Raw File
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# CNTK 201B: Hands On Labs Image Recognition"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This hands-on lab shows how to implement image recognition task using [convolution network][] with CNTK v2 Python API. You will start with a basic feedforward CNN architecture in order to classify Cifar dataset, then you will keep adding advanced feature to your network. Finally, you will implement a VGG net and residual net similar to the one that won ImageNet competition but smaller in size.\n",
    "\n",
    "[convolution network]:https://en.wikipedia.org/wiki/Convolutional_neural_network\n",
    "\n",
    "## Introduction\n",
    "\n",
    "In this hands-on, you will practice the following:\n",
    "\n",
    "* Understanding subset of CNTK python API needed for image classification task.\n",
    "* Write a custom convolution network to classify Cifar dataset.\n",
    "* Modifying the network structure by adding:\n",
    " * [Dropout][] layer.\n",
    " * Batchnormalization layer.\n",
    "* Implement a [VGG][] style network.\n",
    "* Introduction to Residual Nets (RESNET).\n",
    "* Implement and train [RESNET network][].\n",
    "\n",
    "[RESNET network]:https://github.com/Microsoft/CNTK/wiki/Hands-On-Labs-Image-Recognition\n",
    "[VGG]:http://www.robots.ox.ac.uk/~vgg/research/very_deep/\n",
    "[Dropout]:https://en.wikipedia.org/wiki/Dropout_(neural_networks)\n",
    "\n",
    "## Prerequisites\n",
    "\n",
    "CNTK 201A hands-on lab, in which you will download and prepare Cifar dataset is a prerequisites for this lab. This tutorial depends on CNTK v2, so before starting this lab you will need to install CNTK v2. Furthermore, all the tutorials in this lab are done in python, therefore, you will need a basic knowledge of Python.\n",
    "\n",
    "CNTK 102 lab is recommended but not a prerequisites for this tutorials. However, a basic understanding of Deep Learning is needed.\n",
    "\n",
    "## Dataset\n",
    "\n",
    "You will use Cifar 10 dataset, from https://www.cs.toronto.edu/~kriz/cifar.html, during this tutorials. The dataset  contains 50000 training images and 10000 test images, all images are 32x32x3. Each image is classified as one of 10 classes as shown below:\n",
    "\n",
    "<img src=\"https://cntk.ai/jup/201/cifar-10.png\", width=500, height=500>\n",
    "\n",
    "The above image is from: https://www.cs.toronto.edu/~kriz/cifar.html\n",
    "\n",
    "## Convolution Neural Network (CNN)\n",
    "\n",
    "Convolution Neural Network (CNN) is a feedforward network comprise of a bunch of layers in such a way that the output of one layer is fed to the next layer (There are more complex architecture that skip layers, we will discuss one of those at the end of this lab). Usually, CNN start with alternating between convolution layer and pooling layer (downsample), then end up with fully connected layer for the classification part.\n",
    "\n",
    "### Convolution layer\n",
    "\n",
    "Convolution layer consist of multiple 2D convolution kernels applied on the input image or the previous layer, each convolution kernel output a feature map.\n",
    "\n",
    "<img src=\"https://cntk.ai/jup/201/Conv2D.png\">\n",
    "\n",
    "The stack of feature maps output are the input to the next layer.\n",
    "\n",
    "<img src=\"https://cntk.ai/jup/201/Conv2DFeatures.png\">\n",
    "\n",
    "> Gradient-Based Learning Applied to Document Recognition, Proceedings of the IEEE, 86(11):2278-2324, November 1998\n",
    "> Y. LeCun, L. Bottou, Y. Bengio and P. Haffner\n",
    "\n",
    "#### In CNTK:\n",
    "\n",
    "Here the [convolution][] layer in Python:\n",
    "\n",
    "```python\n",
    "def Convolution(filter_shape,        # e.g. (3,3)\n",
    "                num_filters,         # e.g. 64\n",
    "                activation,          # relu or None...etc.\n",
    "                init,                # Random initialization\n",
    "                pad,                 # True or False\n",
    "                strides)             # strides e.g. (1,1)\n",
    "```\n",
    "\n",
    "[convolution]:https://www.cntk.ai/pythondocs/layerref.html#convolution\n",
    "\n",
    "### Pooling layer\n",
    "\n",
    "In most CNN vision architecture, each convolution layer is succeeded by a pooling layer, so they keep alternating until the fully connected layer. \n",
    "\n",
    "The purpose of the pooling layer is as follow:\n",
    "\n",
    "* Reduce the dimensionality of the previous layer, which speed up the network.\n",
    "* Provide a limited translation invariant.\n",
    "\n",
    "Here an example of max pooling with a stride of 2:\n",
    "\n",
    "<img src=\"https://cntk.ai/jup/201/MaxPooling.png\", 200,200>\n",
    "\n",
    "#### In CNTK:\n",
    "\n",
    "Here the [pooling][] layer in Python:\n",
    "\n",
    "```python\n",
    "\n",
    "# Max pooling\n",
    "def MaxPooling(filter_shape,  # e.g. (3,3)\n",
    "               strides,       # (2,2)\n",
    "               pad)           # True or False\n",
    "\n",
    "# Average pooling\n",
    "def AveragePooling(filter_shape,  # e.g. (3,3)\n",
    "                   strides,       # (2,2)\n",
    "                   pad)           # True or False\n",
    "```\n",
    "\n",
    "[pooling]:https://www.cntk.ai/pythondocs/layerref.html#maxpooling-averagepooling\n",
    "\n",
    "### Dropout layer\n",
    "\n",
    "Dropout layer takes a probability value as an input, the value is called the dropout rate. Let's say the dropu rate is 0.5, what this layer does it pick at random 50% of the nodes from the previous layer and drop them out of the nework. This behavior help regularize the network.\n",
    "\n",
    "> Dropout: A Simple Way to Prevent Neural Networks from Overfitting\n",
    "> Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, Ruslan Salakhutdinov\n",
    "\n",
    "\n",
    "#### In CNTK:\n",
    "\n",
    "Dropout layer in Python:\n",
    "\n",
    "```python\n",
    "\n",
    "# Dropout\n",
    "def Dropout(prob)    # dropout rate e.g. 0.5\n",
    "```\n",
    "\n",
    "### Batch normalization (BN)\n",
    "\n",
    "Batch normalization is a way to make the input to each layer has zero mean and unit variance. BN help the network converge faster and keep the input of each layer around zero. BN has two learnable parameters called gamma and beta, the purpose of those parameters is for the network to decide for itself if the normalized input is what is best or the raw input.\n",
    "\n",
    "> Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift\n",
    "> Sergey Ioffe, Christian Szegedy\n",
    "\n",
    "#### In CNTK:\n",
    "\n",
    "[Batch normalization][] layer in Python:\n",
    "\n",
    "```python\n",
    "\n",
    "# Batch normalization\n",
    "def BatchNormalization(map_rank)  # For image map_rank=1\n",
    "```\n",
    "\n",
    "[Batch normalization]:https://www.cntk.ai/pythondocs/layerref.html#batchnormalization-layernormalization-stabilizer\n",
    "\n",
    "## Computational Network Toolkit (CNTK)\n",
    "\n",
