MNIST_Fashon_CNN_using_Keras.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<b> A notebook for the classification of the MNIST fashion dataset.\n",
    "\n",
    "I used Sequential Convolutional Neural Networks (CNN) model using Keras and tensorflow as a backend.\n",
    "\n",
    "\n",
    "\n",
    "Result : |Training Accuracy | Dev(Validation) Accuracy| Testing Accuracy |\n",
    "---------------------------------------------------------------------|\n",
    "         | 92.78 %          | 91.47 %             | 92.00 %          |\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import datetime\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import keras\n",
    "from keras.utils import to_categorical\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense,Dropout, Activation, Flatten\n",
    "from keras.layers import Convolution2D, MaxPooling2D\n",
    "from keras.utils import np_utils\n",
    "from keras import backend as K\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'tf'"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras.backend.image_dim_ordering()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "batch_size     = 32\n",
    "nb_classes     = 10\n",
    "nb_ephochs     = 50\n",
    "img_rows, img_cols = 28,28\n",
    "nb_filters     = 32\n",
    "pool_size      = 2\n",
    "kernel_size    = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "if K.image_dim_ordering() == 'th':\n",
    "    input_shape = (1,img_rows, img_cols)\n",
    "else:\n",
    "    input_shape = (img_rows,img_cols,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_train = pd.read_csv('input_data/fashion-mnist_train.csv')\n",
    "data_test = pd.read_csv('input_data/fashion-mnist_test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>label</th>\n",
       "      <th>pixel1</th>\n",
       "      <th>pixel2</th>\n",
       "      <th>pixel3</th>\n",
       "      <th>pixel4</th>\n",
       "      <th>pixel5</th>\n",
       "      <th>pixel6</th>\n",
       "      <th>pixel7</th>\n",
       "      <th>pixel8</th>\n",
       "      <th>pixel9</th>\n",
       "      <th>...</th>\n",
       "      <th>pixel775</th>\n",
       "      <th>pixel776</th>\n",
       "      <th>pixel777</th>\n",
       "      <th>pixel778</th>\n",
       "      <th>pixel779</th>\n",
       "      <th>pixel780</th>\n",
       "      <th>pixel781</th>\n",
       "      <th>pixel782</th>\n",
       "      <th>pixel783</th>\n",
       "      <th>pixel784</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>30</td>\n",
       "      <td>43</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 785 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   label  pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  pixel8  \\\n",
       "0      2       0       0       0       0       0       0       0       0   \n",
       "1      9       0       0       0       0       0       0       0       0   \n",
       "2      6       0       0       0       0       0       0       0       5   \n",
       "3      0       0       0       0       1       2       0       0       0   \n",
       "4      3       0       0       0       0       0       0       0       0   \n",
       "\n",
       "   pixel9    ...     pixel775  pixel776  pixel777  pixel778  pixel779  \\\n",
       "0       0    ...            0         0         0         0         0   \n",
       "1       0    ...            0         0         0         0         0   \n",
       "2       0    ...            0         0         0        30        43   \n",
       "3       0    ...            3         0         0         0         0   \n",
       "4       0    ...            0         0         0         0         0   \n",
       "\n",
       "   pixel780  pixel781  pixel782  pixel783  pixel784  \n",
       "0         0         0         0         0         0  \n",
       "1         0         0         0         0         0  \n",
       "2         0         0         0         0         0  \n",
       "3         1         0         0         0         0  \n",
       "4         0         0         0         0         0  \n",
       "\n",
       "[5 rows x 785 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "clothes = {0: 'T-shirt/top',\n",
    "           1: 'Trouser',\n",
    "           2: 'Pullover',\n",
    "           3: 'Dress',\n",
    "           4: 'Coat',\n",
    "           5: 'Sandal',\n",
    "           6: 'Shirt',\n",
    "           7: 'Sneaker',\n",
    "           8: 'Bag',\n",
    "           9: 'Ankle boot'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsgAAAJoCAYAAAB7txeYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDCbIAAAAQMb/Beof7nqPtkeDAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x3f819f78d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(3, 4,figsize=(10,10))\n",
