Digit_Classifier.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyMcw8Ftlu3oPpejLVTneMdm",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/chiragtyagi2003/CNN-Clasifier/blob/main/Digit_Classifier.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "kGgyTX80ZwIx"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Dependencies"
      ],
      "metadata": {
        "id": "qD6dKjY2aFOE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Import Tensorflow 2.0\n",
        "%tensorflow_version 2.x\n",
        "import tensorflow as tf\n",
        "\n",
        "!pip install mitdeeplearning\n",
        "import mitdeeplearning as mdl\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import random\n",
        "from tqdm import tqdm\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PuKrZjiIaGwO",
        "outputId": "3259c07c-6c9d-4d4c-8eec-f40d6c531951"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Colab only includes TensorFlow 2.x; %tensorflow_version has no effect.\n",
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting mitdeeplearning\n",
            "  Downloading mitdeeplearning-0.3.0.tar.gz (2.1 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.1/2.1 MB\u001b[0m \u001b[31m23.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from mitdeeplearning) (1.22.4)\n",
            "Requirement already satisfied: regex in /usr/local/lib/python3.10/dist-packages (from mitdeeplearning) (2022.10.31)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from mitdeeplearning) (4.65.0)\n",
            "Requirement already satisfied: gym in /usr/local/lib/python3.10/dist-packages (from mitdeeplearning) (0.25.2)\n",
            "Requirement already satisfied: cloudpickle>=1.2.0 in /usr/local/lib/python3.10/dist-packages (from gym->mitdeeplearning) (2.2.1)\n",
            "Requirement already satisfied: gym-notices>=0.0.4 in /usr/local/lib/python3.10/dist-packages (from gym->mitdeeplearning) (0.0.8)\n",
            "Building wheels for collected packages: mitdeeplearning\n",
            "  Building wheel for mitdeeplearning (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for mitdeeplearning: filename=mitdeeplearning-0.3.0-py3-none-any.whl size=2117305 sha256=f4c2c740aa25c407ad84498c41f88dd3c6be128243964b3e96239b3f87a3f981\n",
            "  Stored in directory: /root/.cache/pip/wheels/9c/9f/b5/0e31f83bc60a83625e37429f923934e26802d0d68cf3ef3216\n",
            "Successfully built mitdeeplearning\n",
            "Installing collected packages: mitdeeplearning\n",
            "Successfully installed mitdeeplearning-0.3.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Downloading the dataset"
      ],
      "metadata": {
        "id": "scn9thT5aKFr"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Data PreProcessing"
      ],
      "metadata": {
        "id": "R9rJcy0Sacd_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# train_images = (np.expand_dims(train_images, axis=-1)/255.).astype(np.float32)\n",
        "# train_labels = (train_labels).astype(np.int64)\n",
        "# test_images = (np.expand_dims(test_images, axis=-1)/255.).astype(np.float32)\n",
        "# test_labels - (test_labels).astype(np.int64)\n",
        "mnist = tf.keras.datasets.mnist\n",
        "\n",
        "(train_images, train_labels), (test_images, test_labels) = mnist.load_data()\n",
        "train_images = (np.expand_dims(train_images, axis=-1)/255.).astype(np.float32)\n",
        "train_labels = (train_labels).astype(np.int64)\n",
        "test_images = (np.expand_dims(test_images, axis=-1)/255.).astype(np.float32)\n",
        "test_labels = (test_labels).astype(np.int64)\n",
        "\n",
        "print(train_images[[0]].shape)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bHXb6oSIaiP3",
        "outputId": "27c5bc84-7058-45b3-c670-00b07ac58daa"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(1, 28, 28, 1)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# plotting some of the images\n",
        "plt.figure(figsize=(10,10))\n",
        "random_inds = np.random.choice(60000,36)\n",
        "for i in range(36):\n",
        "    plt.subplot(6,6,i+1)\n",
        "    plt.xticks([])\n",
        "    plt.yticks([])\n",
        "    plt.grid(False)\n",
        "    image_ind = random_inds[i]\n",
        "    plt.imshow(np.squeeze(train_images[image_ind]), cmap=plt.cm.binary)\n",
        "    plt.xlabel(train_labels[image_ind])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 729
        },
        "id": "CNSX32-wcCr0",
        "outputId": "f999029e-a703-4305-b239-62197175bc08"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py:283: DeprecationWarning: `should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
            "  and should_run_async(code)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x1000 with 36 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Model 1 - simple Neural Network with two Dense layers"
      ],
      "metadata": {
        "id": "BPLMKl8gcVqO"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def build_first_model():\n",
        "  first_model = tf.keras.Sequential([\n",
        "\n",
        "    # layer1 => to flatten the images\n",
        "    # there are no learned params in this layer\n",
        "    # it only reformats the data\n",
        "    tf.keras.layers.Flatten(),\n",
        "\n",
        "    # layer2 => first dense layer\n",
        "    tf.keras.layers.Dense(units=128, activation = tf.nn.relu),\n",
