izderadicka/tf.ipynb

## tf.ipynb
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import tensorflow as tf\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting MNIST_data/train-images-idx3-ubyte.gz\n",
      "Extracting MNIST_data/train-labels-idx1-ubyte.gz\n",
      "Extracting MNIST_data/t10k-images-idx3-ubyte.gz\n",
      "Extracting MNIST_data/t10k-labels-idx1-ubyte.gz\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.examples.tutorials.mnist import input_data\n",
    "mnist = input_data.read_data_sets('MNIST_data', one_hot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sess = tf.InteractiveSession()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "x=tf.placeholder(tf.float32, [None, 784])\n",
    "y_=tf.placeholder(tf.float32, [None, 10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "W=tf.Variable(tf.zeros([784,10]))\n",
    "b=tf.Variable(tf.zeros(10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "sess.run(tf.initialize_all_variables())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "y = tf.nn.softmax(tf.matmul(x, W) + b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "cross_enthropy =  tf.reduce_mean(- tf.reduce_sum(y_ * tf.log(y), reduction_indices=[1]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_enthropy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for i in range(1000):\n",
    "    batch = mnist.train.next_batch(50)\n",
    "    sess.run(train_step, feed_dict={x:batch[0], y_:batch[1]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "correct=tf.equal(tf.argmax(y_,1), tf.argmax(y,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "accuracy=tf.reduce_mean(tf.cast(correct, tf.float32))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.90920001"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy.eval(feed_dict={x: mnist.test.images, y_:mnist.test.labels})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def weighted_variable(shape):\n",
    "    init = tf.truncated_normal(shape, stddev=0.1)\n",
    "    return tf.Variable(init)\n",
    "\n",
    "def bias_variable(shape):\n",
    "    init =  tf.constant(0.1, shape=shape)\n",
    "    return tf.Variable(init)\n",
    "\n",
    "def conv2d(x, W):\n",
    "    return tf.nn.conv2d(x, W, strides=[1,1,1,1], padding='SAME')\n",
    "\n",
    "def max_pool_2x2(x):\n",
    "    return tf.nn.max_pool(x, ksize=[1,2,2,1], strides=[1,2,2,1], padding='SAME')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "FACT = 1\n",
    "L1_SIZE = int(32 * FACT)\n",
    "L2_SIZE = int(64 * FACT)\n",
    "L3_SIZE = int(1024 * FACT)\n",
    "CONV_SIZE = 4 #5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# First convolutional layer\n",
    "W_conv1 = weighted_variable([CONV_SIZE,CONV_SIZE,1,L1_SIZE])\n",
    "b_conv1 = bias_variable([L1_SIZE])\n",
    "x_image = tf.reshape(x, [-1, 28, 28, 1])\n",
    "\n",
    "h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) +b_conv1)\n",
    "h_pool1 =max_pool_2x2(h_conv1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Second convolutional layer\n",
    "W_conv2 = weighted_variable([CONV_SIZE,CONV_SIZE,L1_SIZE,L2_SIZE])\n",
    "b_conv2 = bias_variable([L2_SIZE])\n",
    "\n",
    "h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)\n",
    "h_pool2 = max_pool_2x2(h_conv2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Third fully connected layer\n",
    "W_fc1 = weighted_variable([7*7*L2_SIZE, L3_SIZE])\n",
    "b_fc1 = bias_variable([L3_SIZE])\n",
    "h_pool2_flat = tf.reshape(h_pool2, [-1, 7*7*L2_SIZE])\n",
    "h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# dropout layer\n",
    "keep_prob = tf.placeholder(tf.float32)\n",
    "h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Fourth (readout) softmax layer\n",
    "W_fc2 = weighted_variable([L3_SIZE, 10])\n",
    "b_fc2 = bias_variable([10])\n",
    "\n",
    "y_conv = tf.nn.softmax(tf.matmul(h_fc1_drop, W_fc2) + b_fc2)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "step 0 - accuracy 0.100000\n",
      "step 100 - accuracy 0.620000\n",
      "step 200 - accuracy 0.900000\n",
      "step 300 - accuracy 0.940000\n",
      "step 400 - accuracy 0.900000\n",
      "step 500 - accuracy 0.940000\n",
      "step 600 - accuracy 0.940000\n",
      "step 700 - accuracy 0.840000\n",
      "step 800 - accuracy 0.860000\n",
      "step 900 - accuracy 0.860000\n",
      "step 1000 - accuracy 0.980000\n",
      "step 1100 - accuracy 0.940000\n",
