lab-10-6-mnist_nn_batchnorm.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lab: Batchnormalization Layer\n",
    "\n",
    "## What is a batchnormalization layer?\n",
    "It is a layer that normalize the output before the activation layer. [The original paper](https://arxiv.org/abs/1502.03167) was proposed by Sergey Ioffe in 2015.\n",
    "\n",
    "Batch Normalization Layer looks like this: ![bn](https://kratzert.github.io/images/bn_backpass/bn_algorithm.PNG)\n",
    "\n",
    "## Why batchnormalization?\n",
    "The distribution of each layer's input changes because the weights of the previous layer change as we update weights by the gradient descent. This is called a covariance shift, which makes the network training difficult.\n",
    "\n",
    "For example, if the activation layer is a relu layer and the input of the activation layer is shifted to less than zeros, no weights will be activated!\n",
    "\n",
    "One thing also worth mentioning is that $\\gamma$ and $\\beta$ parameters in $$ y = \\gamma \\hat{x} + \\beta $$ are also trainable. \n",
    "\n",
    "**What it means is that if we don't need the batchnormalization, its parameters will be updated such that it offsets the normalization step.**\n",
    "\n",
    "For example, assume that\n",
    "\n",
    "\\begin{align}\n",
    "\\gamma &= \\sqrt{\\sigma^2_B + \\epsilon}\\\\\n",
    "\\beta &= \\mu_B\n",
    "\\end{align}\n",
    "\n",
    "then\n",
    "\n",
    "$$ y_i = \\gamma \\hat{x_i} + \\beta = x_i $$\n",
    "\n",
    "Also note that $\\mu$ and $\\sigma$ are computed using moving averages during the training step. However, during the test time, the computed $\\mu$ and $\\sigma$ will be used as fixed\n",
    "\n",
    "## Conclusion\n",
    "- Always use the batch normalization!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Enough Talk: how to implement in Tensorflow"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Load Library\n",
    "- We use the famous MNIST data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "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": [
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "from tensorflow.examples.tutorials.mnist import input_data\n",
    "%matplotlib inline\n",
    "\n",
    "mnist = input_data.read_data_sets(\"MNIST_data/\", one_hot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(55000, 784)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mnist.train.images.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Define Model & Solver Class\n",
    "- Object-Oriented-Programming allows to define multiple model easily\n",
    "- Why do we separate model and solver classes?\n",
    "  - We can just swap out the model class in the Solver class when we need a different network architecture\n",
    "  - Usually we need one solver class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "class Model:\n",
    "    \"\"\"Network Model Class\n",
    "    \n",
    "    Note that this class has only the constructor.\n",
    "    The actual model is defined inside the constructor.\n",
    "    \n",
    "    Attributes\n",
    "    ----------\n",
    "    X : tf.float32\n",
    "        This is a tensorflow placeholder for MNIST images\n",
    "        Expected shape is [None, 784]\n",
    "        \n",
    "    y : tf.float32\n",
    "        This is a tensorflow placeholder for MNIST labels (one hot encoded)\n",
    "        Expected shape is [None, 10]\n",
    "        \n",
    "    mode : tf.bool\n",
    "        This is used for the batch normalization\n",
    "        It's `True` at training time and `False` at test time\n",
    "        \n",
    "    loss : tf.float32\n",
    "        The loss function is a softmax cross entropy\n",
    "        \n",
    "    train_op\n",
    "        This is simply the training op that minimizes the loss\n",
    "        \n",
    "    accuracy : tf.float32\n",
    "        The accuracy operation\n",
    "        \n",
    "    \n",
    "    Examples\n",
    "    ----------\n",
    "    >>> model = Model(\"Batch Norm\", 32, 10)\n",
    "\n",
    "    \"\"\"\n",
    "    def __init__(self, name, input_dim, output_dim, hidden_dims=[32, 32], \n",
    "                 use_batchnorm=True, activation_fn=tf.nn.relu, \n",
    "                 optimizer=tf.train.AdamOptimizer, lr=0.01):\n",
    "        \"\"\" Constructor\n",
    "        \n",
    "        Parameters\n",
    "        --------\n",
    "        name : str\n",
    "            The name of this network\n",
    "            The entire network will be created under `tf.variable_scope(name)`\n",