    "CNTK is a highly flexible computation graphs, each node take inputs as tensors and produce tensors as the result of the computation. Each node is exposed in Python API, which give you the flexibility of creating any custom graphs, you can also define your own node in Python or C++ using CPU, GPU or both.\n",
    "\n",
    "For Deep learning, you can use the low level API directly or you can use CNTK layered API. We will start with the low level API, then switch to the layered API in this lab.\n",
    "\n",
    "So let's first import the needed modules for this lab."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from __future__ import print_function\n",
    "import os\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import math\n",
    "\n",
    "from cntk.blocks import default_options\n",
    "from cntk.layers import Convolution, MaxPooling, AveragePooling, Dropout, BatchNormalization, Dense\n",
    "from cntk.models import Sequential, LayerStack\n",
    "from cntk.io import MinibatchSource, ImageDeserializer, StreamDef, StreamDefs\n",
    "from cntk.initializer import glorot_uniform, he_normal\n",
    "from cntk import Trainer\n",
    "from cntk.learner import momentum_sgd, learning_rate_schedule, UnitType, momentum_as_time_constant_schedule\n",
    "from cntk.ops import cross_entropy_with_softmax, classification_error, relu, input_variable, softmax, element_times\n",
    "from cntk.utils import *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we imported the needed modules, let's implement our first CNN, as shown below:\n",
    "\n",
    "<img src=\"https://cntk.ai/jup/201/CNN.png\">\n",
    "\n",
    "Let's implement the above network using CNTK layer API:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def create_basic_model(input, out_dims):\n",
    "    \n",
    "    net = Convolution((5,5), 32, init=glorot_uniform(), activation=relu, pad=True)(input)\n",
    "    net = MaxPooling((3,3), strides=(2,2))(net)\n",
    "\n",
    "    net = Convolution((5,5), 32, init=glorot_uniform(), activation=relu, pad=True)(net)\n",
    "    net = MaxPooling((3,3), strides=(2,2))(net)\n",
    "\n",
    "    net = Convolution((5,5), 64, init=glorot_uniform(), activation=relu, pad=True)(net)\n",
    "    net = MaxPooling((3,3), strides=(2,2))(net)\n",
    "    \n",
    "    net = Dense(64, init=glorot_uniform())(net)\n",
    "    net = Dense(out_dims, init=glorot_uniform(), activation=None)(net)\n",
    "    \n",
    "    return net"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To train the above model we need two things:\n",
    "* Read the training images and their corresponding labels.\n",
    "* Define a cost function, compute the cost for each mini-batch and update the model weights according to the cost value.\n",
    "\n",
    "To read the data in CNTK, we will use CNTK readers which handle data augmentation and can fetch data in parallel.\n",
    "\n",
    "Example of a map text file:\n",
    "\n",
    "    S:\\data\\CIFAR-10\\train\\00001.png\t9\n",
    "    S:\\data\\CIFAR-10\\train\\00002.png\t9\n",
    "    S:\\data\\CIFAR-10\\train\\00003.png\t4\n",
    "    S:\\data\\CIFAR-10\\train\\00004.png\t1\n",
    "    S:\\data\\CIFAR-10\\train\\00005.png\t1\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# model dimensions\n",
    "image_height = 32\n",
    "image_width  = 32\n",
    "num_channels = 3\n",
    "num_classes  = 10\n",
    "\n",
    "#\n",
    "# Define the reader for both training and evaluation action.\n",
    "#\n",
    "def create_reader(map_file, mean_file, train):\n",
    "    if not os.path.exists(map_file) or not os.path.exists(mean_file):\n",
    "        raise RuntimeError(\"This tutorials depends 201A tutorials, please run 201A first.\")\n",
    "\n",
    "    # transformation pipeline for the features has jitter/crop only when training\n",
    "    transforms = []\n",
    "    if train:\n",
    "        transforms += [\n",
    "            ImageDeserializer.crop(crop_type='Random', ratio=0.8, jitter_type='uniRatio') # train uses jitter\n",
    "        ]\n",
    "    transforms += [\n",
    "        ImageDeserializer.scale(width=image_width, height=image_height, channels=num_channels, interpolations='linear'),\n",
    "        ImageDeserializer.mean(mean_file)\n",
    "    ]\n",
    "    # deserializer\n",
    "    return MinibatchSource(ImageDeserializer(map_file, StreamDefs(\n",
    "        features = StreamDef(field='image', transforms=transforms), # first column in map file is referred to as 'image'\n",
    "        labels   = StreamDef(field='label', shape=num_classes)      # and second as 'label'\n",
    "    )))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let us write the the training and validation loop."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#\n",
    "# Train and evaluate the network.\n",
    "#\n",
    "def train_and_evaluate(reader_train, reader_test, max_epochs, model_func):\n",
    "    # Input variables denoting the features and label data\n",
    "    input_var = input_variable((num_channels, image_height, image_width))\n",
    "    label_var = input_variable((num_classes))\n",
    "\n",
    "    # Normalize the input\n",
    "    feature_scale = 1.0 / 256.0\n",
    "    input_var_norm = element_times(feature_scale, input_var)\n",
    "    \n",
    "    # apply model to input\n",
    "    z = model_func(input_var_norm, out_dims=10)\n",
    "\n",
    "    #\n",
    "    # Training action\n",
    "    #\n",
    "\n",
    "    # loss and metric\n",
    "    ce = cross_entropy_with_softmax(z, label_var)\n",
    "    pe = classification_error(z, label_var)\n",
    "\n",
    "    # training config\n",
    "    epoch_size     = 50000\n",
    "    minibatch_size = 64\n",
    "\n",
    "    # Set training parameters\n",
    "    lr_per_minibatch       = learning_rate_schedule([0.01]*10 + [0.003]*10 + [0.001], UnitType.minibatch, epoch_size)\n",
    "    momentum_time_constant = momentum_as_time_constant_schedule(-minibatch_size/np.log(0.9))\n",
    "    l2_reg_weight          = 0.001\n",
    "    \n",
    "    # trainer object\n",
    "    learner     = momentum_sgd(z.parameters, \n",
    "                               lr = lr_per_minibatch, momentum = momentum_time_constant, \n",
    "                               l2_regularization_weight=l2_reg_weight)\n",
    "    trainer     = Trainer(z, ce, pe, [learner])\n",
    "\n",
    "    # define mapping from reader streams to network inputs\n",
    "    input_map = {\n",
    "        input_var: reader_train.streams.features,\n",
    "        label_var: reader_train.streams.labels\n",
    "    }\n",
    "\n",
    "    log_number_of_parameters(z) ; print()\n",
    "    progress_printer = ProgressPrinter(tag='Training')\n",
    "\n",
    "    # perform model training\n",
    "    batch_index = 0\n",
    "    plot_data = {'batchindex':[], 'loss':[], 'error':[]}\n",
    "    for epoch in range(max_epochs):       # loop over epochs\n",
    "        sample_count = 0\n",
    "        while sample_count < epoch_size:  # loop over minibatches in the epoch\n",
    "            data = reader_train.next_minibatch(min(minibatch_size, epoch_size - sample_count), input_map=input_map) # fetch minibatch.\n",
    "            trainer.train_minibatch(data)                                   # update model with it\n",
