Question_2/Q2_CNN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams.update({'font.size': 18})\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from keras.utils.np_utils import to_categorical\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "from tensorflow.keras.models import Sequential, load_model\n",
    "from keras.datasets import mnist\n",
    "import random\n",
    "import sys\n",
    "sys.path.append('../')\n",
    "from Core_Functions import neural_networks, generate_image_set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load data and split into training and test\n",
    "(train_images, train_labels), (test_images, test_labels) = mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split training set further into training and validation sets\n",
    "sample_size = 40000\n",
    "X_train, X_val, y_train, y_val = train_images[:sample_size], train_images[sample_size:sample_size*2],  train_labels[:sample_size], train_labels[sample_size:sample_size*2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "combo_train_X, combo_train_y = generate_image_set.get_left_min_image_combinations(40000, X_train, y_train)\n",
    "combo_val_X, combo_val_y = generate_image_set.get_right_min_image_combinations(20000, X_val, y_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reshape the data so they can be fed into the network, one-hot-encode the labels\n",
    "\n",
    "train_images = combo_train_X.reshape((combo_train_X.shape[0], 28, 56, 1))\n",
    "train_images = train_images/ 255\n",
    "\n",
    "val_images = combo_val_X.reshape((combo_val_X.shape[0], 28, 56, 1))\n",
    "val_images = val_images/255\n",
    "\n",
    "train_labels = to_categorical(combo_train_y)\n",
    "val_labels = to_categorical(combo_val_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_2\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_6 (Conv2D)            (None, 26, 54, 32)        320       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_4 (MaxPooling2 (None, 13, 27, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_7 (Conv2D)            (None, 11, 25, 64)        18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_5 (MaxPooling2 (None, 5, 12, 64)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 3, 10, 64)         36928     \n",
      "_________________________________________________________________\n",
      "flatten_2 (Flatten)          (None, 1920)              0         \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 64)                122944    \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 10)                650       \n",
      "=================================================================\n",
      "Total params: 179,338\n",
      "Trainable params: 179,338\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = neural_networks.build_CNN((28, 56, 1))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = neural_networks.build_CNN((28, 56, 1))\n",
    "history = model.fit(train_images,\n",
    "          train_labels,\n",
    "          validation_data = (val_images, val_labels),\n",
    "          epochs = 20,\n",
    "          batch_size = 64,\n",
    "          verbose = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(1, len(history.history['mae']) + 1), history.history['mae'], label = 'Training', c ='mediumseagreen', lw=2.5)\n",
    "plt.plot(range(1, len(history.history['val_mae']) + 1), history.history['val_mae'], label='Validation', c = 'mediumvioletred', lw=2.5)\n",
    "\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Mean Absolute Error')\n",
    "plt.legend()\n",
    "plt.title('CNN Mean Absolute Error')\n",
    "plt.savefig('Q2_Figures/Q2_CNN_MAE.png', dpi = 400, bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(1, len(history.history['accuracy']) + 1), history.history['accuracy'], label = 'Training', c ='mediumseagreen', lw=2.5)\n",
    "plt.plot(range(1, len(history.history['val_accuracy']) + 1), history.history['val_accuracy'], label='Validation', c = 'mediumvioletred', lw=2.5)\n",
    "\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "plt.tight_layout()\n",
    "plt.title('CNN Accuracy')\n",
    "plt.savefig('Q2_Figures/Q2_CNN_Accuracy.png', dpi = 400, bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted: 3\n",
      "True: 0\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Verify above by visually inspecting images from the predicted set\n",
    "y_predict = model.predict(val_images)\n",
    "num_val_images = 10000\n",
    "\n",
    "# Run this cell repeatedly to see multiple examples\n",
    "check_index = random.randint(0,num_val_images)\n",
    "plt.imshow(val_images[check_index].reshape(28,56), cmap = plt.cm.binary)\n",
    "print(\"Predicted: \" + str(np.argmax(y_predict[check_index])))\n",
    "print(\"True: \" + str(np.argmax(val_labels[check_index])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved model to disk\n"
     ]
    }
   ],
   "source": [
    "# Save trained model for later use\n",
    "\n",
    "# serialize model to JSON\n",
    "model_json = model.to_json()\n",
    "with open(\"Q2_Trained_Models/CNN_Model.json\", \"w\") as json_file:\n",
    "    json_file.write(model_json)\n",
    "# serialize weights to HDF5\n",
    "model.save_weights(\"Q2_Trained_Models/CNN_Model.h5\")\n",
    "print(\"Saved model to disk\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicted: 3\n",
      "True: 3\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Verify above by visually inspecting images from the predicted set\n",
    "y_predict = model.predict(train_images)\n",
    "num_val_images = 10000\n",
    "\n",
    "# Run this cell repeatedly to see multiple examples\n",
    "check_index = random.randint(0,num_val_images)\n",
    "plt.imshow(train_images[check_index].reshape(28,56), cmap = plt.cm.binary)\n",
    "print(\"Predicted: \" + str(np.argmax(y_predict[check_index])))\n",
    "print(\"True: \" + str(np.argmax(train_labels[check_index])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "general_ML",
   "language": "python",
   "name": "general_ml"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}