    "for row in axes:\n",
    "    for axe in row:\n",
    "        index = np.random.randint(10000)\n",
    "        img = data_train.iloc[index, 1:].values.reshape((28, 28))\n",
    "        obj = data_train.iloc[index, 0]\n",
    "        axe.imshow(img, cmap='gray')\n",
    "        axe.set_title(clothes[obj])\n",
    "        plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X = data_train.iloc[:,1:].values\n",
    "y = data_train.iloc[:,0].values\n",
    "y = to_categorical(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.2, random_state=101)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_test = data_test.iloc[:,1:].values\n",
    "y_test = data_test.iloc[:,0]\n",
    "y_test  = to_categorical(y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reshaping Data +  Normalization "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(48000, 784)\n",
      "(12000, 784)\n",
      "(10000, 784)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)\n",
    "print(X_val.shape)\n",
    "print(X_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_train = X_train.reshape(X_train.shape[0], img_rows, img_cols, 1)\n",
    "X_val = X_val.reshape(X_val.shape[0], img_rows, img_cols, 1)\n",
    "X_test = X_test.reshape(X_test.shape[0], img_rows, img_cols, 1)\n",
    "\n",
    "\n",
    "X_train = X_train.astype('float32')\n",
    "X_val = X_val.astype('float32')\n",
    "X_test = X_test.astype('float32')\n",
    "\n",
    "X_train = X_train  / 255\n",
    "X_val   = X_val   / 255\n",
    "X_test  = X_test / 255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(48000, 28, 28, 1)\n",
      "(12000, 28, 28, 1)\n",
      "(10000, 28, 28, 1)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)\n",
    "print(X_val.shape)\n",
    "print(X_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1.0, 1.0, 1.0)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.max(),X_val.max(), X_test.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 0.0, 0.0)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.min(),X_val.min(), X_test.min()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model Architecture\n",
    "\n",
    "## Now we need to follow the following steps:\n",
    "    \n",
    "    1. Convolution \n",
    "    2. Max Pooling\n",
    "    3. Convolution\n",
    "    4. Max pooling\n",
    "    5. Dropout\n",
    "    6  Flattening\n",
    "    7. Full Connection (Dense)\n",
    "    8. Dropout\n",
    "    9. Full Connection\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\agebresi\\AppData\\Local\\Continuum\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:11: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(activation=\"relu\", units=128)`\n",
      "  # This is added back by InteractiveShellApp.init_path()\n",
      "C:\\Users\\agebresi\\AppData\\Local\\Continuum\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:13: UserWarning: Update your `Dense` call to the Keras 2 API: `Dense(activation=\"softmax\", units=10)`\n",
      "  del sys.path[0]\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(Convolution2D(filters= nb_filters,kernel_size=(kernel_size,kernel_size),input_shape = input_shape,activation='relu'))\n",
    "model.add(MaxPooling2D(pool_size=(pool_size,pool_size)))\n",
    "\n",
    "model.add(Convolution2D(filters=nb_filters, kernel_size=(kernel_size,kernel_size),activation='relu'))\n",
    "model.add(MaxPooling2D(pool_size=(pool_size,pool_size)))\n",
    "\n",
    "model.add(Dropout(0.25))\n",
    "model.add(Flatten())\n",
    "\n",
    "model.add(Dense(output_dim = 128 , activation=\"relu\"))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(output_dim = nb_classes, activation='softmax'))\n",
    "model.compile(loss = 'categorical_crossentropy', optimizer = 'adam', metrics = ['accuracy'])\n",
    "\n",
    "#model.compile(loss = keras.losses.categorical_crossentropy, optimizer= keras.optimizers.adam(), metrics = ['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_1 (Conv2D)            (None, 26, 26, 32)        320       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 13, 13, 32)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 11, 11, 32)        9248      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 5, 5, 32)          0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 5, 5, 32)          0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 800)               0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 128)               102528    \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 128)               0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 10)                1290      \n",
      "=================================================================\n",
      "Total params: 113,386\n",
