ [Using Augmented data with CNN]:\")\n",
        "print(report)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Gg5g2xW__diH"
      },
      "source": [
        "We can see quite a few no. of 1.0 which indicates perfect precision/ recall/ f1-score."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "k1Jif96-qOcR"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}