    "\n",
    "            sample_count += data[label_var].num_samples                     # count samples processed so far\n",
    "            \n",
    "            # For visualization...            \n",
    "            plot_data['batchindex'].append(batch_index)\n",
    "            plot_data['loss'].append(trainer.previous_minibatch_loss_average)\n",
    "            plot_data['error'].append(trainer.previous_minibatch_evaluation_average)\n",
    "            \n",
    "            progress_printer.update_with_trainer(trainer, with_metric=True) # log progress\n",
    "            batch_index += 1\n",
    "        progress_printer.epoch_summary(with_metric=True)\n",
    "        \n",
    "    #\n",
    "    # Evaluation action\n",
    "    #\n",
    "    epoch_size     = 10000\n",
    "    minibatch_size = 16\n",
    "\n",
    "    # process minibatches and evaluate the model\n",
    "    metric_numer    = 0\n",
    "    metric_denom    = 0\n",
    "    sample_count    = 0\n",
    "    minibatch_index = 0\n",
    "\n",
    "    while sample_count < epoch_size:\n",
    "        current_minibatch = min(minibatch_size, epoch_size - sample_count)\n",
    "\n",
    "        # Fetch next test min batch.\n",
    "        data = reader_test.next_minibatch(current_minibatch, input_map=input_map)\n",
    "\n",
    "        # minibatch data to be trained with\n",
    "        metric_numer += trainer.test_minibatch(data) * current_minibatch\n",
    "        metric_denom += current_minibatch\n",
    "\n",
    "        # Keep track of the number of samples processed so far.\n",
    "        sample_count += data[label_var].num_samples\n",
    "        minibatch_index += 1\n",
    "\n",
    "    print(\"\")\n",
    "    print(\"Final Results: Minibatch[1-{}]: errs = {:0.1f}% * {}\".format(minibatch_index+1, (metric_numer*100.0)/metric_denom, metric_denom))\n",
    "    print(\"\")\n",
    "    \n",
    "    # Visualize training result:\n",
    "    window_width            = 32\n",
    "    loss_cumsum             = np.cumsum(np.insert(plot_data['loss'], 0, 0)) \n",
    "    error_cumsum            = np.cumsum(np.insert(plot_data['error'], 0, 0)) \n",
    "\n",
    "    # Moving average.\n",
    "    plot_data['batchindex'] = np.insert(plot_data['batchindex'], 0, 0)[window_width:]\n",
    "    plot_data['avg_loss']   = (loss_cumsum[window_width:] - loss_cumsum[:-window_width]) / window_width\n",
    "    plot_data['avg_error']  = (error_cumsum[window_width:] - error_cumsum[:-window_width]) / window_width\n",
    "    \n",
    "    plt.figure(1)\n",
    "    plt.subplot(211)\n",
    "    plt.plot(plot_data[\"batchindex\"], plot_data[\"avg_loss\"], 'b--')\n",
    "    plt.xlabel('Minibatch number')\n",
    "    plt.ylabel('Loss')\n",
    "    plt.title('Minibatch run vs. Training loss ')\n",
    "\n",
    "    plt.show()\n",
    "\n",
    "    plt.subplot(212)\n",
    "    plt.plot(plot_data[\"batchindex\"], plot_data[\"avg_error\"], 'r--')\n",
    "    plt.xlabel('Minibatch number')\n",
    "    plt.ylabel('Label Prediction Error')\n",
    "    plt.title('Minibatch run vs. Label Prediction Error ')\n",
    "    plt.show()\n",
    "    \n",
    "    return softmax(z)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training 116906 parameters in 10 parameter tensors.\n",
      "\n",
      "Finished Epoch [1]: [Training] loss = 1.996150 * 50000, metric = 72.5% * 50000 5.085s (9833.1 samples per second)\n",
      "Finished Epoch [2]: [Training] loss = 1.630085 * 50000, metric = 59.3% * 50000 5.113s (9779.7 samples per second)\n",
      "Finished Epoch [3]: [Training] loss = 1.505027 * 50000, metric = 54.7% * 50000 5.147s (9713.6 samples per second)\n",
      "Finished Epoch [4]: [Training] loss = 1.414553 * 50000, metric = 50.8% * 50000 5.105s (9795.0 samples per second)\n",
      "Finished Epoch [5]: [Training] loss = 1.324908 * 50000, metric = 47.2% * 50000 5.089s (9825.3 samples per second)\n",
      "\n",
      "Final Results: Minibatch[1-626]: errs = 42.7% * 10000\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nOeecc7HiwYlzzjnnYuX/AUXNbe8RayK+AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x6ab9b70>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYjiRJElNMZxIkqSmGE4kSVJTPgDwqwvmHgkQhAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x93c9a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_path = os.path.join('data', 'CIFAR-10')\n",
    "reader_train = create_reader(os.path.join(data_path, 'train_map.txt'), os.path.join(data_path, 'CIFAR-10_mean.xml'), True)\n",
    "reader_test  = create_reader(os.path.join(data_path, 'test_map.txt'), os.path.join(data_path, 'CIFAR-10_mean.xml'), False)\n",
    "\n",
    "pred = train_and_evaluate(reader_train, reader_test, max_epochs=5, model_func=create_basic_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although, this model is very simple, it still has too much code, we can do better. Here the same model in more terse format:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def create_basic_model_terse(input, out_dims):\n",
    "\n",
    "    with default_options(activation=relu):\n",
    "        model = Sequential([\n",
    "            LayerStack(3, lambda i: [\n",
    "                Convolution((5,5), [32,32,64][i], init=glorot_uniform(), pad=True),\n",
    "                MaxPooling((3,3), strides=(2,2))\n",
    "            ]),\n",
    "            Dense(64, init=glorot_uniform()),\n",
    "            Dense(out_dims, init=glorot_uniform(), activation=None)\n",
    "        ])\n",
    "\n",
    "    return model(input)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training 116906 parameters in 10 parameter tensors.\n",
      "\n",
      "Finished Epoch [1]: [Training] loss = 2.087211 * 50000, metric = 76.8% * 50000 5.140s (9728.0 samples per second)\n",
      "Finished Epoch [2]: [Training] loss = 1.703542 * 50000, metric = 63.1% * 50000 5.124s (9757.6 samples per second)\n",
      "Finished Epoch [3]: [Training] loss = 1.555579 * 50000, metric = 57.4% * 50000 5.128s (9750.6 samples per second)\n",
      "Finished Epoch [4]: [Training] loss = 1.461751 * 50000, metric = 53.3% * 50000 5.176s (9659.5 samples per second)\n",
      "Finished Epoch [5]: [Training] loss = 1.382035 * 50000, metric = 49.7% * 50000 5.129s (9747.7 samples per second)\n",
      "Finished Epoch [6]: [Training] loss = 1.305139 * 50000, metric = 46.8% * 50000 5.176s (9659.5 samples per second)\n",
      "Finished Epoch [7]: [Training] loss = 1.243207 * 50000, metric = 44.0% * 50000 5.264s (9498.1 samples per second)\n",
      "Finished Epoch [8]: [Training] loss = 1.183436 * 50000, metric = 41.6% * 50000 5.378s (9297.5 samples per second)\n",
      "Finished Epoch [9]: [Training] loss = 1.139780 * 50000, metric = 40.0% * 50000 5.342s (9359.1 samples per second)\n",
      "Finished Epoch [10]: [Training] loss = 1.097680 * 50000, metric = 38.7% * 50000 5.328s (9383.8 samples per second)\n",
      "\n",
      "Final Results: Minibatch[1-626]: errs = 34.2% * 10000\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x9ff6c18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa0ed9b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pred_basic_model = train_and_evaluate(reader_train, reader_test, max_epochs=10, model_func=create_basic_model_terse)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we have a trained model, let's classify the following image:\n",