    "            \n",
    "        input_dim : int\n",
    "            The input dimension\n",
    "            In this example, 784\n",
    "        \n",
    "        output_dim : int\n",
    "            The number of output labels\n",
    "            There are 10 labels\n",
    "            \n",
    "        hidden_dims : list (default: [32, 32])\n",
    "            len(hidden_dims) = number of layers\n",
    "            each element is the number of hidden units\n",
    "            \n",
    "        use_batchnorm : bool (default: True)\n",
    "            If true, it will create the batchnormalization layer\n",
    "            \n",
    "        activation_fn : TF functions (default: tf.nn.relu)\n",
    "            Activation Function\n",
    "            \n",
    "        optimizer : TF optimizer (default: tf.train.AdamOptimizer)\n",
    "            Optimizer Function\n",
    "            \n",
    "        lr : float (default: 0.01)\n",
    "            Learning rate\n",
    "        \n",
    "        \"\"\"\n",
    "        with tf.variable_scope(name):\n",
    "            # Placeholders are defined\n",
    "            self.X = tf.placeholder(tf.float32, [None, input_dim], name='X')\n",
    "            self.y = tf.placeholder(tf.float32, [None, output_dim], name='y')\n",
    "            self.mode = tf.placeholder(tf.bool, name='train_mode')            \n",
    "            \n",
    "            # Loop over hidden layers\n",
    "            net = self.X\n",
    "            for i, h_dim in enumerate(hidden_dims):\n",
    "                with tf.variable_scope('layer{}'.format(i)):\n",
    "                    net = tf.layers.dense(net, h_dim)\n",
    "                    \n",
    "                    if use_batchnorm:\n",
    "                        net = tf.layers.batch_normalization(net, training=self.mode)\n",
    "                        \n",
    "                    net = activation_fn(net)\n",
    "            \n",
    "            # Attach fully connected layers\n",
    "            net = tf.contrib.layers.flatten(net)\n",
    "            net = tf.layers.dense(net, output_dim)\n",
    "            \n",
    "            self.loss = tf.nn.softmax_cross_entropy_with_logits(logits=net, labels=self.y)\n",
    "            self.loss = tf.reduce_mean(self.loss, name='loss')    \n",
    "            \n",
    "            # When using the batchnormalization layers,\n",
    "            # it is necessary to manually add the update operations\n",
    "            # because the moving averages are not included in the graph            \n",
    "            update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS, scope=name)\n",
    "            with tf.control_dependencies(update_ops):                     \n",
    "                self.train_op = optimizer(lr).minimize(self.loss)\n",
    "            \n",
    "            # Accuracy etc \n",
    "            softmax = tf.nn.softmax(net, name='softmax')\n",
    "            self.accuracy = tf.equal(tf.argmax(softmax, 1), tf.argmax(self.y, 1))\n",
    "            self.accuracy = tf.reduce_mean(tf.cast(self.accuracy, tf.float32))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "class Solver:\n",
    "    \"\"\"Solver class\n",
    "    \n",
    "    This class will contain the model class and session\n",
    "    \n",
    "    Attributes\n",
    "    ----------\n",
    "    model : Model class\n",
    "    sess : TF session\n",
    "        \n",
    "    Methods\n",
    "    ----------\n",
    "    train(X, y)\n",
    "        Run the train_op and Returns the loss\n",
    "        \n",
    "    evalulate(X, y, batch_size=None)\n",
    "        Returns \"Loss\" and \"Accuracy\"\n",
    "        If batch_size is given, it's computed using batch_size\n",
    "        because most GPU memories cannot handle the entire training data at once\n",
    "            \n",
    "    Example\n",
    "    ----------\n",
    "    >>> sess = tf.InteractiveSession()\n",
    "    >>> model = Model(\"BatchNorm\", 32, 10)\n",
    "    >>> solver = Solver(sess, model)\n",
    "    \n",
    "    # Train\n",
    "    >>> solver.train(X, y)\n",
    "    \n",
    "    # Evaluate\n",
    "    >>> solver.evaluate(X, y)\n",
    "    \"\"\"\n",
    "    def __init__(self, sess, model):\n",
    "        self.model = model\n",
    "        self.sess = sess\n",
    "        \n",
    "    def train(self, X, y):\n",
    "        feed = {\n",
    "            self.model.X: X,\n",
    "            self.model.y: y,\n",
    "            self.model.mode: True\n",
    "        }\n",
    "        train_op = self.model.train_op\n",
