MNIST_deep_learning.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ppEVZ1fEp8jS"
      },
      "source": [
        "**Import libraries**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "0ZNuSQZeBLHX"
      },
      "outputs": [],
      "source": [
        "pip install tensorflow[and-cuda]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "U2S4UFI6B-jt"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "from mpl_toolkits.mplot3d import Axes3D\n",
        "import tensorflow as tf\n",
        "import tensorflow_datasets as tfds"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YyVl9IzwqDzR"
      },
      "source": [
        "**Import dataset**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 84,
          "referenced_widgets": [
            "cbc86a453a4745bc92fcf7f3bc74fb57",
            "fdeddba42bbf48ddab8e98355ac238e6",
            "35a50fafbe1f4a4390d41be9aa6ed528",
            "e0a7fdccb74b4b0f81ffc9058abb0bc6",
            "dfe06ea3ecd84ab8b9ddb270f861eefe",
            "017958137b3e49e8813b6843d9a0015a",
            "cc3d22e652e64ebab4b21df72d01d97e",
            "0a401aaad47541b286cb4d1955a1283c",
            "f7e7d1e64ab64787a4b335eead8273fb",
            "c0754629efb54e60b3f52b57827d5b12",
            "6c49c7c636ef46b6a4178a9e382570a7"
          ]
        },
        "id": "cWQM7x07F31r",
        "outputId": "d64b6f6f-a16f-4de7-e5b9-8cfef45631e8"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Downloading and preparing dataset 11.06 MiB (download: 11.06 MiB, generated: 21.00 MiB, total: 32.06 MiB) to /root/tensorflow_datasets/mnist/3.0.1...\n"
          ]
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "cbc86a453a4745bc92fcf7f3bc74fb57",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Dl Completed...:   0%|          | 0/5 [00:00<?, ? file/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Dataset mnist downloaded and prepared to /root/tensorflow_datasets/mnist/3.0.1. Subsequent calls will reuse this data.\n"
          ]
        }
      ],
      "source": [
        "# Load MNIST dataset \n",
        "\n",
        "mnist_dataset, mnist_info = tfds.load(name='mnist', with_info=True, as_supervised=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "RmMBwPoaG1W_"
      },
      "outputs": [],
      "source": [
        "# This function computes the number of examples to be used for validation and test. \n",
        "# We do this by taking the split of the training dataset and splitting it into two sets one for validation and one for test\n",
        "mnist_train, mnist_test = mnist_dataset['train'], mnist_dataset['test']\n",
        "\n",
        "num_validation_samples = 0.1 * mnist_info.splits['train'].num_examples\n",
        "num_validation_samples = tf.cast(num_validation_samples, tf.int64)\n",
        "\n",
        "num_test_samples = mnist_info.splits['test'].num_examples\n",
        "num_test_samples = tf.cast(num_test_samples, tf.int64)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "JU7LLImKIhe9"
      },
      "outputs": [],
      "source": [
        "def scale(image, label):\n",
        "  \"\"\"\n",
        "   Scale the image to 0 - 255. This is useful for image segmentation where you don't know the color of the image.\n",
        "   \n",
        "   @param image - A tensor of shape [ height width channels ] representing an image.\n",
        "   @param label - A tensor of shape [ height width ] representing the ground truth labels.\n",
        "   \n",
        "   @return A tuple of the scaled image and label tensors. The first dimension is the same as the input and the second dimension is the same as the label\n",
        "  \"\"\"\n",
        "  image = tf.cast(image, tf.float32)\n",
        "  image /= 255.\n",
        "  return image, label"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "tRQlLns8I4YK"
      },
      "outputs": [],
      "source": [
        "# Scale data to be used for training and validation.\n",
        "scaled_train_and_validation_data = mnist_train.map(scale)\n",
        "\n",
        "test_data = mnist_test.map(scale)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "nbX9gkM2aM9P"
      },
      "outputs": [],
      "source": [
        "# Randomly shuffles the training and validation data to avoid overfitting the data. \n",
        "# This is useful for generating random data\n",
        "BUFFER_SIZE = 10000\n",
        "\n",
        "suffled_train_and_validation_data = scaled_train_and_validation_data.shuffle(BUFFER_SIZE)\n",
        "\n",
        "validation_data = suffled_train_and_validation_data.take(num_validation_samples)\n",
        "\n",
        "train_data= suffled_train_and_validation_data.skip(num_validation_samples)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "id": "rAkaa-sYbTQT"
      },
      "outputs": [],
      "source": [
        "# Generate batches of training validation and test data for node index generation. This is to avoid running out of memory\n",
        "BATCH_SIZE = 100\n",
        "train_data = train_data.batch(BATCH_SIZE)\n",
        "validation_data = validation_data.batch(num_validation_samples)\n",
        "test_data = test_data.batch(num_test_samples)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "395sIq-IcEJW"
      },
      "outputs": [],
      "source": [
        "# The function returns validation inputs and targets. \n",
        "# This is a generator function that iterates over the data and returns the first item\n",
        "validation_inputs, validation_targets = next(iter(validation_data))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 22,
      "metadata": {
        "id": "8RHO6Elucwd7"
      },
      "outputs": [],