        "\n",
        "    # layer3 => output layer\n",
        "    tf.keras.layers.Dense(units=10, activation=tf.nn.softmax)\n",
        "  ])\n",
        "\n",
        "  return first_model"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ElGo2Mafca7p",
        "outputId": "e89830b2-c56d-4331-b75b-9ece6dc800da"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py:283: DeprecationWarning: `should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
            "  and should_run_async(code)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# building the first model\n",
        "model = build_first_model()\n"
      ],
      "metadata": {
        "id": "1BKQqv6KcCtZ"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Compiling the model"
      ],
      "metadata": {
        "id": "VHQwzNUWfU6a"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(optimizer=tf.keras.optimizers.SGD(learning_rate=1e-1),\n",
        "                                                loss = 'sparse_categorical_crossentropy',\n",
        "                                                metrics = ['accuracy']\n",
        "                                                )"
      ],
      "metadata": {
        "id": "39l2RctffW-9"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "3CVXTOyOf71g"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Training the model"
      ],
      "metadata": {
        "id": "2reTean7f73i"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# hpyerparameters\n",
        "BATCH_SIZE = 64\n",
        "EPOCHS = 100\n",
        "\n",
        "model.fit(train_images, train_labels, batch_size = BATCH_SIZE, epochs = EPOCHS)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7sy25g7Xf9Xn",
        "outputId": "67b2f82a-e2fe-43af-d4fe-a3e8ebc6936d"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3908 - accuracy: 0.8823\n",
            "Epoch 2/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3940 - accuracy: 0.8822\n",
            "Epoch 3/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3882 - accuracy: 0.8850\n",
            "Epoch 4/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3930 - accuracy: 0.8828\n",
            "Epoch 5/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3864 - accuracy: 0.8856\n",
            "Epoch 6/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3867 - accuracy: 0.8842\n",
            "Epoch 7/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3850 - accuracy: 0.8855\n",
            "Epoch 8/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3846 - accuracy: 0.8862\n",
            "Epoch 9/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3812 - accuracy: 0.8873\n",
            "Epoch 10/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3824 - accuracy: 0.8860\n",
            "Epoch 11/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3810 - accuracy: 0.8858\n",
            "Epoch 12/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3845 - accuracy: 0.8850\n",
            "Epoch 13/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3858 - accuracy: 0.8842\n",
            "Epoch 14/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3829 - accuracy: 0.8854\n",
            "Epoch 15/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3834 - accuracy: 0.8853\n",
            "Epoch 16/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3783 - accuracy: 0.8873\n",
            "Epoch 17/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3786 - accuracy: 0.8864\n",
            "Epoch 18/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3786 - accuracy: 0.8868\n",
            "Epoch 19/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3761 - accuracy: 0.8881\n",
            "Epoch 20/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3769 - accuracy: 0.8885\n",
            "Epoch 21/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3762 - accuracy: 0.8882\n",
            "Epoch 22/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3757 - accuracy: 0.8885\n",
            "Epoch 23/100\n",
            "938/938 [==============================] - 4s 5ms/step - loss: 0.3745 - accuracy: 0.8903\n",
            "Epoch 24/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3743 - accuracy: 0.8877\n",
            "Epoch 25/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3732 - accuracy: 0.8884\n",
            "Epoch 26/100\n",
            "938/938 [==============================] - 3s 4ms/step - loss: 0.3712 - accuracy: 0.8892\n",
            "Epoch 27/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3762 - accuracy: 0.8881\n",
            "Epoch 28/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3705 - accuracy: 0.8901\n",
            "Epoch 29/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3718 - accuracy: 0.8884\n",
            "Epoch 30/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3727 - accuracy: 0.8895\n",
            "Epoch 31/100\n",
            "938/938 [==============================] - 3s 4ms/step - loss: 0.3721 - accuracy: 0.8887\n",
            "Epoch 32/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3700 - accuracy: 0.8893\n",
            "Epoch 33/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3695 - accuracy: 0.8897\n",
            "Epoch 34/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3686 - accuracy: 0.8902\n",
            "Epoch 35/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3671 - accuracy: 0.8915\n",