      "Trainable params: 113,386\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "print(model.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 48000 samples, validate on 12000 samples\n",
      "Epoch 1/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.6432 - acc: 0.7668 - val_loss: 0.4000 - val_acc: 0.8521\n",
      "Epoch 2/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.4405 - acc: 0.8391 - val_loss: 0.3499 - val_acc: 0.8697\n",
      "Epoch 3/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.3869 - acc: 0.8594 - val_loss: 0.3297 - val_acc: 0.8792\n",
      "Epoch 4/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.3580 - acc: 0.8700 - val_loss: 0.2996 - val_acc: 0.8909\n",
      "Epoch 5/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.3310 - acc: 0.8764 - val_loss: 0.2853 - val_acc: 0.8928\n",
      "Epoch 6/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.3204 - acc: 0.8833 - val_loss: 0.2917 - val_acc: 0.8925\n",
      "Epoch 7/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.3085 - acc: 0.8870 - val_loss: 0.2756 - val_acc: 0.8974\n",
      "Epoch 8/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2986 - acc: 0.8904 - val_loss: 0.2643 - val_acc: 0.9054\n",
      "Epoch 9/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2848 - acc: 0.8935 - val_loss: 0.2572 - val_acc: 0.9072\n",
      "Epoch 10/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2841 - acc: 0.8948 - val_loss: 0.2573 - val_acc: 0.9050\n",
      "Epoch 11/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2741 - acc: 0.8985 - val_loss: 0.2584 - val_acc: 0.9037\n",
      "Epoch 12/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2702 - acc: 0.9009 - val_loss: 0.2496 - val_acc: 0.9109\n",
      "Epoch 13/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2638 - acc: 0.9021 - val_loss: 0.2533 - val_acc: 0.9065\n",
      "Epoch 14/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2599 - acc: 0.9037 - val_loss: 0.2483 - val_acc: 0.9105\n",
      "Epoch 15/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2539 - acc: 0.9056 - val_loss: 0.2457 - val_acc: 0.9087\n",
      "Epoch 16/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2499 - acc: 0.9069 - val_loss: 0.2561 - val_acc: 0.9101\n",
      "Epoch 17/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2462 - acc: 0.9079 - val_loss: 0.2527 - val_acc: 0.9110\n",
      "Epoch 18/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2418 - acc: 0.9090 - val_loss: 0.2496 - val_acc: 0.9107\n",
      "Epoch 19/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2395 - acc: 0.9100 - val_loss: 0.2457 - val_acc: 0.9135\n",
      "Epoch 20/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2366 - acc: 0.9113 - val_loss: 0.2373 - val_acc: 0.9145\n",
      "Epoch 21/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2312 - acc: 0.9135 - val_loss: 0.2458 - val_acc: 0.9142\n",
      "Epoch 22/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2288 - acc: 0.9147 - val_loss: 0.2472 - val_acc: 0.9127\n",
      "Epoch 23/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2251 - acc: 0.9153 - val_loss: 0.2444 - val_acc: 0.9128\n",
      "Epoch 24/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2241 - acc: 0.9151 - val_loss: 0.2430 - val_acc: 0.9142\n",
      "Epoch 25/50\n",
      "48000/48000 [==============================] - 28s - loss: 0.2214 - acc: 0.9162 - val_loss: 0.2499 - val_acc: 0.9147\n",
      "Epoch 26/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2180 - acc: 0.9176 - val_loss: 0.2544 - val_acc: 0.9116\n",
      "Epoch 27/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2197 - acc: 0.9168 - val_loss: 0.2489 - val_acc: 0.9144\n",
      "Epoch 28/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2162 - acc: 0.9182 - val_loss: 0.2496 - val_acc: 0.9132\n",
      "Epoch 29/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2127 - acc: 0.9198 - val_loss: 0.2380 - val_acc: 0.9170\n",
      "Epoch 30/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2109 - acc: 0.9201 - val_loss: 0.2439 - val_acc: 0.9142\n",
      "Epoch 31/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2108 - acc: 0.9198 - val_loss: 0.2512 - val_acc: 0.9115\n",
      "Epoch 32/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2067 - acc: 0.9217 - val_loss: 0.2543 - val_acc: 0.9113\n",
      "Epoch 33/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2098 - acc: 0.9198 - val_loss: 0.2448 - val_acc: 0.9165\n",
      "Epoch 34/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2058 - acc: 0.9218 - val_loss: 0.2452 - val_acc: 0.9180\n",
      "Epoch 35/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2060 - acc: 0.9231 - val_loss: 0.2472 - val_acc: 0.9152\n",
      "Epoch 36/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2017 - acc: 0.9226 - val_loss: 0.2447 - val_acc: 0.9175\n",
