CNN_MNIST_KERAS.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 165
    },
    "colab_type": "code",
    "id": "z_3cM_LJhM54",
    "outputId": "38a8172a-5ad4-429f-a974-fe2719c4eb06"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://s3.amazonaws.com/img-datasets/mnist.npz\n",
      "11493376/11490434 [==============================] - 0s 0us/step\n",
      "X_train shape:  (60000, 28, 28)\n",
      "Y_train shape:  (60000,)\n",
      "--------------------------------------------------\n",
      "X_test shape:  (10000, 28, 28)\n",
      "y_test shape:  (10000,)\n"
     ]
    }
   ],
   "source": [
    "from __future__ import print_function\n",
    "import keras\n",
    "from keras.datasets import mnist\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout, Flatten, Activation\n",
    "from keras.layers import Conv2D, MaxPooling2D\n",
    "from keras import backend as K\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "(X_train,y_train),(X_test,y_test)=mnist.load_data()\n",
    "print('X_train shape: ',X_train.shape)\n",
    "print('Y_train shape: ',y_train.shape)\n",
    "print('-'*50)\n",
    "print('X_test shape: ',X_test.shape)\n",
    "print('y_test shape: ',y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 72
    },
    "colab_type": "code",
    "id": "Xo3sD1G_hOGb",
    "outputId": "46bbf4a3-d02c-4dad-d163-da8bfb3dbbcc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train shape: (60000, 28, 28, 1)\n",
      "60000 train samples\n",
      "10000 test samples\n"
     ]
    }
   ],
   "source": [
    "# input image dimensions\n",
    "img_rows, img_cols = 28, 28\n",
    "if K.image_data_format() == 'channels_first':\n",
    "    X_train = X_train.reshape(X_train.shape[0], 1, img_rows, img_cols)\n",
    "    X_test = X_test.reshape(X_test.shape[0], 1, img_rows, img_cols)\n",
    "    input_shape = (1, img_rows, img_cols)\n",
    "else:\n",
    "    X_train = X_train.reshape(X_train.shape[0], img_rows, img_cols, 1)\n",
    "    X_test = X_test.reshape(X_test.shape[0], img_rows, img_cols, 1)\n",
    "    input_shape = (img_rows, img_cols, 1)\n",
    "\n",
    "X_train = X_train.astype('float32')\n",
    "X_test = X_test.astype('float32')\n",
    "X_train /= 255\n",
    "X_test /= 255\n",
    "print('X_train shape:', X_train.shape)\n",
    "print(X_train.shape[0], 'train samples')\n",
    "print(X_test.shape[0], 'test samples')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "colab_type": "code",
    "id": "dgU-UBdfh0P_",
    "outputId": "d5431065-ed05-4da6-b693-2cd4526c75fc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0. 0. 0. 0. 0. 1. 0. 0. 0. 0.]\n"
     ]
    }
   ],
   "source": [
    "#Some initialization\n",
    "batch_size = 128\n",
    "num_classes = 10\n",
    "epochs = 12\n",
    "# convert class vectors to binary class matrices\n",
    "y_train = keras.utils.to_categorical(y_train, num_classes)\n",
    "y_test = keras.utils.to_categorical(y_test, num_classes)\n",
    "\n",
    "print(y_train[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "1a_iar6jn2Hr"
   },
   "source": [
    "**Applying CNN with 3 layers**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 589
    },
    "colab_type": "code",
    "id": "ZosfS1XjiBQ2",
    "outputId": "3e91eebc-1d6a-4da6-a089-eff7179c3dc4"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_13 (Conv2D)           (None, 26, 26, 32)        320       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_13 (MaxPooling (None, 13, 13, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_14 (Conv2D)           (None, 11, 11, 64)        18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_14 (MaxPooling (None, 5, 5, 64)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_15 (Conv2D)           (None, 3, 3, 64)          36928     \n",
      "_________________________________________________________________\n",
      "batch_normalization_9 (Batch (None, 3, 3, 64)          12        \n",
      "_________________________________________________________________\n",
      "max_pooling2d_15 (MaxPooling (None, 1, 1, 64)          0         \n",