    "\n",
    "<img src=\"https://cntk.ai/jup/201/00014.png\", width=64, height=64>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "\n",
    "def eval(pred_op, image_path):\n",
    "    label_lookup = [\"airplane\", \"automobile\", \"bird\", \"cat\", \"deer\", \"dog\", \"frog\", \"horse\", \"ship\", \"truck\"]\n",
    "    image_mean   = 133.0\n",
    "    image_data   = np.array(Image.open(image_path), dtype=np.float32)\n",
    "    image_data  -= image_mean\n",
    "    image_data   = np.ascontiguousarray(np.transpose(image_data, (2, 0, 1)))\n",
    "    \n",
    "    result       = np.squeeze(pred_op.eval({pred_op.arguments[0]:[image_data]}))\n",
    "    \n",
    "    # Return top 3 results:\n",
    "    top_count = 3\n",
    "    result_indices = (-np.array(result)).argsort()[:top_count]\n",
    "\n",
    "    print(\"Top 3 predictions:\")\n",
    "    for i in range(top_count):\n",
    "        print(\"\\tLabel: {:10s}, confidence: {:.2f}%\".format(label_lookup[result_indices[i]], result[result_indices[i]] * 100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top 3 predictions:\n",
      "\tLabel: truck     , confidence: 96.60%\n",
      "\tLabel: cat       , confidence: 1.11%\n",
      "\tLabel: ship      , confidence: 0.96%\n"
     ]
    }
   ],
   "source": [
    "eval(pred_basic_model, \"data/CIFAR-10/test/00014.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Adding dropout layer, with drop rate of 0.25, before the last dense layer:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def create_basic_model_with_dropout(input, out_dims):\n",
    "\n",
    "    with default_options(activation=relu):\n",
    "        model = Sequential([\n",
    "            LayerStack(3, lambda i: [\n",
    "                Convolution((5,5), [32,32,64][i], init=glorot_uniform(), pad=True),\n",
    "                MaxPooling((3,3), strides=(2,2))\n",
    "            ]),\n",
    "            Dense(64, init=glorot_uniform()),\n",
    "            Dropout(0.25),\n",
    "            Dense(out_dims, init=glorot_uniform(), activation=None)\n",
    "        ])\n",
    "\n",
    "    return model(input)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training 116906 parameters in 10 parameter tensors.\n",
      "\n",
      "Finished Epoch [1]: [Training] loss = 2.121308 * 50000, metric = 78.6% * 50000 5.960s (8388.7 samples per second)\n",
      "Finished Epoch [2]: [Training] loss = 1.810775 * 50000, metric = 67.2% * 50000 6.054s (8259.0 samples per second)\n",
      "Finished Epoch [3]: [Training] loss = 1.656878 * 50000, metric = 61.4% * 50000 6.098s (8199.4 samples per second)\n",
      "Finished Epoch [4]: [Training] loss = 1.565552 * 50000, metric = 57.8% * 50000 6.035s (8285.1 samples per second)\n",
      "Finished Epoch [5]: [Training] loss = 1.494175 * 50000, metric = 54.6% * 50000 5.960s (8388.8 samples per second)\n",
      "\n",
      "Final Results: Minibatch[1-626]: errs = 48.7% * 10000\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzt3XecVNX5x/HPQ48oYAVFomKhiRqwK/bYe4tgBWuwYov+\nTGyxJ3bFEiMWdNVYsURR1KjYdzUWsKBAEKQoRXrb8/vjuePcHWZ3Z+vcmf2+X695zdxz2zlzF/bZ\nUy2EgIiIiEhSNMt3BkRERETiFJyIiIhIoig4ERERkURRcCIiIiKJouBEREREEkXBiYiIiCSKghMR\nERFJFAUnIiIikigKTkRERCRRFJxIUTCzcjO7tJbnTjCz+2Pbx0fX61N/Oaw9M1svys+5+c6LVM7M\n3jezl2p57mNmNra+85TjvWudb5GGouBEEiMWFJSb2faVHDMp2j8iY1eIXrVRnuXcel3Xwcz6m9nZ\n9XlNqZyZDYv9LFX1ur/6q+WsLj+DAf85zAetYSKJ0yLfGRDJYiEwAHg3nmhmOwOdgUVZzvkNsKyW\n9+tGw/9iGAD0Am5t4PuIuxt4Nba9AXAlcC/wdiz9u3q8Zz9q/4v+GMDqMS8iBU3BiSTRS8ARZnZW\nCCEeNAwAPgbWyDwhhLCktjcLISyt7bmFysxWCiEsyHc+GkoI4QPgg9S2mfUF/gq8F0J4NJdrmFmb\nEEK2QLiye9Y2OCaEsLy254oUIzXrSNIEoARYHfh9KtHMWgKHA4+S5S/MzD4nZnZ5lLahmT1gZrPM\nbLaZ3W9mbTLOnVBJ9X5bM7vHzH4yszlm9qCZdcg490Aze8HMJpvZIjMbZ2Z/NrNmsWPeAPYDUn1H\nys3s+9j+1lF+vzazhWY2xcyeMrMNspTz5Ogei8zsQzPbsrovNNZctpOZDTWzacCkaN8DZjY+yzmX\nm1l5Rlq5md1mZgeZ2edRHr4ws72quf9aZrbUzP6SZd8m0XUHR9stzOwyM/sm+i5+MrO3zWz36spZ\nF2Y21cyeMLP9zKzUzBYBx0X7Tjaz181sWpSnz81sUJZrVOi7YWZ7RWU7MPo+J5vZAjN7xczWyzi3\nQp8TM+uW+l6i13fRvd81s82z3HuAmY2Njvk0Kket+7GYWcfoZ2N6dM1PzKx/luOOM7MyM5sb/fv6\nr5n9Mba/lZldZWbfRteZYWb/MbOdapMvaTpUcyJJNAF4H+gPvBKl7Qu0Ax4Dcum7kapefwL4HrgI\n6AOcBEwDLs5ybJwBdwCzgMvwpp/BwG+BXWPHnQDMBW4E5gG74c0HqwB/io65CmiPN0mdE117HkAU\nxLwYXbMEuCU69/fApkA8cDgaWBlvsgjR9Z8ys645/uU9FJgOXAGsFCt7tvJXlt4PODS61lzgLOBJ\nM/ttCGFWtpuGEKab2X+AI/Hai7ij8Oa4J6LtK/BndS/wEf7Mt8Sf3ajqi1hrAdgMeBAv293Al9G+\nwVFensGb/w4G7jOzEEIYlnGNbC4DFgPX4UH3hcADVPw5quz7PhFoA9wJNMef+ZNmtkkIIQCY2aHA\ncLxW8U94zeLDwJQq8lQpM2sLvIP/vN4G/AD8AXjEzFYOIfwjOu6AqBz/Bu7B/9jtBWwH3BVd7lr8\n3+tdwCf4v4OtgS2At2qaN2lCQgh66ZWIF3A8sBz/RTQYmA20jvY9DrwWfR4PjMg4txy4NLZ9WZR2\nb8ZxTwHTM9LGA/dn5KMcbxZoHks/P8rf/rG01lnKcRf+i7tlLO154Pssxw6M7nVWFd/LetEx04F2\nsfQDovzsm8P3Wg68CVjGvmGV5OsyYHmW73ghsH4srXeUPriaPJwc5bVnRvoXwKux7U8yn209/Wz1\njfJ5XCX7f4zyt2OWfdme8evA5xlp7wEvxbb3iu5ZlvFzdEF0r66xtBJgTGw71Q9qMtA2ln5EdO5u\nsbSvgW/j+cSD2/L4Nav4bjLz/afoHgfH0lrgwc/PQJvYz/nUaq49Fniivp+nXsX/UrOOJNUT+F/3\n+5vZysD+wCM1vEbA/6KLextYPbpmde4NFWsk7iIKBn69QQiLU5/NbGUzWx3/q3MloHsO9zgUmIHX\n0lTnsRDCL7Htt/FamK45nBuAf4QQ6joy49UQwoRfLxrC58AvOeThafy7+0Mqwcx6AT3x2rCU2UAv\nM9uojvmsjbEhhHcyEzOecXszWwP/q7+HmbXK4br3ZfwcpTrk5vLcHgkhzM8499dnHjX9bQwMi+cz\nhPAqHrDUxj7AxBDCs7HrLQNuBzoAqZF0s4H2ZrZbFdeaDWyWrYlSpCoKTiSRQgg/Aa/hnWAPxX9W\nn6zFpf6XsZ1qeli1uiwA4zLyNB//C3v9VJqZ9TSzZ8xsNv5LegZepQ5ehV2dDYGvQ8WOv5WZlJGf\n2dHH6sqSMiHH43LOQ2RWdXkIIfyMN8scGUs+CliKN5ekXIr/AvzGzD4zsxvMrHfdspyzFfregI8S\nM7M3zGw+XtbpUT4Nb3aqTuZ3Nis6N5fnlu1cYuem+q5kG3U0LktaLtYDvsmSPhbPd+qetwMTgVfN\nbKKZ/cPM9sg45xKgI/Bd1BfmWjPrWct8SROi4ESS7FG8luI04N8hhLm1uEZlfTHqPGzTzNrjf0H3\nBv6M1+7sQbqvSX3/+6prWRZmSausJqV5A+ThMWATM9ss2j4CGBVCmPlrZkJ4Gw/YBgKf430uyrJ1\nQG0AK3w/ZtYdGAm0xftO7Is/41RNVy7PuC7fWYP9/NZVCGEK/rN/CN5vag9gpJndFTvmdfx5nogH\nN6cCn5rZ0Y2fYykkCk4kyVIdELfBA5XGZHh1eTrBOwquTboGYhf8L9jjQwh3hBBeiv4zns2KKgsC\nvgO6mVllwUBDm4XXVGRavwHu9SxeU/KHaMTJJnhfiwpCCLNDCA+GEI4GugCfAZc3QH5ycRDe32Lf\nEMJ9IYSXo2eclOHnE6P3bM1gtW0am4g/m0w98J/j1D0JISwNIYwIIQzGm5oeAE4xs3Vix8wMIQwL\nIfTHO5R/jfdpEqmUghNJrKgZ5TT8F9PzecjCKWYWH9E2GK9RSA0XXY4HMfFhw62i4zLNJ3szz1PA\nmsAZ9ZHhWvgO7zewaSrBzNbGR6TUqxDCHHz01ZF4k85i4Ln4MWa2WsY5C/DmidaxY9pFQ21zaVKp\nq1TNRfwZr45PmpaLBp19NYQwHu9bcoLFhsibD+/euNITq/YSPuz9oNj1WuA/o7OB0VFa5rMKeAdn\niJ5XlmPm4aPnWiNSBQ0llqSpUF0dQni4sgMbQStglJk9gXdu/SPwdgjhhWj/u3jNw0NmdluUdgzZ\nfyGVAkea2Y34sNR50XUewufTuMnMtsE7PK4M7A7cGUKor6CssmaAx4DrgWejMrTFA8Kv8VFT9e1x\nfNjrYOCVjA6+AGPM7E38+5oJbIXPb3Nb7JhD8FFGJ+DfX0N6GbgG+LeZ3YfXMp2Cj6JZYTLALBqj\n+eUS/Ht9x8weAtbCf1a/pHZ/gN6JD7l/1MzuwPu9HIX/PJwW63g73Mxa46PAJuM1J2cAH0RBE3hf\nk3/jI5Zm4cOM9wduqEW+pAlRcCJJk8tfmtnmhKjruibZrncGPrfIFUBLfLTQr3OshBBmmtl++Bwn\nf8X/830YH2b6Ssb1hgKb479Qz8Grxl8IIZSb2T74L5hU59+f8SDl8xzKl2u5sx4TleFg4CY8SBmP\nzzOyCSsGJ3XNA8AIvG9HWyqO0km5FTgQHwrbGv+e/g/4e5Z71lRV52QtQwjhCzM7An++N+K/hG/G\na32G5nCPyu5Z2feY67m/7gshPGlmxwJ/wZ/h13iQPBhYJ+sVqrh3CGG+mfXD52UZiM+7MxY4OoQQ\nf2YP4H1JBuNB2494sHhF7Jib8QkI98Kf53h8KPUtOeZLmiir+8hCERFJmmh22G9CCAdVe7BIwuS9\nz4mZXWw+Dfcv5tNDP2Nm2TpjVXb+DuZTY5c1ZD5FRJLIfMr/Zhlpe+MTub2Rn1yJ1E3ea07M16Io\nwWcfbIFPd7wp0COEkG3oY/zc9njb9LdAxxBCQ7SRi4gklpl1w0dCleBNK73wIbtTgc1qOQRfJK/y\nHpxkimZfnA