    "        loss = self.model.loss\n",
    "        \n",
    "        return self.sess.run([train_op, loss], feed_dict=feed)\n",
    "    \n",
    "    def evaluate(self, X, y, batch_size=None):\n",
    "        if batch_size:\n",
    "            N = X.shape[0]\n",
    "            \n",
    "            total_loss = 0\n",
    "            total_acc = 0\n",
    "            \n",
    "            for i in range(0, N, batch_size):\n",
    "                X_batch = X[i:i + batch_size]\n",
    "                y_batch = y[i:i + batch_size]\n",
    "                \n",
    "                feed = {\n",
    "                    self.model.X: X_batch,\n",
    "                    self.model.y: y_batch,\n",
    "                    self.model.mode: False\n",
    "                }\n",
    "                \n",
    "                loss = self.model.loss\n",
    "                accuracy = self.model.accuracy\n",
    "                \n",
    "                step_loss, step_acc = self.sess.run([loss, accuracy], feed_dict=feed)\n",
    "                \n",
    "                total_loss += step_loss * X_batch.shape[0]\n",
    "                total_acc += step_acc * X_batch.shape[0]\n",
    "            \n",
    "            total_loss /= N\n",
    "            total_acc /= N\n",
    "            \n",
    "            return total_loss, total_acc\n",
    "            \n",
    "            \n",
    "        else:\n",
    "            feed = {\n",
    "                self.model.X: X,\n",
    "                self.model.y: y,\n",
    "                self.model.mode: False\n",
    "            }\n",
    "            \n",
    "            loss = self.model.loss            \n",
    "            accuracy = self.model.accuracy\n",
    "\n",
    "            return self.sess.run([loss, accuracy], feed_dict=feed)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3. Instantiate Model/Solver classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "input_dim = 784\n",
    "output_dim = 10\n",
    "N = 55000\n",
    "\n",
    "tf.reset_default_graph()\n",
    "sess = tf.InteractiveSession()\n",
    "\n",
    "# We create two models: one with the batch norm and other without\n",
    "bn = Model('batchnorm', input_dim, output_dim, use_batchnorm=True)\n",
    "nn = Model('no_norm', input_dim, output_dim, use_batchnorm=False)\n",
    "\n",
    "# We create two solvers: to train both models at the same time for comparison\n",
    "# Usually we only need one solver class\n",
    "bn_solver = Solver(sess, bn)\n",
    "nn_solver = Solver(sess, nn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "epoch_n = 10\n",
    "batch_size = 32\n",
    "\n",
    "# Save Losses and Accuracies every epoch\n",
    "# We are going to plot them later\n",
    "train_losses = []\n",
    "train_accs = []\n",
    "\n",
    "valid_losses = []\n",
    "valid_accs = []"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4. Run the train step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Epoch 0-TRAIN] Batchnorm Loss(Acc): 0.17608(94.53%) vs No Batchnorm Loss(Acc): 0.22892(93.45%)\n",
      "[Epoch 0-VALID] Batchnorm Loss(Acc): 0.18063(94.56%) vs No Batchnorm Loss(Acc): 0.24752(93.32%)\n",
      "\n",
      "[Epoch 1-TRAIN] Batchnorm Loss(Acc): 0.11353(96.43%) vs No Batchnorm Loss(Acc): 0.17606(94.92%)\n",
      "[Epoch 1-VALID] Batchnorm Loss(Acc): 0.13095(96.02%) vs No Batchnorm Loss(Acc): 0.19394(94.64%)\n",
      "\n",
      "[Epoch 2-TRAIN] Batchnorm Loss(Acc): 0.08712(97.26%) vs No Batchnorm Loss(Acc): 0.16224(95.49%)\n",
      "[Epoch 2-VALID] Batchnorm Loss(Acc): 0.11474(96.72%) vs No Batchnorm Loss(Acc): 0.19018(95.46%)\n",
      "\n",
      "[Epoch 3-TRAIN] Batchnorm Loss(Acc): 0.08645(97.35%) vs No Batchnorm Loss(Acc): 0.14193(96.08%)\n",
      "[Epoch 3-VALID] Batchnorm Loss(Acc): 0.11125(96.76%) vs No Batchnorm Loss(Acc): 0.16944(95.68%)\n",
      "\n",
      "[Epoch 4-TRAIN] Batchnorm Loss(Acc): 0.07827(97.50%) vs No Batchnorm Loss(Acc): 0.14416(95.95%)\n",
      "[Epoch 4-VALID] Batchnorm Loss(Acc): 0.11577(96.78%) vs No Batchnorm Loss(Acc): 0.18173(95.34%)\n",
      "\n",
      "[Epoch 5-TRAIN] Batchnorm Loss(Acc): 0.07221(97.71%) vs No Batchnorm Loss(Acc): 0.13727(96.31%)\n",
      "[Epoch 5-VALID] Batchnorm Loss(Acc): 0.11270(96.62%) vs No Batchnorm Loss(Acc): 0.17928(95.88%)\n",
      "\n",
      "[Epoch 6-TRAIN] Batchnorm Loss(Acc): 0.06104(98.06%) vs No Batchnorm Loss(Acc): 0.14972(95.91%)\n",
      "[Epoch 6-VALID] Batchnorm Loss(Acc): 0.10893(96.90%) vs No Batchnorm Loss(Acc): 0.19067(95.30%)\n",
      "\n",