            "Epoch 36/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3677 - accuracy: 0.8900\n",
            "Epoch 37/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3684 - accuracy: 0.8911\n",
            "Epoch 38/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3662 - accuracy: 0.8921\n",
            "Epoch 39/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3672 - accuracy: 0.8903\n",
            "Epoch 40/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3644 - accuracy: 0.8923\n",
            "Epoch 41/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3664 - accuracy: 0.8906\n",
            "Epoch 42/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3645 - accuracy: 0.8911\n",
            "Epoch 43/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3646 - accuracy: 0.8918\n",
            "Epoch 44/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3697 - accuracy: 0.8897\n",
            "Epoch 45/100\n",
            "938/938 [==============================] - 3s 4ms/step - loss: 0.3630 - accuracy: 0.8913\n",
            "Epoch 46/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3602 - accuracy: 0.8928\n",
            "Epoch 47/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3606 - accuracy: 0.8930\n",
            "Epoch 48/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3637 - accuracy: 0.8918\n",
            "Epoch 49/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3639 - accuracy: 0.8917\n",
            "Epoch 50/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3627 - accuracy: 0.8917\n",
            "Epoch 51/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3601 - accuracy: 0.8938\n",
            "Epoch 52/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3592 - accuracy: 0.8939\n",
            "Epoch 53/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3597 - accuracy: 0.8919\n",
            "Epoch 54/100\n",
            "938/938 [==============================] - 3s 4ms/step - loss: 0.3596 - accuracy: 0.8926\n",
            "Epoch 55/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3592 - accuracy: 0.8921\n",
            "Epoch 56/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3592 - accuracy: 0.8930\n",
            "Epoch 57/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3574 - accuracy: 0.8945\n",
            "Epoch 58/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3564 - accuracy: 0.8940\n",
            "Epoch 59/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3573 - accuracy: 0.8932\n",
            "Epoch 60/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3546 - accuracy: 0.8949\n",
            "Epoch 61/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3560 - accuracy: 0.8937\n",
            "Epoch 62/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3568 - accuracy: 0.8942\n",
            "Epoch 63/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3541 - accuracy: 0.8945\n",
            "Epoch 64/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3542 - accuracy: 0.8955\n",
            "Epoch 65/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3531 - accuracy: 0.8947\n",
            "Epoch 66/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3541 - accuracy: 0.8957\n",
            "Epoch 67/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3508 - accuracy: 0.8950\n",
            "Epoch 68/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3506 - accuracy: 0.8954\n",
            "Epoch 69/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3528 - accuracy: 0.8945\n",
            "Epoch 70/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3526 - accuracy: 0.8950\n",
            "Epoch 71/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3485 - accuracy: 0.8960\n",
            "Epoch 72/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3506 - accuracy: 0.8946\n",
            "Epoch 73/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3476 - accuracy: 0.8964\n",
            "Epoch 74/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3504 - accuracy: 0.8943\n",
            "Epoch 75/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3465 - accuracy: 0.8970\n",
            "Epoch 76/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3474 - accuracy: 0.8965\n",
            "Epoch 77/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3481 - accuracy: 0.8974\n",
            "Epoch 78/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3457 - accuracy: 0.8974\n",
            "Epoch 79/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3463 - accuracy: 0.8963\n",
            "Epoch 80/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3461 - accuracy: 0.8978\n",
            "Epoch 81/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3481 - accuracy: 0.8971\n",
            "Epoch 82/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3481 - accuracy: 0.8957\n",
            "Epoch 83/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3446 - accuracy: 0.8976\n",
            "Epoch 84/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3471 - accuracy: 0.8963\n",
            "Epoch 85/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3449 - accuracy: 0.8962\n",
            "Epoch 86/100\n",
            "938/938 [==============================] - 3s 4ms/step - loss: 0.3465 - accuracy: 0.8969\n",
            "Epoch 87/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3442 - accuracy: 0.8975\n",
            "Epoch 88/100\n",