      "step 1200 - accuracy 0.980000\n",
      "step 1300 - accuracy 0.960000\n",
      "step 1400 - accuracy 0.940000\n",
      "step 1500 - accuracy 0.940000\n",
      "step 1600 - accuracy 0.960000\n",
      "step 1700 - accuracy 0.920000\n",
      "step 1800 - accuracy 0.940000\n",
      "step 1900 - accuracy 0.900000\n",
      "step 2000 - accuracy 0.960000\n",
      "step 2100 - accuracy 0.940000\n",
      "step 2200 - accuracy 0.980000\n",
      "step 2300 - accuracy 0.940000\n",
      "step 2400 - accuracy 0.980000\n",
      "step 2500 - accuracy 0.980000\n",
      "step 2600 - accuracy 1.000000\n",
      "step 2700 - accuracy 0.980000\n",
      "step 2800 - accuracy 1.000000\n",
      "step 2900 - accuracy 1.000000\n",
      "step 3000 - accuracy 0.940000\n",
      "step 3100 - accuracy 0.980000\n",
      "step 3200 - accuracy 1.000000\n",
      "step 3300 - accuracy 0.960000\n",
      "step 3400 - accuracy 0.960000\n",
      "step 3500 - accuracy 1.000000\n",
      "step 3600 - accuracy 1.000000\n",
      "step 3700 - accuracy 0.980000\n",
      "step 3800 - accuracy 0.960000\n",
      "step 3900 - accuracy 1.000000\n",
      "step 4000 - accuracy 0.940000\n",
      "step 4100 - accuracy 0.960000\n",
      "step 4200 - accuracy 0.980000\n",
      "step 4300 - accuracy 0.960000\n",
      "step 4400 - accuracy 0.940000\n",
      "step 4500 - accuracy 0.980000\n",
      "step 4600 - accuracy 0.960000\n",
      "step 4700 - accuracy 1.000000\n",
      "step 4800 - accuracy 1.000000\n",
      "step 4900 - accuracy 0.980000\n",
      "step 5000 - accuracy 1.000000\n",
      "step 5100 - accuracy 0.980000\n",
      "step 5200 - accuracy 0.980000\n",
      "step 5300 - accuracy 1.000000\n",
      "step 5400 - accuracy 0.980000\n",
      "step 5500 - accuracy 0.980000\n",
      "step 5600 - accuracy 1.000000\n",
      "step 5700 - accuracy 0.960000\n",
      "step 5800 - accuracy 1.000000\n",
      "step 5900 - accuracy 0.980000\n",
      "step 6000 - accuracy 1.000000\n",
      "step 6100 - accuracy 0.980000\n",
      "step 6200 - accuracy 1.000000\n",
      "step 6300 - accuracy 1.000000\n",
      "step 6400 - accuracy 1.000000\n",
      "step 6500 - accuracy 0.980000\n",
      "step 6600 - accuracy 1.000000\n",
      "step 6700 - accuracy 1.000000\n",
      "step 6800 - accuracy 0.980000\n",
      "step 6900 - accuracy 0.960000\n",
      "step 7000 - accuracy 1.000000\n",
      "step 7100 - accuracy 1.000000\n",
      "step 7200 - accuracy 0.980000\n",
      "step 7300 - accuracy 0.980000\n",
      "step 7400 - accuracy 0.980000\n",
      "step 7500 - accuracy 1.000000\n",
      "step 7600 - accuracy 1.000000\n",
      "step 7700 - accuracy 0.980000\n",
      "step 7800 - accuracy 1.000000\n",
      "step 7900 - accuracy 1.000000\n",
      "step 8000 - accuracy 0.980000\n",
      "step 8100 - accuracy 0.980000\n",
      "step 8200 - accuracy 0.960000\n",
      "step 8300 - accuracy 0.980000\n",
      "step 8400 - accuracy 0.940000\n",
      "step 8500 - accuracy 1.000000\n",
      "step 8600 - accuracy 0.980000\n",
      "step 8700 - accuracy 1.000000\n",
      "step 8800 - accuracy 0.980000\n",
      "step 8900 - accuracy 1.000000\n",
      "step 9000 - accuracy 1.000000\n",
      "step 9100 - accuracy 0.980000\n",
      "step 9200 - accuracy 0.960000\n",
      "step 9300 - accuracy 1.000000\n",
      "step 9400 - accuracy 0.960000\n",
      "step 9500 - accuracy 0.960000\n",
      "step 9600 - accuracy 1.000000\n",
      "step 9700 - accuracy 1.000000\n",
      "step 9800 - accuracy 1.000000\n",
      "step 9900 - accuracy 1.000000\n",
      "step 10000 - accuracy 1.000000\n",
      "step 10100 - accuracy 0.980000\n",
      "step 10200 - accuracy 0.960000\n",
      "step 10300 - accuracy 1.000000\n",
      "step 10400 - accuracy 1.000000\n",
      "step 10500 - accuracy 0.980000\n",
      "step 10600 - accuracy 0.980000\n",
      "step 10700 - accuracy 0.980000\n",
      "step 10800 - accuracy 0.960000\n",
      "step 10900 - accuracy 1.000000\n",
      "step 11000 - accuracy 1.000000\n",
      "step 11100 - accuracy 1.000000\n",
      "step 11200 - accuracy 1.000000\n",
      "step 11300 - accuracy 1.000000\n",
      "step 11400 - accuracy 0.980000\n",
      "step 11500 - accuracy 1.000000\n",
      "step 11600 - accuracy 0.980000\n",
      "step 11700 - accuracy 0.980000\n",
      "step 11800 - accuracy 0.980000\n",
      "step 11900 - accuracy 1.000000\n",