      "_________________________________________________________________\n",
      "flatten_2 (Flatten)          (None, 64)                0         \n",
      "_________________________________________________________________\n",
      "dense_9 (Dense)              (None, 512)               33280     \n",
      "_________________________________________________________________\n",
      "activation_8 (Activation)    (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dropout_5 (Dropout)          (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_10 (Dense)             (None, 10)                5130      \n",
      "=================================================================\n",
      "Total params: 94,166\n",
      "Trainable params: 94,160\n",
      "Non-trainable params: 6\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from keras.layers.normalization import BatchNormalization\n",
    "#Conv2d-->Pooling-->Conv2d-->Pooling-->Conv2d-->BN-->Pooling-->FC-->Dropout-->Softmax\n",
    "#We will use Relu as Activation function and Adam as Optimizer\n",
    "#We will also use BatchNormalization and Dropout at suitable places\n",
    "\n",
    "model=Sequential()\n",
    "#Adding first layer\n",
    "model.add(Conv2D(32, kernel_size=(3, 3),activation='relu',input_shape=input_shape))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "#Second layer\n",
    "model.add(Conv2D(64, kernel_size=(3, 3), activation='relu'))\n",
    "\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "#third layer\n",
    "model.add(Conv2D(64, kernel_size=(3, 3), activation='relu'))\n",
    "model.add(BatchNormalization(axis=1))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "#Fully Connected layer\n",
    "#Flattening\n",
    "model.add(Flatten())\n",
    "model.add(Dense(512))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.5))\n",
    "\n",
    "#Softmax\n",
    "model.add(Dense(num_classes, activation='softmax'))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 554
    },
    "colab_type": "code",
    "id": "_xNm3ynXsHCv",
    "outputId": "e2fb8db6-e486-41c7-cbcb-599127bf772b"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0629 19:38:00.390176 140361178707840 deprecation.py:323] From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/math_grad.py:1250: add_dispatch_support.<locals>.wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/12\n",
      "60000/60000 [==============================] - 51s 853us/step - loss: 0.3726 - acc: 0.8831 - val_loss: 0.1126 - val_acc: 0.9657\n",
      "Epoch 2/12\n",
      "60000/60000 [==============================] - 50s 831us/step - loss: 0.1063 - acc: 0.9674 - val_loss: 0.0756 - val_acc: 0.9752\n",
      "Epoch 3/12\n",
      "60000/60000 [==============================] - 50s 829us/step - loss: 0.0760 - acc: 0.9775 - val_loss: 0.0644 - val_acc: 0.9805\n",
      "Epoch 4/12\n",
      "60000/60000 [==============================] - 50s 825us/step - loss: 0.0597 - acc: 0.9818 - val_loss: 0.0506 - val_acc: 0.9844\n",
      "Epoch 5/12\n",
      "60000/60000 [==============================] - 49s 816us/step - loss: 0.0504 - acc: 0.9846 - val_loss: 0.0596 - val_acc: 0.9822\n",
      "Epoch 6/12\n",
      "60000/60000 [==============================] - 49s 812us/step - loss: 0.0423 - acc: 0.9867 - val_loss: 0.0524 - val_acc: 0.9846\n",
      "Epoch 7/12\n",
      "60000/60000 [==============================] - 49s 809us/step - loss: 0.0373 - acc: 0.9885 - val_loss: 0.0487 - val_acc: 0.9859\n",
      "Epoch 8/12\n",
      "60000/60000 [==============================] - 49s 821us/step - loss: 0.0323 - acc: 0.9901 - val_loss: 0.0471 - val_acc: 0.9856\n",
      "Epoch 9/12\n",
      "60000/60000 [==============================] - 51s 842us/step - loss: 0.0287 - acc: 0.9908 - val_loss: 0.0536 - val_acc: 0.9835\n",
      "Epoch 10/12\n",
      "60000/60000 [==============================] - 51s 852us/step - loss: 0.0243 - acc: 0.9922 - val_loss: 0.0520 - val_acc: 0.9859\n",
      "Epoch 11/12\n",
      "60000/60000 [==============================] - 51s 850us/step - loss: 0.0233 - acc: 0.9927 - val_loss: 0.0437 - val_acc: 0.9872\n",