      "Epoch 37/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.2013 - acc: 0.9238 - val_loss: 0.2505 - val_acc: 0.9146\n",
      "Epoch 38/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.1987 - acc: 0.9241 - val_loss: 0.2512 - val_acc: 0.9153\n",
      "Epoch 39/50\n",
      "48000/48000 [==============================] - 30s - loss: 0.1997 - acc: 0.9232 - val_loss: 0.2613 - val_acc: 0.9127\n",
      "Epoch 40/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.1941 - acc: 0.9260 - val_loss: 0.2508 - val_acc: 0.9129\n",
      "Epoch 41/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.1957 - acc: 0.9264 - val_loss: 0.2514 - val_acc: 0.9153\n",
      "Epoch 42/50\n",
      "48000/48000 [==============================] - 30s - loss: 0.1923 - acc: 0.9264 - val_loss: 0.2446 - val_acc: 0.9174\n",
      "Epoch 43/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.1931 - acc: 0.9275 - val_loss: 0.2439 - val_acc: 0.9177\n",
      "Epoch 44/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.1910 - acc: 0.9278 - val_loss: 0.2504 - val_acc: 0.9181\n",
      "Epoch 45/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.1912 - acc: 0.9264 - val_loss: 0.2511 - val_acc: 0.9183\n",
      "Epoch 46/50\n",
      "48000/48000 [==============================] - 29s - loss: 0.1921 - acc: 0.9273 - val_loss: 0.2613 - val_acc: 0.9133\n",
      "Epoch 47/50\n",
      "48000/48000 [==============================] - 30s - loss: 0.1897 - acc: 0.9284 - val_loss: 0.2530 - val_acc: 0.9130\n",
      "Epoch 48/50\n",
      "48000/48000 [==============================] - 31s - loss: 0.1872 - acc: 0.9290 - val_loss: 0.2601 - val_acc: 0.9180\n",
      "Epoch 49/50\n",
      "48000/48000 [==============================] - 30s - loss: 0.1854 - acc: 0.9296 - val_loss: 0.2657 - val_acc: 0.9132\n",
      "Epoch 50/50\n",
      "48000/48000 [==============================] - 32s - loss: 0.1837 - acc: 0.9278 - val_loss: 0.2664 - val_acc: 0.9147\n",
      "1465.7929389476776\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "start_time = time.time()\n",
    "\n",
    "training_history =model.fit(X_train,y_train,batch_size= batch_size, epochs = nb_ephochs, verbose=1, validation_data=(X_val, y_val))\n",
    "\n",
    "training_time = time.time() - start_time\n",
    "print(training_time)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'keras.callbacks.History'>\n"
     ]
    }
   ],
   "source": [
    "print(type(training_history))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1465.7929389476776"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['val_loss', 'val_acc', 'loss', 'acc'])"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_history.history.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "training_accuracy   = training_history.history['acc']\n",
    "validation_accuracy = training_history.history['val_acc']\n",
    "\n",
    "training_loss       = training_history.history['loss']\n",
    "validation_loss     = training_history.history['val_loss']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x3fbf7074e0>"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJkAAAF3CAYAAAAGvJp+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nERFJ2v8Ht7LB4rmuOzoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x3fc07bf668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(nrows=1, ncols=2,figsize=(20,6))\n",
    "\n",
    "axes[0].plot(training_accuracy,'b')\n",
    "axes[0].plot(validation_accuracy,'r')\n",
    "axes[0].legend(['training_acc', 'validation_acc'])\n",
    "axes[0].set_xlabel('No of Epoch')\n",
    "axes[0].set_ylabel('Accuracy')\n",
    "\n",
    "axes[1].plot(training_loss,'b')\n",
    "axes[1].plot(validation_loss,'r')\n",
    "axes[1].legend(['training_loss', 'validation_loss'])\n",
    "axes[1].set_xlabel('No of Epoch')\n",
    "axes[1].set_ylabel('Loss')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "24.0 24.429882315794625\n"
     ]
    }
   ],
   "source": [
    "print(training_time // 60,training_time/60)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 9888/10000 [============================>.] - ETA: 0s"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(X_test, y_test, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.24394725968837738, 0.92000000000000004]"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test loss: 0.24394725968837738  \n",
      " Test Accuracy: 92.0 %\n"
     ]
    }
   ],
   "source": [
    "print('Test loss: {}  \\n Test Accuracy: {} %'.format(score[0],score[1]*100))\n"
   ]
  }
 ],
 "metadata": {
  "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": "3.6.1"
  },
  "nav_menu": {},
  "toc": {
   "navigate_menu": true,
   "number_sections": true,
   "sideBar": true,
   "threshold": 6,
   "toc_cell": false,
   "toc_section_display": "block",
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}