      "source": [
        "# This is the code that is run when you want to set up the parameters. \n",
        "\n",
        "input_size = 784\n",
        "output_size = 10\n",
        "hidden_layer_size = 200"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 23,
      "metadata": {
        "id": "R9QIE1AOdBij"
      },
      "outputs": [],
      "source": [
        "# Build a Sequential model that can be used for training.\n",
        "model = tf.keras.Sequential([\n",
        "    tf.keras.layers.Flatten(input_shape = (28,28,1)),\n",
        "    tf.keras.layers.Dense(hidden_layer_size, activation = 'relu'),\n",
        "    tf.keras.layers.Dense(hidden_layer_size, activation = 'relu'),\n",
        "    tf.keras.layers.Dense(hidden_layer_size, activation = 'relu'),\n",
        "    tf.keras.layers.Dense(output_size, activation = 'softmax')\n",
        "])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 30,
      "metadata": {
        "id": "gEHy1QVkelpp"
      },
      "outputs": [],
      "source": [
        "# Compiles and compiles Adamagnet model. Args : optimizer : Keras optimizer. learning_rate : Learning rate\n",
        "optimizer = tf.keras.optimizers.Adam(learning_rate=0.0005)\n",
        "model.compile(optimizer=optimizer, loss='sparse_categorical_crossentropy', metrics=['accuracy'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 31,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MxerQdeGfE6V",
        "outputId": "7f10463f-f05b-4796-b8c9-85e73887c2f5"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/5\n",
            "540/540 - 10s - loss: 0.0256 - accuracy: 0.9920 - val_loss: 0.0325 - val_accuracy: 0.9895 - 10s/epoch - 19ms/step\n",
            "Epoch 2/5\n",
            "540/540 - 6s - loss: 0.0120 - accuracy: 0.9964 - val_loss: 0.0244 - val_accuracy: 0.9932 - 6s/epoch - 12ms/step\n",
            "Epoch 3/5\n",
            "540/540 - 6s - loss: 0.0080 - accuracy: 0.9978 - val_loss: 0.0170 - val_accuracy: 0.9955 - 6s/epoch - 12ms/step\n",
            "Epoch 4/5\n",
            "540/540 - 8s - loss: 0.0057 - accuracy: 0.9984 - val_loss: 0.0125 - val_accuracy: 0.9963 - 8s/epoch - 14ms/step\n",
            "Epoch 5/5\n",
            "540/540 - 7s - loss: 0.0043 - accuracy: 0.9987 - val_loss: 0.0085 - val_accuracy: 0.9968 - 7s/epoch - 12ms/step\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "<keras.src.callbacks.History at 0x7e2ccfe705e0>"
            ]
          },
          "execution_count": 31,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Train and validation model with training data and validation data. \n",
        "model.fit(train_data, epochs=5, validation_data=(validation_inputs, validation_targets), verbose=2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 32,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5bhr2aU9nsZG",
        "outputId": "171c5bc4-1464-4776-c38d-1eae8fa59449"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1/1 [==============================] - 2s 2s/step - loss: 0.1069 - accuracy: 0.9836\n"
          ]
        }
      ],
      "source": [
        "# Evaluate the model on test_data and return the loss and accuracy.\n",
        "test_loss, test_accuracy = model.evaluate(test_data)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 33,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Oz_7NNPzn9tF",
        "outputId": "1a986909-6921-40df-b272-06743d3396b4"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Test loss: 0.11. Test Accuracy: 98.36%\n"
          ]
        }
      ],
      "source": [
        "# Prints test loss and accuracy to standard output.\n",
        "print('Test loss: {0:.2f}. Test Accuracy: {1:.2f}%'.format(test_loss, test_accuracy * 100.))"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "017958137b3e49e8813b6843d9a0015a": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0a401aaad47541b286cb4d1955a1283c": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "35a50fafbe1f4a4390d41be9aa6ed528": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0a401aaad47541b286cb4d1955a1283c",
            "max": 5,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_f7e7d1e64ab64787a4b335eead8273fb",
            "value": 5
          }
        },
        "6c49c7c636ef46b6a4178a9e382570a7": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "c0754629efb54e60b3f52b57827d5b12": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "cbc86a453a4745bc92fcf7f3bc74fb57": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_fdeddba42bbf48ddab8e98355ac238e6",
              "IPY_MODEL_35a50fafbe1f4a4390d41be9aa6ed528",
              "IPY_MODEL_e0a7fdccb74b4b0f81ffc9058abb0bc6"
            ],
            "layout": "IPY_MODEL_dfe06ea3ecd84ab8b9ddb270f861eefe"
          }
        },
        "cc3d22e652e64ebab4b21df72d01d97e": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "dfe06ea3ecd84ab8b9ddb270f861eefe": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "e0a7fdccb74b4b0f81ffc9058abb0bc6": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_c0754629efb54e60b3f52b57827d5b12",
            "placeholder": "​",
            "style": "IPY_MODEL_6c49c7c636ef46b6a4178a9e382570a7",
            "value": " 5/5 [00:00&lt;00:00, 12.84 file/s]"
          }
        },
        "f7e7d1e64ab64787a4b335eead8273fb": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "fdeddba42bbf48ddab8e98355ac238e6": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_017958137b3e49e8813b6843d9a0015a",
            "placeholder": "​",
            "style": "IPY_MODEL_cc3d22e652e64ebab4b21df72d01d97e",
            "value": "Dl Completed...: 100%"
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}