      "Epoch 12/12\n",
      "60000/60000 [==============================] - 51s 852us/step - loss: 0.0196 - acc: 0.9934 - val_loss: 0.0416 - val_acc: 0.9876\n"
     ]
    }
   ],
   "source": [
    "#Compiling with Cross-entropy-loss and Adam Optimizer taking Accuracy as a metric\n",
    "from keras.optimizers import Adam\n",
    "model.compile(loss='categorical_crossentropy', optimizer=Adam(), metrics=['accuracy'])\n",
    "history=model.fit(X_train, y_train,batch_size=batch_size,epochs=epochs,verbose=1,validation_data=(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Z_Kk6xnes9yi"
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "def plt_dynamic(x, vy, ty, ax, colors=['b']):\n",
    "    ax.plot(x, vy, 'b', label=\"Validation Loss\")\n",
    "    ax.plot(x, ty, 'r', label=\"Train Loss\")\n",
    "    plt.legend()\n",
    "    plt.grid()\n",
    "    fig.canvas.draw()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 332
    },
    "colab_type": "code",
    "id": "zCU3DFYpyYAz",
    "outputId": "a6567bc8-a931-44c1-8ff5-b6fdcf3d809f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test score: 0.0416184305918403\n",
      "Test accuracy: 0.9876\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nJiJLFsYYYyL6/0BXniXLCuJTAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "score = model.evaluate(X_test, y_test, verbose=0) \n",
    "print('Test score:', score[0]) \n",
    "print('Test accuracy:', score[1])\n",
    "\n",
    "fig,ax = plt.subplots(1,1)\n",
    "ax.set_title('Epoch vs Loss')\n",
    "ax.set_xlabel('epoch') ; ax.set_ylabel('Categorical Crossentropy Loss')\n",
    "\n",
    "# list of epoch numbers\n",
    "x = list(range(1,epochs+1))\n",
    "\n",
    "\n",
    "vy = history.history['val_loss']\n",
    "ty = history.history['loss']\n",
    "plt_dynamic(x, vy, ty, ax)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "cDJwkmcbzLnj"
   },
   "source": [
    "**Applying CNN with 5 layers**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 811
    },
    "colab_type": "code",
    "id": "mV2Cvnspydyu",
    "outputId": "acc6efb5-0b11-4d64-9651-8ccf3246c027"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_95 (Conv2D)           (None, 27, 27, 64)        320       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_93 (MaxPooling (None, 27, 27, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_96 (Conv2D)           (None, 26, 26, 64)        16448     \n",
      "_________________________________________________________________\n",
      "dropout_28 (Dropout)         (None, 26, 26, 64)        0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_94 (MaxPooling (None, 13, 13, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_97 (Conv2D)           (None, 11, 11, 32)        18464     \n",
      "_________________________________________________________________\n",
      "batch_normalization_43 (Batc (None, 11, 11, 32)        44        \n",
      "_________________________________________________________________\n",
      "max_pooling2d_95 (MaxPooling (None, 11, 11, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_98 (Conv2D)           (None, 10, 10, 32)        4128      \n",
      "_________________________________________________________________\n",
      "batch_normalization_44 (Batc (None, 10, 10, 32)        40        \n",
      "_________________________________________________________________\n",
      "max_pooling2d_96 (MaxPooling (None, 5, 5, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_99 (Conv2D)           (None, 4, 4, 16)          2064      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_97 (MaxPooling (None, 2, 2, 16)          0         \n",
      "_________________________________________________________________\n",
      "flatten_9 (Flatten)          (None, 64)                0         \n",
      "_________________________________________________________________\n",
      "dense_23 (Dense)             (None, 512)               33280     \n",
      "_________________________________________________________________\n",