            "938/938 [==============================] - 3s 4ms/step - loss: 0.3439 - accuracy: 0.8974\n",
            "Epoch 89/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3423 - accuracy: 0.8993\n",
            "Epoch 90/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3440 - accuracy: 0.8981\n",
            "Epoch 91/100\n",
            "938/938 [==============================] - 4s 5ms/step - loss: 0.3420 - accuracy: 0.8988\n",
            "Epoch 92/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3424 - accuracy: 0.8979\n",
            "Epoch 93/100\n",
            "938/938 [==============================] - 3s 4ms/step - loss: 0.3416 - accuracy: 0.8979\n",
            "Epoch 94/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3418 - accuracy: 0.8982\n",
            "Epoch 95/100\n",
            "938/938 [==============================] - 4s 4ms/step - loss: 0.3402 - accuracy: 0.8977\n",
            "Epoch 96/100\n",
            "938/938 [==============================] - 3s 4ms/step - loss: 0.3409 - accuracy: 0.8983\n",
            "Epoch 97/100\n",
            "938/938 [==============================] - 3s 4ms/step - loss: 0.3406 - accuracy: 0.8983\n",
            "Epoch 98/100\n",
            "938/938 [==============================] - 4s 5ms/step - loss: 0.3390 - accuracy: 0.8992\n",
            "Epoch 99/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3390 - accuracy: 0.8990\n",
            "Epoch 100/100\n",
            "938/938 [==============================] - 3s 3ms/step - loss: 0.3370 - accuracy: 0.8997\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7f3a4c9f9c60>"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "vPPOSIq_hebJ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Evaluating the model1"
      ],
      "metadata": {
        "id": "u10RC7K6hedA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "test_loss, test_acc = model.evaluate(test_images, test_labels)\n",
        "\n",
        "print('Test Accuracy: ', test_acc)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "E43OL4MahgPG",
        "outputId": "caee4d8f-89b9-44a0-ad0f-be1ad61574e8"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 [==============================] - 1s 2ms/step - loss: 44.1710 - accuracy: 0.8503\n",
            "Test Accuracy:  0.8503000140190125\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "NYibXcAklkUK"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Define the CNN model, model2"
      ],
      "metadata": {
        "id": "guhqAQxClkV1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def build_cnn_model():\n",
        "  cnn_model = tf.keras.Sequential([\n",
        "\n",
        "      # first conv layer\n",
        "      tf.keras.layers.Conv2D(filters=24, kernel_size=(3,3), activation = tf.nn.relu),\n",
        "\n",
        "      # first pooling layer\n",
        "      tf.keras.layers.MaxPool2D(pool_size=(2,2)),\n",
        "\n",
        "      # second conv layer\n",
        "      tf.keras.layers.Conv2D(filters=36, kernel_size=(3,3), activation = tf.nn.relu),\n",
        "\n",
        "      # second pooling layer\n",
        "      tf.keras.layers.MaxPool2D(pool_size=(2,2)),\n",
        "\n",
        "      # flatter layer of fully conncected network\n",
        "      tf.keras.layers.Flatten(),\n",
        "\n",
        "      # first hidden layer\n",
        "      tf.keras.layers.Dense(units = 128, activation = tf.nn.relu),\n",
        "\n",
        "      # output layer\n",
        "      tf.keras.layers.Dense(units = 10, activation = tf.nn.softmax),\n",
        "\n",
        "  ])\n",
        "\n",
        "\n",
        "  return cnn_model;"
      ],
      "metadata": {
        "id": "GPebW5pdll7h"
      },
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "80wr5pasm8lK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        " #building the model\n",
        "cnn_model = build_cnn_model()\n",
        "\n",
        "#initilaze the model by passing some data through\n",
        "cnn_model.predict(train_images[[0]])\n",
        "\n",
        "print(cnn_model.summary())\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "N3CDs-c4m9K1",
        "outputId": "8adde1c5-1f48-4147-c7e6-b36cc73a3cb1"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 1s 742ms/step\n",
            "Model: \"sequential_4\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d_6 (Conv2D)           (None, 26, 26, 24)        240       \n",
            "                                                                 \n",
            " max_pooling2d_6 (MaxPooling  (None, 13, 13, 24)       0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_7 (Conv2D)           (None, 11, 11, 36)        7812      \n",
            "                                                                 \n",
            " max_pooling2d_7 (MaxPooling  (None, 5, 5, 36)         0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten_4 (Flatten)         (None, 900)               0         \n",
            "                                                                 \n",
            " dense_8 (Dense)             (None, 128)               115328    \n",
            "                                                                 \n",
            " dense_9 (Dense)             (None, 10)                1290      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 124,670\n",
            "Trainable params: 124,670\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Train the Cnn model"