7slG22xoxjS/DJgsqBgxSciEhTE42IuQvYAe+k+wvwKnBxCCFzEkKRgpDEDrEd8M5Z\nM6s6yMwGAhvgHRZXWO1URKQpiCax+0O1B4oUkEQFJ2ZmeC/ud0IIY6o4bmN8eN+O0WiHxsqiiIiI\nNLBEBSf40LyeePVkVlHHr0eAy0IIqfUkqo1OoomT9sJn91xU55yKiIg0HW3wmaNfidbKalCJ6XMS\nTfZzANCvqnbSqBPsLGAZ6aCkWfR5GbBnCOHNLOcNoOar2oqIiEja0SGEBl9OJBE1J1FgchCwcw4d\nuH7BR/PEnQ7sChxG5SuvTgAYPnw4PXr0qHVeC8GQIUO4+eab852NBqdyFheVs7g0lXJC0yjr2LFj\nOeaYY6B+VjevVt6DEzMbCvTHZ4Wcb2Ydo11zQgiLomOuATqHEI6P1m8Yk3GN6cCiEMLYKm61CKBH\njx706VPcg3rat29f9GUElbPYqJzFpamUE5pWWWmkbhF5n4QNX8ejHb4+w5TY68jYMWvjq5OKiIhI\nkct7zUkIodoAKYQwsJr9V1BxPQcREREpUEmoORERERH5lYKTItS/f/98Z6FRqJzFReUsLk2lnNC0\nytpYEjOUuKGZWR+gtLS0tCl1XBIREamzsrIy+vbtC9A3hNDgC+2q5kREREQSRcGJiIiIJIqCExER\nEUkUBSciIiKSKE02OLnkEnjllXznQkRERDI1yeAkBBg9GvbeG045BRZpjWIREZHEaJLBiRmMGgW3\n3grDh0Pv3nD++TBlSr5zJiIiIk0yOAFo3hzOOgveegu23x7uuQe6dYMdd4Snnsp37kRERJquJhuc\npGy5JTz4IHz9NQweDC1bwtSp+c6ViIhI05X3hf+SYp114Prr850LERERafI1JzX16qvw6KOwfHm+\ncyIiIlKcVHNSA2+9BXvu6Z/ffRfuuCO/+RERESlGqjmpgd69vbPsaafB3XfDk0/6sGQRERGpPwpO\namDVVeHQQ+Fvf4OttoIjjoCDD4alS/OdMxERkeKR9+DEzC42sw/N7Bczm2Zmz5jZJtWcc4iZjTSz\n6WY2x8zeNbM9GyvPK68M770Hp54KI0bA6qvDzz831t1FRESKWxL6nPQDbgc+xvNzLTDSzHqEEBZW\ncs5OwEjgYmA2MAh43sy2DiH8txHyDHjTzu67w6RJsNpqjXVXERGR4pb34CSEsG9828xOAKYDfYF3\nKjlnSEbSJWZ2EHAA0GjBCXjTjoiIiNSfvDfrZNEBCMDMXE8wMwNWqck5DSkE+OabfOdCRESkMCUq\nOImCjFuAd0IIY2pw6gVAW+CJBslYDd1wA2y+OXz6ab5zIiIiUngSFZwAQ4GewFG5nmBmA4C/AEeE\nEH5qqIzVxJlnQteusNNOMHlyvnMjIiJSWPLe5yTFzO4A9gX6hRB+zPGco4B7gcNDCG/kcs6QIUNo\n3759hbT+/fvTv3//Gua4ciutBP/6F/TqBeuuCwMGwCOP1NvlRUREGkxJSQklJSUV0ubMmdOoebCQ\ngFnEosDkIGDnEML3OZ7TH7gP+EMI4YUcju8DlJaWltKnT5865TdXjz0GqZjngw9g660b5bYiIiL1\nqqysjL59+wL0DSGUNfT98t6sY2ZDgaOBAcB8M+sYvdrEjrnGzB6MbQ8AHgTOAz6KndOusfNflaOO\nghkz/LOmuhcREclN3oMT4DSgHfAmMCX2OjJ2zNpAl9j2yUBz4M6Mc25p+OzWzBprwPjxcM896bQp\nU2DbbWHoUBgyxGtVRERExOW9z0kIodoAKYQwMGN714bLUf1bf/2K26NGeUCSCkpuuQXmz/e+KiIi\nIk1dEmpOmpxjj4Wvvsp3LkRERJIp7zUnTVW3bjB2LKyyCrRvr1oTERGRFAUnedS9e75zICIikjxq\n1kmY667zjrIiIiJNlWpOEmTZMrj4Yv/cuTMcdFB+8yMiIpIPqjlJkBYt4JVXoHdvnxdl4cJ850hE\nRKTxKThJmD33hOOOg9de806yV19dcX9pqdewiIiIFCsFJwl0yinpzy1iDW8//ghbbqmOtCIiUtzU\n5ySB2rWDbEsejRnj7999500+v/lN4+ZLRESkMajmpEBcdx3ssYd/vuACNe2IiEjxUs1JgUiN4rn+\nerjwwvzmRUREpCEpOCkQX3zh/U+6dauY/sknMGwY3HZbfvIlIiJS3xScFIhevVZMCwH69PHPnTr5\nvCjZjhMRESkk6nNSwMzgoov88yWXwKabwvLlvj1sGLz6avaOtSIiIkmm4KTAXXstrLaaf37mGWje\nHBYvhkGDfM6UZs08iFmwIL/5FBERyZWCkyLw889eQ3Lwwb7dujXccEPFY8aNa/x8iYiI1IaCkyJ1\nwQUwa5bXmNx9N2y2Wb5zJCIikpu8BydmdrGZfWhmv5jZNDN7xsw2yeG8Xcys1MwWmdk3ZnZ8Y+S3\nkHTo4BO1nXpqvnMiIiKSu7wHJ0A/4HZgG2APoCUw0swqnf/UzNYHXgBGAZsDtwL3mdnvGzqzhWzU\nKO9/kpppVkREJInyPpQ4hLBvfNvMTgCmA32Bdyo57Y/A9yGE1HRkX5vZjsAQ4NUGymrBmzHD33v1\ngieegCOOyG9+REREsklCzUmmDkAAZlZxzLbAaxlprwDbNVSmisHhh6cnazvySHjrrfzmR0REJJtE\nBSdmZsAtwDshhKoaHzoB0zLSpgHtzKx1Q+Wv0LVoAWeeCT16+PZ99/n73Llw8smw226w664wfnz+\n8igiIpL3Zp0MQ4GewA75zkgxGzMGvvkGNtoonZYKVCC9qOCZZ3oT0GmnNW7+RESkaUtMcGJmdwD7\nAv1CCD9Wc/hUoGNGWkfglxDC4qpOHDJkCO3bt6+Q1r9/f/r371/DHBe2TWLjoVZZBR56CI47zrdT\nQctzz8Edd8CJJ3qty7Jl8O230LNn4+dXREQaR0lJCSUlJRXS5syZ06h5sJCA+c2jwOQgYOcQwvc5\nHH8dsE8IYfNY2qNAh8wOtrH9fYDS0tJS+qQWpJEKTjjBJ3C75x7fPv98uPFG/3z88dCvH5x0Eqy3\nntempKbOFxGR4lZWVkbfvn0B+oYQyhr6fnnvc2JmQ4GjgQHAfDPrGL3axI65xswejJ12N9DVzK43\ns25mNhg4HLipUTNfZB54IB2YAAwZkv5cXg6dO/vniRPh4ov9cwgwfToMH+5r+YiIiNRVrYITM9s7\nGrqb2j7dzD41s0fNbNUaXu40oB3wJjAl9joydszaQJfURghhArAfPi/Kp/gQ4hNDCJkjeKQOOnf2\n2pMzzvBmn733hq22Su+fPRuuuQY6dvQalRdegGnTfC6V1KggERGRmqptn5O/AX8CMLPewI14rcWu\n0fvAXC8UQqg2QAohrHC9EMJb+Fwo0oD+9reK26+9Bl99BS++CO3aeZ8U8MUGt9kGUt15zj4bzjqr\ncfMqIiLFobbNOhsAqaG+hwEvhBD+Dzgd2Kc+MibJ1K4dbL01XHGFr3gcH3bcuTO0aQPnnefbCejO\nJCIiBai2wckSYKXo8x7AyOjzTLyJRpqINm28NuXUU2H77T1th2gg+E8/5S9fIiJSuGrbrPMOcJOZ\njQa2Bv4QpW8C/FAfGZPCsfvu/kpZf31/nzAB1lwz+zkheN8UERGRTLWtOTkDWIaPkPljCGFylL4P\n8HJ9ZEwKVyo42Xrr7Pvff9+bhO66q9GyJCIiBaRWNSchhP8B+2dJH5LlcGliOnTw9802S6fFa0qu\nvNLf77kH/vjHxs2biIgkX22HEveJRumktg8ys2ej+Uha1V/2pBCZeTDy3/+m0664AtZZB5Yv95E9\n4PvPOAOWLMlPPkVEJJlq26xzD96/BDPrCjwGLACOAG6on6xJsfjySw9Opk+H5s1h1ChPA7jzznSw\nAhrhIyIite8Quwk++Rl4QPJWCGGAme2AByrn1EfmpDhsuqm/r7xyOq1nTxg0CLbYAubP9yHKKaWl\nvr7P5MnQtq3XxHTpgoiINBG1rTmx2Ll7AC9FnycBa9Q1U1I84jUho0ZV3PfPf/rKxwsWVEwfNcrX\n8unZ09fx+e1vffr8khIPVMxg7tyGz7uIiORHbYOTj4E/m9mxwM7Ai1H6BsC0+siYFIeFC3048ejR\n0LeS+XzXXx8GDPDPI0bABRfAd9+l93fqBI8+mj4GfIVkEREpTrVt1jkHeAQ4GLg6hDAuSj8ceLc+\nMibFYaWVYOpUHzpcmWbN4JFH/JVy9NHw44++Zs+mm8LSpXDssb5v7tyKTUQiIlJcajuU+DOgd5Zd\nFwDL65QjKTpVBSaV+dOfKm63bAkzZ/q1FJiIiBS32tacAGBmfYEe0eaYEEJZ3bMkkt2qNV3vWkRE\nClKtghMzWwt4HO9vMjtK7mBmbwBHhRBm1FP+RGrk55+hdWvVroiIFLLadoi9HVgZ6BVCWC2EsBqw\nKb7o3231lTmRmli8GNZYA1ZZxfuoiIhIYaptcLI3MDiEMDaVEEIYA5yOr68j0qCuvdaHFH/zTTrt\nnNjsOs891/h5EhGR+lHb4KQZkO1v06V1uKZIzm680d9PPdXfQ4C77/bPZ50Fhx+en3yJiEjd1TaQ\neB241czWSSWYWWfg5mhfjZhZPzMbYWaTzazczA7M4ZyjzexTM5tvZlPM7J9mtlpN7y2FacoUf3/z\nzfScKBts4Csd33pr+phhwzRhm4hIoaltcHIG3r9kgpl9Z2bfAeOBVaJ9NdUWnw5/MFDt6irRNPkP\nAv8AeuLzq2wN3FuLe0sBatUKHn7YPx96qDfxfP89nHZa+pjXXvMp8tu1U4AiIlJIajvPySQz64NP\nXd89Sh4LfAVcCpxSw+u9DLwMYGaWwynbAuNDCHdG2xPN7B7gwprcVwrbgAE+Mdtnn/n09pnzqXTr\nlv683XbwxReNmz8REamdWvcPCe7VEMLt0es1YHXgxPrLXqXeA7qY2T4AZtYRX4DwxSrPkqLSrBnM\nng3jxmWf6G2bbWD8eP/89dcewIiISPLVaRK2fAkhvGtmxwCPm1kbvBwjqF2TkhSw9u39VZn114cf\nfvBhxqkAZtEimDHDVzpevtzX6WndulGyKyIiOSjIkTVm1hO4Fbgc6APshS86eE8esyUJ1bkzdO3q\nn8ePhy228JWOx471vitt2niQsv763k9FRETyy0Kotv9p7hcz2xwoCyE0r8M1yoGDQwgjqjjmIaBN\nCOHIWNoOwNvA2iGEFVZGjvrIlO600060z/hTu3///vTv37+2WZYCMW+eT9CW0rYtzJ/vn6dNg44d\n/fPpp8Mdd3hz0S23wG23ea3LwoUezDSv4qd7/nyfnfbee+HkkxuuLCIiDaWkpISSkpIKaXPmzOGt\nt94C6NsYS9XUKDgxs6erOaQDsHMjBCdPAktCCANiadsB7wCdQwhTs5zTBygtLS2lT58+tc2eFLhU\nd+s//xk22QSOO863X3oJ9t03fdzLL8Pee/vnzz+HXr08QDnlFLinivq5jz+Grbbyz5MnwzrrVH6s\niEihKCsro2/fvtBIwUlNm3XmVPOaCDxU00yYWVsz29zMtoiSukbbXaL915rZg7FTngcOM7PTzGyD\nqNbkVuCDbIGJSMqkSbDLLnD22T7S55BD4OmnYZ994O23/Zj11oMFC9LnPPAAfPutf773Xq9lOeQQ\nuOACrx1ZtAjmzPEmo3bt0ucNGtRYpRIRKS416hAbQhjYQPnYEngDn+MkANH8nzwIDAI6AV1i+XjQ\nzFbGp8v/O7744CjgogbKnxSJddeFN95Ibz8dqwvccUefaRbS7126wFVXpWefBZ9PZeWV4e9/9+3b\nboMOHdLnffMN7LknXHppw5VDRKSYJWK0TgjhP1RRi5MtKIrmOLkzy+EidWYGS5ZAixb++aabPH3S\nJA9wXnklfezRR6c/Dx8OxxyTHsKczZAhcOCBsOuu2fdfc42PHjrvvLqXQ0SkEBXkaB2RxtCyZbqP\nyttve8Cx7rq+veaa/n7UUT7XSsqcOenPIXjtSbymZuBA72S7227ptI8+8toY8H4vl1wC559f/+UR\nESkUiag5EUm69daruD1okNduDBrkU+O//rr3Q4kzg8ce84Bl1119uPIDD6T3f/wxDB3q6/907Aij\nR8O//53e//nn0Lt3gxVJRCSxVHMiUgu/+Q2cdJKP4GnffsXAJGXLLb1mBHwEEEBJiU8Kt3ChBybg\nw5fXXRcOPti3X3kle2Dy5ptw2WX1WhQRkcRRcCLSgJYtg/feg+nT06snH3GEz5fSs2f6uIsv9pqY\nZ57x5qA99/QAZo894J130sftuitceWXD5XfMGK/x+fzzhruHiEh1FJyINKANN/T3W2+FE06ATz5J\nT+K2+urp41pkaWD9979h1Cjo12/FfS1bZr/f+PEeYMT7wcQtXlx1flOLIz7+eNXHiYg0JAUnIg3o\n2mv9/YUXfA6ULbaouP/JJz1gyWavvdKf99nHa1RKS3172TJ/nzzZhzIfdpjPt9K1q08Yt+qq3mwU\nN24crLaa93uZNy/7PVOddw88MOciiojUO3WIFWkElTWTHHZY5ef85jfex2SXXdL9Vfr0gWef9eBi\n3rz06CHwYc5xzZt7gLLXXj5qaNIkn1xu4EC4+mqfjyU1Gglg6dL0fC5bb13TEoqI1B/VnIg0sPJy\nH6lTGzvvDJtt5s0sqUDioIN8bpVWrdLH9ezp6wYdeihss40vYNiqlQc3b78Nv/+9L3SYMm5cxcAE\n0h134wGPiEg+KDgRaWBmKwYCNfHf/8KRR66Y3qoVrLEGdOsGX34JnTrBU0/B++/D7rt7n5XUekHn\nnuvDmtdeG844w/ueTJ3q+boxmo95o43gxBPhq69qn1cRkfqg4ESkgM2YUXkwMW5c+vONN8Jnn3lT\nzu23e2AzYYLvO/98D1bWWgvuu89XawaYOROuv95rfuorr4cdBrNmwd/+lr6/iEgmBSciRSq14vIe\ne/h7hw6+JlDKttumP2eOzlm82EcTXXSR18rMmFHzpqnmzb1mJhXcPP64r2V0111w4YUVp/0XEYlT\ncCJSpNqF+dHbAAAddElEQVS394Di1VcrPyY13f7AjNWr4nOpfPml16pkG+5cmfLydFDy1lv+/mC0\nrnhqFNK77+Z+PRFpWhSciBSxZtX8C2/XDo4/3msy4v76Vx8ZBOmOslCxU23KJ5/49P3gM+WmFkp8\n801P23NPD5I+/ti3+/atcTHqzYIF3odHRJJNwYlIE/fAA+n5WFKaNfM5Ve6806fUv/VWT0/VfsT1\n6eMdcEPwYc4AF1zgI43AhyhvvLF//sc//H3//f29vvqzVOfRRz0Pf/mLzzVz//2Nc18RqR0FJyJS\nqcGDfYbaM8/07euv9/dly3xk0PDh6WOHDk1/fuEFfy8rg733hk039e1ddvH355/3YCazZmfZMl+h\neenS+i3H4497B+GbbvLtE0+s3+uLSP1ScCIi1TLzkTapTrQtW8Lhh8Oxx6ZrIebN8/lVli6F/fbz\ntN/9zoc0P/SQ115stFHV95kwwaf8f+st74T7xhsrznRbU1OnwogR/jkVXAF8/XXdrisiDUczxIpI\nTu64w/trLFiQTrvoIu9Me9xx6TWDsunQAfr3XzF90SKYPz+9zlBqeHHHjt4JFzwI+te/ap/vrl39\nfe+9vW/NccfBDTfABhvU/prgnYnbtavbHDYikp1qTkQkJ506+UibqVN9e9tt031VqgpMKrN0qU/R\nv8Ya/nnZMhg50pt6fvklfdz//pf7NZcs8euEkA6ijjnG3x95JF2Om26qOMNuTS1e7AHXWWelRzyJ\nSP1JRHBiZv3MbISZTTazcjOrdtkxM2tlZleb2QQzW2Rm35vZCY2QXZEmbf31Yfr09CKBtRUfmnzp\npXDOOT45W3k5bL+9p59yCnzwQcXzFizwEUDZFkwcNswDnkcf9cnk/vY3uOce+OknX/Qw09SpcPrp\nNe+Y27q1v99xh8+6KyL1KxHBCdAW+BQYDIQcz/kXsCswENgE6A+oFVmkgTVrBmuuCW3a1O06ZumO\nr9ddB5ddVnF/CB5YZPruO5+7ZfJkb/ox834tP/zgM9yusooHOuDNOGbpZqNMBxzgHXlffDH7/q++\ngr//vepy1LVPTH2aP79is5tIoUpEn5MQwsvAywBm1bfgmtneQD+gawhhdpRcg8pfEUmCeO3Jmmv6\nsOaWLbMf+/DDHmSkOtuus453mgVfQ2jgwPRcKqNHww47VH//JUv8/aOPPFDJ1KOHv++3H3Tv7vdL\n9YVZsgSuuip9TBKkZgD+6itfc0mkUCUiOKmFA4CPgT+Z2bHAfGAE8JcQwqK85kxEauTii9M1G8cf\nv+L+kSPTs8rGrbOOr6D8ww++3atXet/22/s0+dVN+PbBBz4fy1//6nOgVBYYvfqqr/wM8OOP3m+l\nZUu44oqqr9+YQqzOuXt373tTm75AIkmQlGadmuqK15z0Ag4GzgYOB+7MZ6ZEpOauuQbOO6/y/fEZ\nXQcP9vdtt/UAYdIkmDgRvv0WhgyBzTaDUaP8mNNOg622qvrebdp4HxVIz4GydKk3Mb39dvq4+fPT\nn+PpSbJggQclKVUFJi++6M1dmWsqiSSFhZBrF4/GYWblwMEhhBFVHPMKsCPQMYQwL0o7BO+H0jaE\nsDjLOX2A0p122on27dtX2Ne/f3/6ZxvnKCJ5t2yZz6fy2GMeOEya5CN8Vlmlfq7/ww+w3XY+XHnb\nbeHUU+Hee2GnnXz9nwkTvJZm0CBvdurUyWtPGsvSpdC7t8/Sm60GaeZMeOcdODAaRvDll7DSSjB3\nri8rkOp/E9e9e3qel6FDPegrK/N5aRrL4sU+YkpDsZOnpKSEkpKSCmlz5szhLV8oq28Ioayh81Co\nwckDwPYhhE1iad2BL4FNQgjfZTmnD1BaWlpKn9SiISJSEJYv9+HFq67asPc58cT0pHJTplQciROC\nBy1bbZVed6gmZs2C8eNrfm58Jt3U3CopS5emh0R/8016mQDwQGrgQO8s3Lev9+kB/y7XWcdHXIFf\nu7zcy/Xhh7nna/p0Xz/pyCNrVp6Uc86B//wn+6grSZ6ysjL6ejtpowQnhdqsMxpYx8xWiqV1A8qB\nH/KTJRFpKM2bN3xgAunA5NxzVxwibOa1KtmCi9mzvfYiZZdd0gsgjhvnQcVqq9Vu0UOzdO3H88+n\n08eN8yHTKZnrHqXyuc8+8Nxz6fQpU3w233PPhZdegs8+8/SPPvJA4fTTc8vXiSfCH/7gZa+NCRM0\nDFsql4jgxMzamtnmZrZFlNQ12u4S7b/WzOL/9B4FfgaGmVkPM9sJuAH4Z7YmHRGRXKQWLjz88Jqd\n17+/r0Fk5jUdqf4e550HTzyRXjixJsrL4be/9blUUsOsTzopvf+oo9L57dMHzj+/4vmpmXGh4kik\nLl28D82NN3rg0qsX/PGP6esMHZp99em4JUvS6ydVNgy7OhMm+Jw5VVm40EdHjR9fu3tI4UpEcAJs\nCXwClOLznNwIlAGpvvCdgC6pg0MI84HfAx2Aj4CHgefwjrEiIrVy4IHeZ2O77Wp23pVXpj//858V\np9vfY490cNGtm9d+VLaw4cKF8Omn/vn2271/zZln+my0LVr4dP8TJ3pzTWkpbLmlB0OlpX5M3Mor\np/tzdOzo76Wl6dFNcbfc4jUmL7/s299/X3V540PAb7654r5LL/X7pmYSjlu0yDs4b7SRv6+3XtX3\nWWklH74dD7TqSwhe8xTv7CwJEkJoEi+gDxBKS0uDiEh9mzo1BP+V59vbbuuf58wJYdCgEA48MIQf\nf0wfU15e8fxvvknv69o1/fnuu33/6af79qmnpve9/nrVeVqyJIR58/zz5pv7ORttVPnxS5f6MRtv\nXH15f/nFy7jHHum0mTPTeTvjjBXPufzyEFq1Sh9z//0hTJrkecz8PkIIYaedKn6n9WnkyPS1Fy+u\n/+sXm9LS0oBXHvQJjfA7Oyk1JyIiBa1jR5/75MQTffu997yGpF07r0157rl0DQakhzyXl/t8KfE5\nXo46Kv351FP9/dprfRjzX/+a3rfrrlXnqWVLn8Yf0kOyK6u1gXSNyLffZt8fHz+xyio+fPuww9Jp\n06alP7/22ornjx7t/XHefhvOOMMnt+vSxWt5dtwxfdwnn/j8Ny+/DJ07w4AB3qcmXkNVV6nnBBW/\nU0mGQp2ETUQkcWbN8qaIlBYZ/8Oa+TGrrur9Wl56KT2T7fPP+3De1AKLc+f6L+iUVVZJ/wL/6quq\ng4yqVNdH5NxzYfhw/7x0acWJ6Z5/3vunfP65d/DNHKnTvbsHMGee6RPkZZo2zfvm7Lijv1Iz9EJ6\nKPR116XLfc016WaoVBPV//3fit9rbZx7rgdXv/sd7LZb3a8n9awxqmeS8ELNOiKSEKnmhK22Sn8e\nN65h7zljRghPPln9cffdF8KGG4Ywe3YIhx3mTTUprVtnb5I65xxPf/fdyq97wgl+zF/+UjE9Vf5p\n00L49NP0dmZTTirt1FOrL0MuysuzNyWFEMKyZSFceKE3w02eHEJ9/9pYsKDi91oI1KwjIlLkDjjA\nR8q89JJvr7UWbLhhw95zjTUqNsFUZs89ffTRDz/AU0+lhxqDT5wGK06cdsst/p5aTTpuzBh4+mlf\nGwnS6/+kjBzp+9daq2KzV3xoNsD11/v76NHVlyEXZtkngCsv91FRN9zgq2J37uxDwLN18K2tk07y\nmqfly317yRLPS7t26qCbouBERKSRjRjhgckaa3h9QLyvRr516eJ9MDbZxKf2jwcJ3btnn3E21TSV\nOYPtrFnwzDO+lIBPLuojjOJ+/3s45BD/3KlTut4kc+HGCy+Eq6+G/9VxideyMrjttoppy5al+wiN\nHOkT2EHFeWWqag77xz/gT3/KPQ+bRNOHDhvm7xde6O9z59ZPk1Ux0NcgIiIraNnSF2T885999tgW\nLbyvyy67rHjsG294/5ALLkinDRvmU/6DL5q4/fZeKxDvk1NTRxzhU/mXl8P772evqanOX//qQcBZ\nZ6XTPvnErzVkCOy+ezp93Dgf9jx6tC9tkGnaNF/i4JRTfPv0031umur07u3vJ5/sAWBqHpzPP4fW\nrWtWnuXLPZgrtqBGNSciIpJVaor7667z4ANgiy1WPK5lS5/LJR547Lxz+vPee/t7XQIT8On5DzgA\nzj7bO9YuyrIG/bhxlc/T8uCDPnFdarHHlE039febb/ZA5IMPPADacEP/xb/99j7N/913V5wQbq+9\n4NBD00FZdfPDpBxySHrG42OO8ffzzkvnoyb23Tf7mksFrzE6tiThhTrEiojUyH33hbDSSiF8/71v\nL1hQeSfSTIsXewPNUUf5fCv16eOP/drvvFMxvbw83XH22mtXPC+179lnK9/30UfZ7zl5su/fffd0\nWmoum9mz/X2LLbzjb6622y6EZ54JYfnydNqgQSEcfHDu12ioeWAyqUOsiIgkwoknelPMBhv49m9+\nk/sqwq1a+a/NkpKKw5Hrw+abey3Me+9VTF+wIP05PgwbKq4BdNBBlV+7spWZf/654vtXX3nT0rnn\nQmqh+08/9f4qla1a/dZbFWtX3n0XDj44vbBjyrPPpjsf5yrVubZYKDgREZGC0qKFdyrNnCxu2bL0\n5yOOSH/ecUef+O2qq1YMaFLmzvVmotS6SJl69/ap+T/91Fdv7tHD01OdW+PWWcffFyxIBxn77ONN\nXZmdcTOlJs1r0yadr6oWV0xdb9w473icWvOo0Ck4ERGRgtO164oLArZv78NyQ/AFF0eO9Jqe0aO9\nBuiSS7J3bAUf4lxdZ9RU0LHNNnD55f45tRhjCL76dMr8+R5opIZOp9Ytateu6ntcdVX684wZfvxe\ne3mfnvistim9evl79+5wzz0+824xUHAiIiIFZ4MNsndAjTchXXRR+vNWW9X9nvvv7+8ffODBQggV\na1ratfMFG8FHOYHX5syblz7muOOqvke7dumhxam5XT780AOf++/3gCtu1109D88+C5tt5gtDhkDB\nU3AiIiIFp2tXD06WLEkPLc78pXz66f7+zDP1M8ld585+j623rvyYwYN9uPLuu6eH9152mQcp8+f7\n0OTqXH01fPFFuu/Kjjt6zQikR+ZccIGvNWTm1z7ooPT6TJk1SoXIQjGEWDkwsz5AaWlpKX369Ml3\ndkREpA7ee89/ab//vtcc9O0L//qXr1mUFEuWpJuKavOr9tBDPbBavtwniGvTxudRmTgx3TE5ft2J\nE2H99b0J6d13vZ9LZc1YNVVWVkbfvn0B+oYQyurnqpUrsmlbRESkKdh6a6+JaNMmPafKNtvkN0+Z\nWrXyQGHmzNqdf9VVcOyxPpqndWsYOBDGjvW+KJCeyTYlNf3///2fz4R75ZXp4GXJEp80bt11vZPu\n1197M1Cuo68am5p1RESk4DRv7oHJggXpqfFXXz2/ecpmr718raLa6NkzPbU/+GRw/fr5OkSw4hT/\nqRE+8Sas8nIPQFq39lqXsjKfyG6LLeD882uXr8ag4ERERArWgw+mP9d1BtqkSy1ImJKt/8qPP8Lw\n4el1geLrAwGstx689pp/rm5Ycz4lIjgxs35mNsLMJptZuZkdWINzdzCzpWbW4G1ghaKkpCTfWWgU\nKmdxUTmLS2OV87TTvK9JTSctq0+N/Uzbt/dROtl06uTNSfvu69uvvpred8EFFRdOjE9alzSJCE6A\ntsCnwGB8etycmFl74EHgtQbKV0HSf37FReUsLipn/TLzTrCtWjXK7bJq7Gf6888Vg45smjXz4dNj\nx3ozzrffeq1Lnz7w9NMwZkz9z9xbnxLRITaE8DLwMoBZjbrn3A08ApQDVUxILCIiUhwqm8U2U79+\nsMYaK07lH+/HklSJCE5qw8wGAhsARwN/yXN2REREEuXGG/Odg9oryODEzDYGrgF2DCGU16yyRURE\nRJKs4IITM2uGN+VcFkL4LpWcw6ltAMaOHdtQWUuMOXPmUFZW/P2DVc7ionIWl6ZSTmgaZY397mzT\nGPdL3AyxZlYOHBxCGFHJ/vbALGAZ6aCkWfR5GbBnCOHNLOcNwIMaERERqZ2jQwiPNvRNCq7mBPgF\n2DQj7XRgV+AwYEIl572C90+ZACxqoLyJiIgUozbA+vjv0gaXiODEzNoCG5GuCelqZpsDM0MIk8zs\nWmCdEMLxwat6xmScPx1YFEKotM0mhPAz0ODRnoiISJF6t7FulIjgBNgSeAOf4yQAqT7GDwKDgE5A\nl/xkTURERBpT4vqciIiISNOWlBliRURERIAmEpyY2elmNt7MFprZ+2a2Vb7zlCszuyxabyj+yuxz\nc6WZTTGzBWb2qpltlLG/tZndaWY/mdlcM3vSzNZq3JKsKJc1leqjbGa2qpk9YmZzzGyWmd0X9XNq\nFNWV08yGZXnGL2Uck+hymtnFZvahmf1iZtPM7Bkz2yTLcQX9PHMpZzE8z+j+p5nZf6P7zzGzd81s\n74xjCvp5RvevspzF8jwzmdlFUVluykhPxjMNIRT1C/gDPjrnOKA7cA8wE1gj33nLMf+XAZ8BawJr\nRa/VYvv/FJVnf3wU07PAd0Cr2DF34aOUdgZ+h3dqejsBZdsbuBJfemA5cGDG/nopG/BvoAzv27Q9\n8A0wPEHlHAa8mPGM22cck+hyAi8BxwI9gN7AC1F+f1NMzzPHchb884zuv1/0s7shPmDhKmAx0KNY\nnmeO5SyK55mRl62A74FPgJti6Yl5po3+peThIbwP3BrbNuAH4MJ85y3H/F8GlFWxfwowJLbdDlgI\nHBnbXgwcEjumG74e0db5Ll8sT+Ws+Eu7zmXDf4mUA7+LHbMXPidOp4SUcxjwdBXnFGI514jys2OR\nP89s5Sy65xnLw8/AwGJ9npWUs6ieJ7Ay8DWwGz4QJR6cJOaZFnWzjpm1BPoCo1Jpwb+p14Dt8pWv\nWtjYvEngOzMbbmZdAMxsA3wkU7x8vwAfkC7flviorPgxXwP/I8HfQT2WbVtgVgjhk9jlX8NHhW3T\nUPmvhV2iZoKvzGyoma0W29eXwitnh+jeM6Gon2eFcsYU1fM0s2ZmdhSwEvBusT7PzHLGdhXT87wT\neD6E8Ho8MWnPNClDiRvKGkBzYFpG+jQ82isE7wMn4JHu2sDlwFtmtin+gxTIXr5O0eeOwJLoh6yy\nY5KovsrWCZge3xlCWG5mM0lO+f8NPAWMx6uWrwVeMrPtomC6EwVUTjMz4BbgnRBCqn9U0T3