      "activation_15 (Activation)   (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dropout_29 (Dropout)         (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_24 (Dense)             (None, 10)                5130      \n",
      "=================================================================\n",
      "Total params: 79,918\n",
      "Trainable params: 79,876\n",
      "Non-trainable params: 42\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "#Conv2d-->Pooling-->Conv2d-->Dropout-->Pooling-->Conv2d-->BN-->Pooling-->Dropout-->Conv2D-->BN-->Pooling-->Conv2D-->Pooling-->FC-->Dropout-->Softmax\n",
    "#We will use Relu as Activation function and Adam as Optimizer\n",
    "#We will also use BatchNormalization and Dropout at suitable places\n",
    "\n",
    "model=Sequential()\n",
    "#Adding first layer\n",
    "model.add(Conv2D(64, kernel_size=(2, 2),activation='relu',input_shape=input_shape))\n",
    "model.add(MaxPooling2D(pool_size=(1, 1)))\n",
    "\n",
    "#Second layer\n",
    "model.add(Conv2D(64, kernel_size=(2, 2), activation='relu'))\n",
    "model.add(Dropout(0.25))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "#third layer\n",
    "model.add(Conv2D(32, kernel_size=(3, 3), activation='relu'))\n",
    "model.add(BatchNormalization(axis=1))\n",
    "model.add(MaxPooling2D(pool_size=(1,1)))\n",
    "\n",
    "#Fourth Layer\n",
    "model.add(Conv2D(32,kernel_size=(2,2),activation='relu'))\n",
    "model.add(BatchNormalization(axis=1))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "#Fifth Layer\n",
    "model.add(Conv2D(16,kernel_size=(2,2),activation='relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "#Fully Connected layer\n",
    "#Flattening\n",
    "model.add(Flatten())\n",
    "model.add(Dense(512))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.5))\n",
    "\n",
    "#Softmax\n",
    "model.add(Dense(num_classes, activation='softmax'))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 479
    },
    "colab_type": "code",
    "id": "JcOBpqfX2y3J",
    "outputId": "0c21377d-087c-4c50-fd06-f05bea4e9e54"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/12\n",
      "60000/60000 [==============================] - 246s 4ms/step - loss: 0.2913 - acc: 0.9065 - val_loss: 0.0726 - val_acc: 0.9789\n",
      "Epoch 2/12\n",
      "60000/60000 [==============================] - 239s 4ms/step - loss: 0.0743 - acc: 0.9771 - val_loss: 0.0622 - val_acc: 0.9821\n",
      "Epoch 3/12\n",
      "60000/60000 [==============================] - 237s 4ms/step - loss: 0.0514 - acc: 0.9844 - val_loss: 0.0655 - val_acc: 0.9805\n",
      "Epoch 4/12\n",
      "60000/60000 [==============================] - 241s 4ms/step - loss: 0.0436 - acc: 0.9867 - val_loss: 0.0488 - val_acc: 0.9844\n",
      "Epoch 5/12\n",
      "60000/60000 [==============================] - 237s 4ms/step - loss: 0.0366 - acc: 0.9886 - val_loss: 0.0416 - val_acc: 0.9867\n",
      "Epoch 6/12\n",
      "60000/60000 [==============================] - 237s 4ms/step - loss: 0.0322 - acc: 0.9900 - val_loss: 0.0430 - val_acc: 0.9864\n",
      "Epoch 7/12\n",
      "60000/60000 [==============================] - 237s 4ms/step - loss: 0.0294 - acc: 0.9910 - val_loss: 0.0307 - val_acc: 0.9905\n",
      "Epoch 8/12\n",
      "60000/60000 [==============================] - 239s 4ms/step - loss: 0.0273 - acc: 0.9913 - val_loss: 0.0266 - val_acc: 0.9925\n",
      "Epoch 9/12\n",
      "60000/60000 [==============================] - 237s 4ms/step - loss: 0.0253 - acc: 0.9919 - val_loss: 0.0267 - val_acc: 0.9917\n",
      "Epoch 10/12\n",
      "60000/60000 [==============================] - 237s 4ms/step - loss: 0.0242 - acc: 0.9926 - val_loss: 0.0328 - val_acc: 0.9906\n",
      "Epoch 11/12\n",
      "60000/60000 [==============================] - 237s 4ms/step - loss: 0.0209 - acc: 0.9930 - val_loss: 0.0309 - val_acc: 0.9904\n",
      "Epoch 12/12\n",
      "60000/60000 [==============================] - 237s 4ms/step - loss: 0.0196 - acc: 0.9939 - val_loss: 0.0365 - val_acc: 0.9908\n"
     ]
    }
   ],
   "source": [
