Handwritten_digit_Recognition.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "95149e5e",
   "metadata": {},
   "source": [
    "\n",
    "## Handwritten Digit recognition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "9aa9636c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import random\n",
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Flatten, Conv2D, Dense, MaxPooling2D\n",
    "from tensorflow.keras.optimizers import SGD #stochastic gradient descent\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from tensorflow.keras.datasets import mnist #dataset of handwritten digits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ffac3511",
   "metadata": {},
   "outputs": [],
   "source": [
    "(X_train, y_train), (X_test, y_test)= mnist.load_data() #test train split inbuilt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "8c124b31",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 20 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_digit(image, digit, plt, i):\n",
    "    plt.subplot(4, 5, i + 1)\n",
    "    plt.imshow(image, cmap=plt.get_cmap('gray'))\n",
    "    plt.title(f\"Digit: {digit}\")\n",
    "    plt.xticks([])\n",
    "    plt.yticks([])\n",
    "\n",
    "\n",
    "plt.figure(figsize=(16, 10))\n",
    "for i in range(20):\n",
    "    plot_digit(X_train[i], y_train[i], plt, i)\n",
    "    \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d76b2c04",
   "metadata": {},
   "outputs": [],
   "source": [
    "#data is in the form of pixels, if we scale down the pixels it will be easier for the model to handle\n",
    "X_train= X_train/255.0\n",
    "X_test= X_test/255.0\n",
    "\n",
    "\n",
    "#add an additional dimension of colour to the dataset because the images are in grayscale\n",
    "X_train= X_train.reshape((X_train.shape + (1,)))\n",
    "X_test= X_test.reshape((X_test.shape + (1,)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "46ada4e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "#creating the model\n",
    "model= Sequential([ #sequential nn \n",
    "    Conv2D(32, (3,3), activation= \"relu\", input_shape= (28,28,1)),\n",
    "    MaxPooling2D((2,2)),\n",
    "    Flatten(),\n",
    "    Dense(100, activation= \"relu\"),\n",
    "    Dense(10, activation= \"softmax\"), #for converting the numerical outputs to probabilities, 10 inputs for digits from 0-9 and their probabilities\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d6c4edea",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 26, 26, 32)        320       \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 13, 13, 32)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 5408)              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 100)               540900    \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 10)                1010      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 542,230\n",
      "Trainable params: 542,230\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "#training the model\n",
    "optimiser= SGD(learning_rate= 0.005, momentum= 0.3) #the closer momentum is to 1 the quicker the moving average window moves, \n",
    "model.compile(\n",
    "    optimizer= optimiser,\n",
    "    loss= \"sparse_categorical_crossentropy\",\n",
    "    metrics= [\"accuracy\"]\n",
    ")\n",
    "\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6276201f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/15\n",
      "1875/1875 [==============================] - 20s 10ms/step - loss: 0.5244 - accuracy: 0.8619\n",
      "Epoch 2/15\n",
      "1875/1875 [==============================] - 19s 10ms/step - loss: 0.2440 - accuracy: 0.9295\n",
      "Epoch 3/15\n",
      "1875/1875 [==============================] - 20s 11ms/step - loss: 0.1922 - accuracy: 0.9430\n",
      "Epoch 4/15\n",
      "1875/1875 [==============================] - 21s 11ms/step - loss: 0.1595 - accuracy: 0.9525\n",
      "Epoch 5/15\n",
      "1875/1875 [==============================] - 21s 11ms/step - loss: 0.1355 - accuracy: 0.9596\n",
      "Epoch 6/15\n",
      "1491/1875 [======================>.......] - ETA: 4s - loss: 0.1193 - accuracy: 0.9645"
     ]
    }
   ],
   "source": [
    "model.fit(X_train, y_train, epochs= 15, batch_size= 32) #batch size of window = 32"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "94ae54fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "#running test\n",
    "plt.figure(figsize=(16, 10))\n",
    "for i in range(20):\n",
    "    image = random.choice(X_test).squeeze()\n",
    "    digit = np.argmax(model.predict(image.reshape((1, 28, 28, 1)))[0], axis=-1)\n",
    "    plot_digit(image, digit, plt, i)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "52696430",
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions= np.argmax(model.predict(X_test), axis= -1) #lets check on test data\n",
    "accuracy_score(y_test, predictions)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c8ae5fa3",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}