PSsoJ\nRfQ8o/9n3sMn4JqL/8X8tZltRxE9z8rKGe0upud5FLAFHmRkStS/0WIPTgpeCCE+G98XZvYhMBE4\nEvgqP7mS+hRCeCK2+aWZfY638+6CV7sWmqFAT2CHfGekgWUtZ5E9z6+AzYH2wOHAQ2a2U36z1CCy\nljOE8FWxPE8zWxcPpvcIISzNd36qU9TNOsBPeAfEjhnpHYGpjZ+dugshzME7F22El8GounxTgVZm\n1q6KY5Kovso2Fe/A9iszaw6sRkLLH0IYj//spnrJF0w5zewOYF9glxDCj7FdRfU8qyjnCgr5eYYQ\nloUQvg8hfBJCuAT4L3A2RfY8qyhntmML9Xn2xTv1lpnZUjNbindqPdvMluC1H4l5pkUdnETRYSmw\neyotqordnUachrc+mdnK+D+KKdE/kqlULF87vF0vVb5SvCNS/JhuwG/xasxEqseyvQd0MLPfxS6/\nO/6P8IOGyn9dRH/hrA6kfukVRDmjX9gHAbuGEP4X31dMz7OqclZyfEE+z0o0A1oX0/OsRDOgdbYd\nBfw8X8NHmG2B1xJtDnwMDAc2DyF8T5KeaWP2Es7HC2/+WEDFocQ/A2vmO2855v9vwE7AeviQrFfx\nCHf1aP+FUXkOiH7wngW+peLQr6F4e+kuePQ8mmQMJW4b/QPZAu/dfU603aU+y4YP//wYHz63A95/\n5+EklDPadwP+H8B60T/ij4GxQMtCKWeUv1lAP/yvqNSrTeyYgn+e1ZWzWJ5ndP9ronKuhw8rvRb/\nxbRbsTzP6spZTM+zkrJnjtZJzDPN25fSyA9gMD4ueyEe1W2Z7zzVIO8l+NDnhXiP6EeBDTKOuRwf\nArYAXzFyo4z9rYHb8arIucC/gLUSULad8V/WyzNe99dn2fARFcOBOfgvln8AKyWhnHgHvJfxv1gW\n4XMP3EVG8Jz0clZSvuXAcfX9s5rkchbL84zuf1+U/4VReUYSBSbF8jyrK2cxPc9Kyv46seAkSc9U\na+uIiIhIohR1nxMREREpPApOREREJFEUnIiIiEiiKDgRERGRRFFwIiIiIomi4EREREQSRcGJiIiI\nJIqCExEREUkUBSciIiKSKApORAqcmb1hZjfV4Pj1zKzczDaLtneOtjNXGm1wZjbMzJ5u7PvWlpld\nZmaf5DsfIsVOwYlIwpjZA1GwMDTLvjujfffHkg8B/lKDW/wP6AR8EUur8zoWNQ2SCpjW/BBpYApO\nRJIn4AHEUWb267Lt0ef+wMQKB4cwO4QwP+eLu+khhPL6yrDUjZm1yHceRJJEwYlIMn0CTAIOjaUd\nigcmFZoVMmsszGy8mV1sZv80s1/MbKKZnRzbX6FZJ2ZHM/uvmS00s/fMrFfsnNXM7FEz+8HM5pvZ\nZ2Z2VGz/MHz15bOjay83s99G+3qZ2fNmNifKz3/MbIOMMpxnZlPM7Cczu8PMmlf2xaSaVszsmKis\ns82sxMzaZnwHZ2Wc94mZXRrbLjezU6K8zTezMWa2rZltGH2n88xsdGZeo3NPMbP/Rec9bmarZOw/\nKbrewuj9j1m+/yPN7E0zWwAMqKy8Ik2RghORZArA/cCgWNogYBhgOZx/LvARsAUwFLjLzDbOuH6c\nATcAQ4AtgRnAiFiQ0Ab4GNgH6AXcAzxkZltG+88G3sOXRu8IrA1MMrN1gP/gy9HvAvwuOiZeU7Ab\n0DXafxxwQvSqyobAQcC+wH54YHRRNedk82fgAWBzYCzwKHA3cDXQF/9e7sg4Z2PgiOi+e+Fl+rUJ\nzsyOxpedvxjoDvwfcKWZHZtxnWuBm4Ee+NL0IhJRVaJIcj0CXGdmXfA/JLYH/gDsmsO5L4YQ7o4+\nX29mQ6Lzvo3SsgU4l4cQXgcws+OBH/D+LE+GEKYA8f4kd5rZ3sCRwMchhF/MbAmwIIQwI3WQmZ0B\nzAb6hxCWR8nfZdx3JnBGCCEA35jZi8DuwD+rKJ8Bx4cQFkT3eTg6pyZ9bwDuDyE8FV3jBjzAuiKE\n8FqUdiseJMa1Bo4NIUyNjjkTeNHMzgshTMcDk/NCCM9Fx0+MaqFOAx6OXefm2DEiEqPgRCShQgg/\nmdkLwED8l/GLIYSZZrlUnPB5xvZUYK2qbge8H7v3LDP7Gv+rHjNrBlyC1xh0BlpFr+r6umwOvB0L\nTLL5MgpMUn4ENq3muhNSgUnsnKrKV5n49zQtev8iI62Nma0cQpgXpf0vFZhE3sODx25mNg+v1fmn\nmd0XO6Y5HqTFldYivyJNgoITkWQbhjcrBGBwDc5bmrEdqFsz7oXAmXjzzRd4UHIrHqBUZWEO165N\nXqs7p5wVa4daVnOdUEVart/dytH7ScCHGfsyA7ScOzGLNDXqcyKSbC/jAUALYGQD3seAbX/dMFsV\n2AQYEyVtDzwXQigJIXwOjI/2xy3BawjiPgP6VdXBtYHMwPu9ABDN4bJCx9Yschkm/Fsz6xTb3g4P\nPL6KmnWmABuGEL7PeMVHWWk4skgVFJyIJFg03Lc70Cuj6aMhXGpmu5nZpngn0RlAqk/Et8DvzWw7\nM+uBd4jtmHH+BGCbaDTK6lHaHUA74HEz62tmG0WjbDamYb0OHGtmO5pZ76g8y3I4L1ubWWbaYuBB\nM9vMzPrhNUiPx/raXAZcbGZnmtnGZrapmZ1gZudUcx8RiSg4EUm4EMK8WH+HrIdUs53LMQEf7XIr\nPspnTeCAEELqF/pVQBlek/M63sfjmYxr/B2vQRgDTDez34YQZuKjcdoCb+Ijfk5ixWaZ+nYtPkro\n+ej1DCt2xM3le8qW9i3wNPAS/n18Cpz+68Eh/BMv40C85uhN4Hi8tqmq+4hIxBr+jzERERGR3Knm\nRERERBJFwYmIiIgkioITERERSRQFJyIiIpIoCk5EREQkURSciIiISKIoOBEREZFEUXAiIiIiiaLg\nRERERBJFwYmIiIgkioITERERSRQFJyIiIpIo/w8Ep12WDyJk7AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x69b0710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAArcgIArASzzhT3a0asAwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa1dff98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pred_basic_model_dropout = train_and_evaluate(reader_train, reader_test, max_epochs=5, model_func=create_basic_model_with_dropout)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Add batch normalization after each convolution and before the last dense layer:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def create_basic_model_with_batch_normalization(input, out_dims):\n",
    "\n",
    "    with default_options(activation=relu):\n",
    "        model = Sequential([\n",
    "            LayerStack(3, lambda i: [\n",
    "                Convolution((5,5), [32,32,64][i], init=glorot_uniform(), pad=True),\n",
    "                BatchNormalization(map_rank=1),\n",
    "                MaxPooling((3,3), strides=(2,2))\n",
    "            ]),\n",
    "            Dense(64, init=glorot_uniform()),\n",
    "            BatchNormalization(map_rank=1),\n",
    "            Dense(out_dims, init=glorot_uniform(), activation=None)\n",
    "        ])\n",
    "\n",
    "    return model(input)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training 117290 parameters in 18 parameter tensors.\n",
      "\n",
      "Finished Epoch [1]: [Training] loss = 1.572498 * 50000, metric = 56.4% * 50000 5.768s (8667.9 samples per second)\n",
      "Finished Epoch [2]: [Training] loss = 1.235720 * 50000, metric = 43.7% * 50000 5.817s (8595.0 samples per second)\n",
      "Finished Epoch [3]: [Training] loss = 1.096666 * 50000, metric = 38.5% * 50000 5.811s (8604.3 samples per second)\n",
      "Finished Epoch [4]: [Training] loss = 1.021178 * 50000, metric = 35.6% * 50000 5.837s (8566.3 samples per second)\n",
      "Finished Epoch [5]: [Training] loss = 0.965618 * 50000, metric = 33.6% * 50000 5.765s (8672.7 samples per second)\n",
      "\n",
      "Final Results: Minibatch[1-626]: errs = 32.3% * 10000\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQqb8f2jISJ85mklNAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa143e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nprwDdH1qolW1G7QAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa2e2710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pred_basic_model_bn = train_and_evaluate(reader_train, reader_test, max_epochs=5, model_func=create_basic_model_with_batch_normalization)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's implement an inspired VGG style network, using layer API, here the architecture:\n",
    "\n",
    "| VGG9          |\n",
    "| ------------- |\n",
    "| conv3-64      |\n",
    "| conv3-64      |\n",
    "| max3          |\n",
    "|               |\n",
    "| conv3-96      |\n",
    "| conv3-96      |\n",
    "| max3          |\n",
    "|               |\n",
    "| conv3-128     |\n",
    "| conv3-128     |\n",
    "| max3          |\n",
    "|               |\n",
    "| FC-1024       |\n",
    "| FC-1024       |\n",
    "|               |\n",
    "| FC-10         |\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def create_vgg9_model(input, out_dims):\n",
    "    with default_options(activation=relu):\n",
    "        model = Sequential([\n",
    "            LayerStack(3, lambda i: [\n",
    "                Convolution((3,3), [64,96,128][i], init=glorot_uniform(), pad=True),\n",
    "                Convolution((3,3), [64,96,128][i], init=glorot_uniform(), pad=True),\n",
    "                MaxPooling((3,3), strides=(2,2))\n",
    "            ]),\n",
    "            LayerStack(2, lambda : [\n",
    "                Dense(1024, init=glorot_uniform())\n",
    "            ]),\n",
    "            Dense(out_dims, init=glorot_uniform(), activation=None)\n",
    "        ])\n",
    "        \n",
    "    return model(input)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training 2675978 parameters in 18 parameter tensors.\n",
      "\n",
      "Finished Epoch [1]: [Training] loss = 2.258327 * 50000, metric = 83.0% * 50000 11.357s (4402.7 samples per second)\n",