    "#Compiling with Cross-entropy-loss and Adam Optimizer taking Accuracy as a metric\n",
    "\n",
    "model.compile(loss='categorical_crossentropy', optimizer=Adam(), metrics=['accuracy'])\n",
    "history=model.fit(X_train, y_train,batch_size=batch_size,epochs=epochs,verbose=1,validation_data=(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 332
    },
    "colab_type": "code",
    "id": "yatN7eM94Aqm",
    "outputId": "6275db7c-f071-42a0-a14c-6c2a7180588e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test score: 0.03646882001824051\n",
      "Test accuracy: 0.9908\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEWCAYAAACXGLsWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "score = model.evaluate(X_test, y_test, verbose=0) \n",
    "print('Test score:', score[0]) \n",
    "print('Test accuracy:', score[1])\n",
    "\n",
    "fig,ax = plt.subplots(1,1)\n",
    "ax.set_title('Epoch vs Loss')\n",
    "ax.set_xlabel('epoch') ; ax.set_ylabel('Categorical Crossentropy Loss')\n",
    "\n",
    "# list of epoch numbers\n",
    "x = list(range(1,epochs+1))\n",
    "\n",
    "\n",
    "vy = history.history['val_loss']\n",
    "ty = history.history['loss']\n",
    "plt_dynamic(x, vy, ty, ax)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "AAh9VFISDJs9"
   },
   "source": [
    "**Applying CNN with 7 layers**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "OHkiaRwHC_vM",
    "outputId": "5ca90ff7-d49d-4cf9-fbb9-6e4c54676d41"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_333 (Conv2D)          (None, 24, 24, 64)        1664      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_325 (MaxPoolin (None, 24, 24, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_334 (Conv2D)          (None, 22, 22, 64)        36928     \n",
      "_________________________________________________________________\n",
      "batch_normalization_142 (Bat (None, 22, 22, 64)        88        \n",
      "_________________________________________________________________\n",
      "max_pooling2d_326 (MaxPoolin (None, 11, 11, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_335 (Conv2D)          (None, 9, 9, 32)          18464     \n",
      "_________________________________________________________________\n",
      "dropout_122 (Dropout)        (None, 9, 9, 32)          0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_327 (MaxPoolin (None, 9, 9, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_336 (Conv2D)          (None, 7, 7, 32)          9248      \n",
      "_________________________________________________________________\n",
      "dropout_123 (Dropout)        (None, 7, 7, 32)          0         \n",
      "_________________________________________________________________\n",
      "max_pooling2d_328 (MaxPoolin (None, 7, 7, 32)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_337 (Conv2D)          (None, 6, 6, 16)          2064      \n",
      "_________________________________________________________________\n",
      "batch_normalization_143 (Bat (None, 6, 6, 16)          24        \n",
      "_________________________________________________________________\n",
      "max_pooling2d_329 (MaxPoolin (None, 6, 6, 16)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_338 (Conv2D)          (None, 5, 5, 16)          1040      \n",
      "_________________________________________________________________\n",
      "batch_normalization_144 (Bat (None, 5, 5, 16)          20        \n",
      "_________________________________________________________________\n",
      "max_pooling2d_330 (MaxPoolin (None, 5, 5, 16)          0         \n",
      "_________________________________________________________________\n",
      "conv2d_339 (Conv2D)          (None, 3, 3, 8)           1160      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_331 (MaxPoolin (None, 1, 1, 8)           0         \n",
      "_________________________________________________________________\n",
      "flatten_33 (Flatten)         (None, 8)                 0         \n",
      "_________________________________________________________________\n",