      ],
      "metadata": {
        "id": "crQ4lRRrqEPS"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#compiling it\n",
        "cnn_model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=1e-3),\n",
        "                  loss = 'sparse_categorical_crossentropy',\n",
        "                  metrics = ['accuracy'])\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-S8_0jrrqFbr",
        "outputId": "a48e3eae-a116-4469-8a2a-f41b55c8b80e"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py:283: DeprecationWarning: `should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
            "  and should_run_async(code)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# training it\n",
        "cnn_model.fit(x=train_images, y=train_labels, batch_size = 64, epochs = 5)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uv8oAFyCqZ3C",
        "outputId": "e516b5bd-78fd-4dbd-8870-adcc0385615c"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/5\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py:283: DeprecationWarning: `should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
            "  and should_run_async(code)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "938/938 [==============================] - 52s 54ms/step - loss: 0.1778 - accuracy: 0.9473\n",
            "Epoch 2/5\n",
            "938/938 [==============================] - 44s 47ms/step - loss: 0.0530 - accuracy: 0.9837\n",
            "Epoch 3/5\n",
            "938/938 [==============================] - 43s 46ms/step - loss: 0.0379 - accuracy: 0.9883\n",
            "Epoch 4/5\n",
            "938/938 [==============================] - 42s 45ms/step - loss: 0.0288 - accuracy: 0.9908\n",
            "Epoch 5/5\n",
            "938/938 [==============================] - 43s 46ms/step - loss: 0.0224 - accuracy: 0.9929\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7f3a4c988a90>"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [],
      "metadata": {
        "id": "_G4HWVL7qtnZ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Make predictions"
      ],
      "metadata": {
        "id": "T3rAzSWWqtp_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "predictions = cnn_model.predict(test_images)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uFpqo-5FqvaP",
        "outputId": "655b4c98-045a-4ac7-ef64-654ea00197dd"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 [==============================] - 4s 11ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictions[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Y3JS1QAHqz1y",
        "outputId": "52dd4ad4-015c-45bb-e39a-fac4281e7c77"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([5.93748384e-09, 1.78508071e-08, 2.96628684e-07, 1.80457590e-07,\n",
              "       7.08961917e-11, 1.71611181e-09, 1.40614415e-14, 9.99998033e-01,\n",
              "       1.04734532e-08, 1.25124723e-06], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predicted_img0 = np.argmax(predictions[0])\n",
        "\n",
        "print(predicted_img0)\n",
        "\n",
        "\n",
        "\n",
        "plt.imshow(test_images[0,:,:,0], cmap = plt.cm.binary)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 519
        },
        "id": "uKIuu0H4rhzn",
        "outputId": "8e53902f-e627-482d-cf2b-c830797c87dc"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "7\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py:283: DeprecationWarning: `should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
            "  and should_run_async(code)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f3a4db7a4a0>"
            ]
          },
          "metadata": {},
          "execution_count": 37
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Visualizer"
      ],
      "metadata": {
        "id": "7FymU9nmr5r9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#@title Change the slider to look at the model's predictions! { run: \"auto\" }\n",
        "\n",
        "image_index = 16 #@param {type:\"slider\", min:0, max:100, step:1}\n",
        "plt.subplot(1,2,1)\n",
        "mdl.lab2.plot_image_prediction(image_index, predictions, test_labels, test_images)\n",
        "plt.subplot(1,2,2)\n",
        "mdl.lab2.plot_value_prediction(image_index, predictions, test_labels)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 406
        },
        "id": "lVPCmUsbr7hY",
        "outputId": "28243889-7ef6-4c70-f661-5a45ee7791f2"
      },
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plots the first X test images, their predicted label, and the true label\n",
        "# Color correct predictions in blue, incorrect predictions in red\n",
        "num_rows = 5\n",
        "num_cols = 4\n",
        "num_images = num_rows*num_cols\n",
        "plt.figure(figsize=(2*2*num_cols, 2*num_rows))\n",
        "for i in range(num_images):\n",
        "  plt.subplot(num_rows, 2*num_cols, 2*i+1)\n",
        "  mdl.lab2.plot_image_prediction(i, predictions, test_labels, test_images)\n",
        "  plt.subplot(num_rows, 2*num_cols, 2*i+2)\n",
        "  mdl.lab2.plot_value_prediction(i, predictions, test_labels)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 431
        },
        "id": "hvU0NMr7r4jy",
        "outputId": "3bc44870-c17f-4d75-cb3c-9336f6b11787"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1600x1000 with 40 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}