      "Finished Epoch [2]: [Training] loss = 1.928672 * 50000, metric = 72.0% * 50000 11.328s (4414.0 samples per second)\n",
      "Finished Epoch [3]: [Training] loss = 1.709348 * 50000, metric = 63.5% * 50000 11.514s (4342.5 samples per second)\n",
      "Finished Epoch [4]: [Training] loss = 1.571040 * 50000, metric = 57.7% * 50000 12.057s (4147.1 samples per second)\n",
      "Finished Epoch [5]: [Training] loss = 1.467748 * 50000, metric = 53.6% * 50000 11.984s (4172.3 samples per second)\n",
      "\n",
      "Final Results: Minibatch[1-626]: errs = 49.6% * 10000\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRKSsKDgRERGRsqLgRERERMrK/wPXD88FuOeZhwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa2aa160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xa131ba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pred_vgg = train_and_evaluate(reader_train, reader_test, max_epochs=5, model_func=create_vgg9_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Residual Network (ResNet)\n",
    "\n",
    "One of the main problem of a Deep Neural Network is how to propagate the error all the way to the first layer. For a deep network, the gradient keep getting smaller until it has no effect on the network weights. [ResNet](https://arxiv.org/abs/1512.03385) was designed to overcome such problem, by defining a block with identity path, as shown below:\n",
    "\n",
    "<img src=\"https://cntk.ai/jup/201/ResNetBlock2.png\">\n",
    "\n",
    "The idea of the above block is 2 folds:\n",
    "\n",
    "* During back propagation the gradient have a path that doesn't affect its magnitude.\n",
    "* The network need to learn residual mapping (delta to x).\n",
    "\n",
    "So let's implements ResNet blocks using CNTK:\n",
    "\n",
    "            ResNetNode                   ResNetNodeInc\n",
    "                |                              |\n",
    "         +------+------+             +---------+----------+\n",
    "         |             |             |                    |\n",
    "         V             |             V                    V\n",
    "    +----------+       |      +--------------+   +----------------+\n",
    "    | Conv, BN |       |      | Conv x 2, BN |   | SubSample, BN  |\n",
    "    +----------+       |      +--------------+   +----------------+\n",
    "         |             |             |                    |\n",
    "         V             |             V                    |\n",
    "     +-------+         |         +-------+                |\n",
    "     | ReLU  |         |         | ReLU  |                |\n",
    "     +-------+         |         +-------+                |\n",
    "         |             |             |                    |\n",
    "         V             |             V                    |\n",
    "    +----------+       |        +----------+              |\n",
    "    | Conv, BN |       |        | Conv, BN |              |\n",
    "    +----------+       |        +----------+              |\n",
    "         |             |             |                    |\n",
    "         |    +---+    |             |       +---+        |\n",
    "         +--->| + |<---+             +------>+ + +<-------+\n",
    "              +---+                          +---+\n",
    "                |                              |\n",
    "                V                              V\n",
    "            +-------+                      +-------+\n",
    "            | ReLU  |                      | ReLU  |\n",
    "            +-------+                      +-------+\n",
    "                |                              |\n",
    "                V                              V\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from cntk.ops import combine, times, element_times, AVG_POOLING\n",
    "\n",
    "def convolution_bn(input, filter_size, num_filters, strides=(1,1), init=he_normal(), activation=relu):\n",
    "    if activation is None:\n",
    "        activation = lambda x: x\n",
    "        \n",
    "    r = Convolution(filter_size, num_filters, strides=strides, init=init, activation=None, pad=True, bias=False)(input)\n",
    "    r = BatchNormalization(map_rank=1)(r)\n",
    "    r = activation(r)\n",
    "    \n",
    "    return r\n",
    "\n",
    "def resnet_basic(input, num_filters):\n",
    "    c1 = convolution_bn(input, (3,3), num_filters)\n",
    "    c2 = convolution_bn(c1, (3,3), num_filters, activation=None)\n",
    "    p  = c2 + input\n",
    "    return relu(p)\n",
    "\n",
    "def resnet_basic_inc(input, num_filters):\n",
    "    c1 = convolution_bn(input, (3,3), num_filters, strides=(2,2))\n",
    "    c2 = convolution_bn(c1, (3,3), num_filters, activation=None)\n",
    "\n",
    "    s = convolution_bn(input, (1,1), num_filters, strides=(2,2), activation=None)\n",
    "    \n",
    "    p = c2 + s\n",
    "    return relu(p)\n",
    "\n",
    "def resnet_basic_stack(input, num_filters, num_stack):\n",
    "    assert (num_stack > 0)\n",
    "    \n",
    "    r = input\n",
    "    for _ in range(num_stack):\n",
    "        r = resnet_basic(r, num_filters)\n",
    "    return r"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's write the full model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def create_resnet_model(input, out_dims):\n",
    "    conv = convolution_bn(input, (3,3), 16)\n",
    "    r1_1 = resnet_basic_stack(conv, 16, 3)\n",
    "\n",
    "    r2_1 = resnet_basic_inc(r1_1, 32)\n",
    "    r2_2 = resnet_basic_stack(r2_1, 32, 2)\n",
    "\n",
    "    r3_1 = resnet_basic_inc(r2_2, 64)\n",
    "    r3_2 = resnet_basic_stack(r3_1, 64, 2)\n",
    "\n",
    "    # Global average pooling\n",
    "    pool = AveragePooling(filter_shape=(8,8), strides=(1,1))(r3_2)    \n",
    "    net = Dense(out_dims, init=he_normal(), activation=None)(pool)\n",
    "    \n",
    "    return net"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training 272474 parameters in 65 parameter tensors.\n",
      "\n",
      "Finished Epoch [1]: [Training] loss = 1.888094 * 50000, metric = 70.5% * 50000 15.690s (3186.8 samples per second)\n",
      "Finished Epoch [2]: [Training] loss = 1.545802 * 50000, metric = 57.7% * 50000 15.700s (3184.6 samples per second)\n",
      "Finished Epoch [3]: [Training] loss = 1.421820 * 50000, metric = 51.9% * 50000 15.715s (3181.6 samples per second)\n",
      "Finished Epoch [4]: [Training] loss = 1.333261 * 50000, metric = 48.4% * 50000 15.514s (3223.0 samples per second)\n",
      "Finished Epoch [5]: [Training] loss = 1.255839 * 50000, metric = 45.6% * 50000 15.436s (3239.2 samples per second)\n",
      "\n",
      "Final Results: Minibatch[1-626]: errs = 45.1% * 10000\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n4fOyOyHsygAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x6a18908>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAicAAADeCAYAAADmUqAlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xfde9b38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pred_resnet = train_and_evaluate(reader_train, reader_test, max_epochs=5, model_func=create_resnet_model)"
   ]
  }
 ],
 "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": "3.4.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}
back to top