      "dense_71 (Dense)             (None, 128)               1152      \n",
      "_________________________________________________________________\n",
      "activation_39 (Activation)   (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dropout_124 (Dropout)        (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense_72 (Dense)             (None, 10)                1290      \n",
      "=================================================================\n",
      "Total params: 73,142\n",
      "Trainable params: 73,076\n",
      "Non-trainable params: 66\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "#Conv2d-->Pooling-->Conv2d-->BN-->Pooling-->Conv2d-->Dropout-->Pooling-->Conv2d-->Dropout-->Pooling-->Conv2d-->BN-->Pooling-->Dropout-->Conv2D-->BN-->Pooling-->Conv2D-->Pooling-->FC-->Dropout-->Softmax\n",
    "#We will use Relu as Activation function and Adam as Optimizer\n",
    "#We will also use BatchNormalization and Dropout at suitable places\n",
    "\n",
    "model=Sequential()\n",
    "#Adding first layer\n",
    "model.add(Conv2D(64, kernel_size=(5, 5),activation='relu',input_shape=input_shape))\n",
    "model.add(MaxPooling2D(pool_size=(1, 1)))\n",
    "\n",
    "#Adding Second Layer\n",
    "model.add(Conv2D(64,kernel_size=(3,3),activation='relu'))\n",
    "model.add(BatchNormalization(axis=1))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "#Adding Third Layer\n",
    "model.add(Conv2D(32,kernel_size=(3,3),activation='relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(MaxPooling2D(pool_size=(1,1)))\n",
    "\n",
    "#Fourth layer\n",
    "model.add(Conv2D(32, kernel_size=(3, 3), activation='relu'))\n",
    "model.add(Dropout(0.25))\n",
    "model.add(MaxPooling2D(pool_size=(1,1)))\n",
    "\n",
    "#Fifth layer\n",
    "model.add(Conv2D(16, kernel_size=(2, 2), activation='relu'))\n",
    "model.add(BatchNormalization(axis=1))\n",
    "model.add(MaxPooling2D(pool_size=(1,1)))\n",
    "\n",
    "#Sixth Layer\n",
    "model.add(Conv2D(16,kernel_size=(2,2),activation='relu'))\n",
    "model.add(BatchNormalization(axis=1))\n",
    "model.add(MaxPooling2D(pool_size=(1,1)))\n",
    "\n",
    "#Seventh Layer\n",
    "model.add(Conv2D(8,kernel_size=(3,3),activation='relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "\n",
    "#Fully Connected layer\n",
    "#Flattening\n",
    "model.add(Flatten())\n",
    "model.add(Dense(128))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dropout(0.5))\n",
    "\n",
    "#Softmax\n",
    "model.add(Dense(num_classes, activation='softmax'))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 479
    },
    "colab_type": "code",
    "id": "r5422nDnEcz-",
    "outputId": "79384732-4e84-41fa-ac08-5be4ab87f131"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/12\n",
      "60000/60000 [==============================] - 294s 5ms/step - loss: 0.8691 - acc: 0.6921 - val_loss: 0.1066 - val_acc: 0.9679\n",
      "Epoch 2/12\n",
      "60000/60000 [==============================] - 288s 5ms/step - loss: 0.1616 - acc: 0.9552 - val_loss: 0.0551 - val_acc: 0.9826\n",
      "Epoch 3/12\n",
      "60000/60000 [==============================] - 288s 5ms/step - loss: 0.1080 - acc: 0.9702 - val_loss: 0.0644 - val_acc: 0.9813\n",
      "Epoch 4/12\n",
      "60000/60000 [==============================] - 285s 5ms/step - loss: 0.0826 - acc: 0.9773 - val_loss: 0.0385 - val_acc: 0.9879\n",
      "Epoch 5/12\n",
      "60000/60000 [==============================] - 288s 5ms/step - loss: 0.0736 - acc: 0.9794 - val_loss: 0.0404 - val_acc: 0.9888\n",
      "Epoch 6/12\n",
      "60000/60000 [==============================] - 284s 5ms/step - loss: 0.0646 - acc: 0.9819 - val_loss: 0.0340 - val_acc: 0.9904\n",
      "Epoch 7/12\n",
      "60000/60000 [==============================] - 288s 5ms/step - loss: 0.0575 - acc: 0.9840 - val_loss: 0.0312 - val_acc: 0.9904\n",
      "Epoch 8/12\n",
      "60000/60000 [==============================] - 294s 5ms/step - loss: 0.0543 - acc: 0.9847 - val_loss: 0.0328 - val_acc: 0.9914\n",
      "Epoch 9/12\n",