      "[Epoch 7-TRAIN] Batchnorm Loss(Acc): 0.06444(97.97%) vs No Batchnorm Loss(Acc): 0.14997(96.07%)\n",
      "[Epoch 7-VALID] Batchnorm Loss(Acc): 0.11659(96.66%) vs No Batchnorm Loss(Acc): 0.21077(95.06%)\n",
      "\n",
      "[Epoch 8-TRAIN] Batchnorm Loss(Acc): 0.05585(98.16%) vs No Batchnorm Loss(Acc): 0.12542(96.63%)\n",
      "[Epoch 8-VALID] Batchnorm Loss(Acc): 0.10542(97.10%) vs No Batchnorm Loss(Acc): 0.18511(95.50%)\n",
      "\n",
      "[Epoch 9-TRAIN] Batchnorm Loss(Acc): 0.05356(98.27%) vs No Batchnorm Loss(Acc): 0.14035(96.05%)\n",
      "[Epoch 9-VALID] Batchnorm Loss(Acc): 0.09461(97.10%) vs No Batchnorm Loss(Acc): 0.20565(94.76%)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "init = tf.global_variables_initializer()\n",
    "sess.run(init)\n",
    "\n",
    "for epoch in range(epoch_n):\n",
    "    for _ in range(N//batch_size):\n",
    "        X_batch, y_batch = mnist.train.next_batch(batch_size)\n",
    "        \n",
    "        _, bn_loss = bn_solver.train(X_batch, y_batch)\n",
    "        _, nn_loss = nn_solver.train(X_batch, y_batch)       \n",
    "    \n",
    "    b_loss, b_acc = bn_solver.evaluate(mnist.train.images, mnist.train.labels, batch_size)\n",
    "    n_loss, n_acc = nn_solver.evaluate(mnist.train.images, mnist.train.labels, batch_size)\n",
    "    \n",
    "    # Save train losses/acc\n",
    "    train_losses.append([b_loss, n_loss])\n",
    "    train_accs.append([b_acc, n_acc])\n",
    "    print(f'[Epoch {epoch}-TRAIN] Batchnorm Loss(Acc): {b_loss:.5f}({b_acc:.2%}) vs No Batchnorm Loss(Acc): {n_loss:.5f}({n_acc:.2%})')\n",
    "    \n",
    "    b_loss, b_acc = bn_solver.evaluate(mnist.validation.images, mnist.validation.labels)\n",
    "    n_loss, n_acc = nn_solver.evaluate(mnist.validation.images, mnist.validation.labels)\n",
    "    \n",
    "    # Save valid losses/acc\n",
    "    valid_losses.append([b_loss, n_loss])\n",
    "    valid_accs.append([b_acc, n_acc])\n",
    "    print(f'[Epoch {epoch}-VALID] Batchnorm Loss(Acc): {b_loss:.5f}({b_acc:.2%}) vs No Batchnorm Loss(Acc): {n_loss:.5f}({n_acc:.2%})')\n",
    "    print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5. Performance Comparison\n",
    "* With the batchnormalization, the loss is lower and it's more accurate too!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.096957415, 0.9702]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bn_solver.evaluate(mnist.test.images, mnist.test.labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.20128743, 0.95359999]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nn_solver.evaluate(mnist.test.images, mnist.test.labels)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def plot_compare(loss_list: list, ylim=None, title=None) -> None:\n",
    "    \n",
    "    bn = [i[0] for i in loss_list]\n",
    "    nn = [i[1] for i in loss_list]\n",
    "    \n",
    "    plt.figure(figsize=(15, 10))\n",
    "    plt.plot(bn, label='With BN')\n",
    "    plt.plot(nn, label='Without BN')\n",
    "    if ylim:\n",
    "        plt.ylim(ylim)\n",
    "        \n",
    "    if title:\n",
    "        plt.title(title)\n",
    "    plt.legend()\n",
    "    plt.grid('on')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3oAAAJOCAYAAADhz3V3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\negAAAJNR9AAAACaj6AEAAEzmC0viDJwSa9KJAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2b95283358>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_compare(train_losses, title='Training Loss at Epoch')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA20AAAJOCAYAAAAkve/mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\naSAVmYxb+wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2b91d118d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_compare(train_accs, [0, 1.0], title=\"Training Acc at Epoch\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3QAAAJOCAYAAAD/BkXEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nFDoAAICB+het8nKq9yaRZgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2b91c60860>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_compare(valid_losses, title='Validation Loss at Epoch')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA20AAAJOCAYAAAAkve/mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAwmGgDAAAYTLQBAAAMJtoAAAAGewMoxeSCLITdCQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2b90970eb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_compare(valid_accs, [0, 1.], title='Validation Acc at Epoch')"
   ]
  },
  {
   "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": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}