TF10 Classification on MNIST Save & Restore.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from tensorflow.examples.tutorials.mnist import input_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "DATA_DIR = r'D:\\tensorflow'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Extracting D:\\tensorflow\\train-images-idx3-ubyte.gz\n",
      "Extracting D:\\tensorflow\\train-labels-idx1-ubyte.gz\n",
      "Extracting D:\\tensorflow\\t10k-images-idx3-ubyte.gz\n",
      "Extracting D:\\tensorflow\\t10k-labels-idx1-ubyte.gz\n"
     ]
    }
   ],
   "source": [
    "data = input_data.read_data_sets(DATA_DIR,one_hot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(55000, 784)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.train.images.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 784)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.test.images.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(55000, 10)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.train.labels.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "x = tf.placeholder(tf.float32, [None, 784], name='x')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "W = tf.Variable(tf.zeros([784,10]), name='W')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "y_true = tf.placeholder(tf.float32, [None, 10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = tf.matmul(x,W)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#softmax is good for multi class clasification\n",
    "loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=y_pred,labels=y_true))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "gd = tf.train.GradientDescentOptimizer(0.5).minimize(loss)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on function argmax in module tensorflow.python.ops.math_ops:\n",
      "\n",
      "argmax(input, axis=None, name=None, dimension=None, output_type=tf.int64)\n",
      "    Returns the index with the largest value across axes of a tensor. (deprecated arguments)\n",
      "    \n",
      "    SOME ARGUMENTS ARE DEPRECATED. They will be removed in a future version.\n",
      "    Instructions for updating:\n",
      "    Use the `axis` argument instead\n",
      "    \n",
      "    Note that in case of ties the identity of the return value is not guaranteed.\n",
      "    \n",
      "    Args:\n",
      "      input: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`.\n",
      "      axis: A `Tensor`. Must be one of the following types: `int32`, `int64`.\n",
      "        int32 or int64, 0 <= axis < rank(input).  Describes\n",
      "        which axis of the input Tensor to reduce across. For vectors,\n",
      "        use axis = 0.\n",
      "      output_type: An optional `tf.DType` from: `tf.int32, tf.int64`. Defaults to `tf.int64`.\n",
      "      name: A name for the operation (optional).\n",
      "    \n",
      "    Returns:\n",
      "      A `Tensor` of type `output_type`.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(tf.argmax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "correct_mask = tf.equal(tf.argmax(y_pred,1) , tf.argmax(y_true,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "accuracy = tf.reduce_mean(tf.cast(correct_mask,tf.float32))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "saver = tf.train.Saver(max_to_keep=7, keep_checkpoint_every_n_hours=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "NUM_STEPS = 1000\n",
    "BATCH_SIZE = 1000\n",
    "with tf.Session() as sess:\n",
    "    sess.run(tf.global_variables_initializer())\n",
    "    for step in range(NUM_STEPS):\n",
    "        batch_x, batch_y = data.train.next_batch(BATCH_SIZE)\n",
    "        sess.run(gd, feed_dict={x:batch_x,y_true:batch_y})\n",
    "        \n",
    "        if step % 10 == 0 :\n",
    "            saver.save(sess,\"D:/tensorflow/mnist-model\",global_step=step)\n",
    "    \n",
    "    ans = sess.run(accuracy,feed_dict={x:data.test.images, y_true:data.test.labels})\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.92049998"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}