      "60000/60000 [==============================] - 294s 5ms/step - loss: 0.0500 - acc: 0.9857 - val_loss: 0.0419 - val_acc: 0.9895\n",
      "Epoch 10/12\n",
      "60000/60000 [==============================] - 294s 5ms/step - loss: 0.0482 - acc: 0.9863 - val_loss: 0.0298 - val_acc: 0.9927\n",
      "Epoch 11/12\n",
      "60000/60000 [==============================] - 294s 5ms/step - loss: 0.0481 - acc: 0.9865 - val_loss: 0.0300 - val_acc: 0.9911\n",
      "Epoch 12/12\n",
      "60000/60000 [==============================] - 295s 5ms/step - loss: 0.0417 - acc: 0.9887 - val_loss: 0.0322 - val_acc: 0.9914\n"
     ]
    }
   ],
   "source": [
    "#Compiling with Cross-entropy-loss and Adam Optimizer taking Accuracy as a metric\n",
    "\n",
    "model.compile(loss='categorical_crossentropy', optimizer=Adam(), metrics=['accuracy'])\n",
    "history=model.fit(X_train, y_train,batch_size=batch_size,epochs=epochs,verbose=1,validation_data=(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 332
    },
    "colab_type": "code",
    "id": "Etks05viGntN",
    "outputId": "dbc25804-537d-4dc7-8b1d-3049a9d2bd08"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test score: 0.032224515436636285\n",
      "Test accuracy: 0.9914\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "score = model.evaluate(X_test, y_test, verbose=0) \n",
    "print('Test score:', score[0]) \n",
    "print('Test accuracy:', score[1])\n",
    "\n",
    "fig,ax = plt.subplots(1,1)\n",
    "ax.set_title('Epoch vs Loss')\n",
    "ax.set_xlabel('epoch') ; ax.set_ylabel('Categorical Crossentropy Loss')\n",
    "\n",
    "# list of epoch numbers\n",
    "x = list(range(1,epochs+1))\n",
    "\n",
    "\n",
    "vy = history.history['val_loss']\n",
    "ty = history.history['loss']\n",
    "plt_dynamic(x, vy, ty, ax)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "DfA-PxoHULhb"
   },
   "source": [
    "**Observations**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 146
    },
    "colab_type": "code",
    "id": "r7JOkY59T9Ta",
    "outputId": "5ab00eb2-712c-4695-b5e1-96c564ee3074"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+--------------+------------+\n",
      "| No of layers | Accuracy % |\n",
      "+--------------+------------+\n",
      "|      3       |   98.76    |\n",
      "|      5       |   99.08    |\n",
      "|      7       |   99.14    |\n",
      "+--------------+------------+\n"
     ]
    }
   ],
   "source": [
    "from prettytable import PrettyTable    \n",
    "x = PrettyTable()\n",
    "x.field_names = [\"No of layers\", \"Accuracy %\"]\n",
    "\n",
    "x.add_row([\"3\",\"98.76\"])\n",
    "x.add_row([\"5\",\"99.08\"])\n",
    "x.add_row([\"7\",\"99.14\"])\n",
    "\n",
    "\n",
    "\n",
    "print(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Fa8JoCxOUxuK"
   },
   "source": [
    "**Our 3 layer model arranged in this order:**\n",
    "\n",
    "*Conv2d-->Max Pooling-->Conv2d-->Max Pooling-->Conv2d-->BN-->Max      Pooling-->FC-->Dropout-->Softmax*\n",
    "\n",
    "**Our 5 layer model arranged in :** \n",
    " \n",
    "*Conv2d-->Pooling-->Conv2d-->Dropout-->Pooling-->Conv2d-->BN-->Pooling-->Dropout-->Conv2D-->BN-->Pooling-->Conv2D-->Pooling-->FC-->Dropout-->Softmax*\n",
    "\n",
    "**Our 7 layer model arranged in:** \n",
    "\n",
    "*Conv2d-->Pooling-->Conv2d-->BN-->Pooling-->Conv2d-->Dropout-->Pooling-->Conv2d-->Dropout-->Pooling-->Conv2d-->BN-->Pooling-->Dropout-->Conv2D-->BN-->Pooling-->Conv2D-->Pooling-->FC-->Dropout-->Softmax*\n",
    "\n",
    "Where BN= Batch Normalization and FC= Fully Connected.\n",
    "\n",
    "We used different kernel_size and pool_size for our different layers and we can observe that after increasing number of layers our accuracy increases. We applied batch normalisation not at all layers to reduce internal covariate shift."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "-9in8ivZUvon"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "CNN_MNIST_KERAS.ipynb",
   "provenance": [],
   "version": "0.3.2"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}