      "step 12000 - accuracy 1.000000\n",
      "step 12100 - accuracy 1.000000\n",
      "step 12200 - accuracy 1.000000\n",
      "step 12300 - accuracy 1.000000\n",
      "step 12400 - accuracy 0.940000\n",
      "step 12500 - accuracy 1.000000\n",
      "step 12600 - accuracy 1.000000\n",
      "step 12700 - accuracy 1.000000\n",
      "step 12800 - accuracy 1.000000\n",
      "step 12900 - accuracy 1.000000\n",
      "step 13000 - accuracy 1.000000\n",
      "step 13100 - accuracy 1.000000\n",
      "step 13200 - accuracy 1.000000\n",
      "step 13300 - accuracy 1.000000\n",
      "step 13400 - accuracy 1.000000\n",
      "step 13500 - accuracy 0.980000\n",
      "step 13600 - accuracy 0.980000\n",
      "step 13700 - accuracy 1.000000\n",
      "step 13800 - accuracy 1.000000\n",
      "step 13900 - accuracy 0.980000\n",
      "step 14000 - accuracy 1.000000\n",
      "step 14100 - accuracy 1.000000\n",
      "step 14200 - accuracy 1.000000\n",
      "step 14300 - accuracy 1.000000\n",
      "step 14400 - accuracy 1.000000\n",
      "step 14500 - accuracy 1.000000\n",
      "step 14600 - accuracy 1.000000\n",
      "step 14700 - accuracy 1.000000\n",
      "step 14800 - accuracy 0.980000\n",
      "step 14900 - accuracy 1.000000\n",
      "step 15000 - accuracy 0.960000\n",
      "step 15100 - accuracy 1.000000\n",
      "step 15200 - accuracy 1.000000\n",
      "step 15300 - accuracy 0.980000\n",
      "step 15400 - accuracy 0.980000\n",
      "step 15500 - accuracy 1.000000\n",
      "step 15600 - accuracy 1.000000\n",
      "step 15700 - accuracy 1.000000\n",
      "step 15800 - accuracy 1.000000\n",
      "step 15900 - accuracy 1.000000\n",
      "step 16000 - accuracy 1.000000\n",
      "step 16100 - accuracy 0.980000\n",
      "step 16200 - accuracy 1.000000\n",
      "step 16300 - accuracy 1.000000\n",
      "step 16400 - accuracy 1.000000\n",
      "step 16500 - accuracy 1.000000\n",
      "step 16600 - accuracy 1.000000\n",
      "step 16700 - accuracy 1.000000\n",
      "step 16800 - accuracy 0.980000\n",
      "step 16900 - accuracy 1.000000\n",
      "step 17000 - accuracy 1.000000\n",
      "step 17100 - accuracy 1.000000\n",
      "step 17200 - accuracy 1.000000\n",
      "step 17300 - accuracy 0.980000\n",
      "step 17400 - accuracy 1.000000\n",
      "step 17500 - accuracy 1.000000\n",
      "step 17600 - accuracy 0.980000\n",
      "step 17700 - accuracy 1.000000\n",
      "step 17800 - accuracy 1.000000\n",
      "step 17900 - accuracy 1.000000\n",
      "step 18000 - accuracy 1.000000\n",
      "step 18100 - accuracy 1.000000\n",
      "step 18200 - accuracy 1.000000\n",
      "step 18300 - accuracy 1.000000\n",
      "step 18400 - accuracy 1.000000\n",
      "step 18500 - accuracy 1.000000\n",
      "step 18600 - accuracy 1.000000\n",
      "step 18700 - accuracy 1.000000\n",
      "step 18800 - accuracy 1.000000\n",
      "step 18900 - accuracy 1.000000\n",
      "step 19000 - accuracy 1.000000\n",
      "step 19100 - accuracy 1.000000\n",
      "step 19200 - accuracy 1.000000\n",
      "step 19300 - accuracy 0.980000\n",
      "step 19400 - accuracy 1.000000\n",
      "step 19500 - accuracy 1.000000\n",
      "step 19600 - accuracy 1.000000\n",
      "step 19700 - accuracy 1.000000\n",
      "step 19800 - accuracy 1.000000\n",
      "step 19900 - accuracy 1.000000\n",
      "Test accuracy 0.992500 \n"
     ]
    }
   ],
   "source": [
    "cross_entropy = tf.reduce_mean( - tf.reduce_sum( y_ * tf.log(y_conv), reduction_indices=[1]))\n",
    "train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)\n",
    "correct_prediction = tf.equal( tf.argmax(y_, 1), tf.argmax(y_conv, 1))\n",
    "accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))\n",
    "sess.run(tf.initialize_all_variables())\n",
    "results_accuracy = np.zeros(200)\n",
    "for i in range(20000):\n",
    "    batch = mnist.train.next_batch(50)\n",
    "        \n",
    "    if i % 100 == 0:\n",
    "        train_accuracy = accuracy.eval(feed_dict={x:batch[0], y_: batch[1], keep_prob: 0.5})\n",
    "        print \"step %d - accuracy %f\" %(i,train_accuracy)\n",
    "        results_accuracy[i / 100] = train_accuracy\n",
    "\n",
    "    train_step.run(feed_dict={x:batch[0], y_: batch[1], keep_prob: 0.5})\n",
    "        \n",
    "        \n",
    "print \"Test accuracy %f \" % accuracy.eval( feed_dict={x: mnist.test.images, \n",
    "                                                      y_:mnist.test.labels, keep_prob: 1.0})\n",
    "            \n",
    "        \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "missed = np.logical_not(correct_prediction.eval(feed_dict={x: mnist.test.images, \n",
    "                                                      y_:mnist.test.labels, keep_prob: 1.0}))\n",
    "guessed = sess.run(tf.argmax(y_conv, 1), feed_dict={x: mnist.test.images, \n",
    "                                                      y_:mnist.test.labels, keep_prob: 1.0})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99.25\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>missed</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>digit</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>15.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>11.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       missed\n",
       "digit        \n",
       "4        15.0\n",
       "6        12.0\n",
       "9        11.0\n",
       "8         8.0\n",
       "5         7.0\n",
       "7         7.0\n",
       "0         5.0\n",
       "2         4.0\n",
       "3         4.0\n",
       "1         2.0"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print 100*(1- (missed.sum() / float(missed.size)))\n",
    "df = pd.DataFrame({'digit': np.argmax(mnist.test.labels, 1), 'missed': missed, 'guessed':guessed})\n",
    "df.groupby('digit').sum().sort_values('missed', ascending=False)[['missed']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def display_image(idx):\n",
    "    plt.imshow(mnist.test.images[idx].reshape([28,28]), cmap='Greys')\n",
    "    plt.title('Is %d but guessed as %d'%(df.digit[idx], df.guessed[idx]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ivan/.virtualenvs/tensorflow/lib/python2.7/site-packages/ipykernel/__main__.py:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  if __name__ == '__main__':\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>digit</th>\n",
       "      <th>guessed</th>\n",
       "      <th>missed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1232</th>\n",
       "      <td>9</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1247</th>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1709</th>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1901</th>\n",
       "      <td>9</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2939</th>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3060</th>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3503</th>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4224</th>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4761</th>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6571</th>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9692</th>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      digit  guessed missed\n",
       "1232      9        4   True\n",
       "1247      9        5   True\n",
       "1709      9        5   True\n",
       "1901      9        4   True\n",
       "2939      9        5   True\n",
       "3060      9        7   True\n",
       "3503      9        1   True\n",
       "4224      9        7   True\n",
       "4761      9        8   True\n",
       "6571      9        7   True\n",
       "9692      9        7   True"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.missed][df.digit == 9]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def display_missed(digit):\n",
    "    missed=df[df.missed][df.digit == digit]\n",
    "    length = missed.shape[0]\n",
    "    rows= np.ceil(length / 4.0 )\n",
    "    zoom=3\n",
    "    plt.figure(figsize=(zoom*4,zoom*rows))\n",
    "    for count,i in enumerate(missed.index):\n",
    "        plt.subplot(rows, 4, count+1)\n",
    "        display_image(i)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ivan/.virtualenvs/tensorflow/lib/python2.7/site-packages/ipykernel/__main__.py:2: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  from ipykernel import kernelapp as app\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArwAAAIfCAYAAACSMg8DAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nERGpNBW8IiIiIlJpKnhFREREpNL+D34euu+7YxAYAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7ac125dd50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display_missed(9)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#W_conv1.eval()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}