Chandrak1907/lesson1-Copy2.ipynb Secret

## lesson1-Copy2.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Using Convolutional Neural Networks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Welcome to the first week of the first deep learning certificate! We're going to use convolutional neural networks (CNNs) to allow our computer to see - something that is only possible thanks to deep learning."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Introduction to this week's task: 'Dogs vs Cats'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We're going to try to create a model to enter the [Dogs vs Cats](https://www.kaggle.com/c/dogs-vs-cats) competition at Kaggle. There are 25,000 labelled dog and cat photos available for training, and 12,500 in the test set that we have to try to label for this competition. According to the Kaggle web-site, when this competition was launched (end of 2013): *\"**State of the art**: The current literature suggests machine classifiers can score above 80% accuracy on this task\"*. So if we can beat 80%, then we will be at the cutting edge as at 2013!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Basic setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There isn't too much to do to get started - just a few simple configuration steps.\n",
    "\n",
    "This shows plots in the web page itself - we always wants to use this when using jupyter notebook:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Chandra/anaconda/lib/python2.7/site-packages/matplotlib/font_manager.py:273: UserWarning: Matplotlib is building the font cache using fc-list. This may take a moment.\n",
      "  warnings.warn('Matplotlib is building the font cache using fc-list. This may take a moment.')\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Define path to data: (It's a good idea to put it in a subdirectory of your notebooks folder, and then exclude that directory from git control by adding it to .gitignore.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "path = \"data/dogscats/\"\n",
    "#path = \"data/dogscats/sample/\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A few basic libraries that we'll need for the initial exercises:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from __future__ import division,print_function\n",
    "\n",
    "import os, json\n",
    "from glob import glob\n",
    "import numpy as np\n",
    "np.set_printoptions(precision=4, linewidth=100)\n",
    "from matplotlib import pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have created a file most imaginatively called 'utils.py' to store any little convenience functions we'll want to use. We will discuss these as we use them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using Theano backend.\n"
     ]
    }
   ],
   "source": [
    "import utils; reload(utils)\n",
    "from utils import plots"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Use a pretrained VGG model with our **Vgg16** class"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our first step is simply to use a model that has been fully created for us, which can recognise a wide variety (1,000 categories) of images. We will use 'VGG', which won the 2014 Imagenet competition, and is a very simple model to create and understand. The VGG Imagenet team created both a larger, slower, slightly more accurate model (*VGG  19*) and a smaller, faster model (*VGG 16*). We will be using VGG 16 since the much slower performance of VGG19 is generally not worth the very minor improvement in accuracy.\n",
    "\n",
    "We have created a python class, *Vgg16*, which makes using the VGG 16 model very straightforward. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The punchline: state of the art custom model in 7 lines of code\n",
    "\n",
    "Here's everything you need to do to get >97% accuracy on the Dogs vs Cats dataset - we won't analyze how it works behind the scenes yet, since at this stage we're just going to focus on the minimum necessary to actually do useful work."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# As large as you can, but no larger than 64 is recommended. \n",
    "# If you have an older or cheaper GPU, you'll run out of memory, so will have to decrease this.\n",
    "batch_size=5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Errno 2] No such file or directory: 'z3_Projects/1_DeepLearning/FastAI/courses/deeplearning1/nbs/'\n",
      "/Users/Chandra/z3_Projects/1_DeepLearning/FastAI/courses/deeplearning1/nbs\n"
     ]
    }
   ],
   "source": [
    "% cd z3_Projects/1_DeepLearning/FastAI/courses/deeplearning1/nbs/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "u'/Users/Chandra/z3_Projects/1_DeepLearning/FastAI/courses/deeplearning1/nbs'"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "% pwd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Import our class, and instantiate\n",
    "import vgg16; reload(vgg16)\n",
    "from vgg16 import Vgg16"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "u'/Users/Chandra/z3_Projects/1_DeepLearning/FastAI/courses/deeplearning1/nbs'"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "% pwd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "% path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "path = \"/Users/Chandra/z3_Projects/1_DeepLearning/FastAI/courses/deeplearning1/nbs/MyExercises/dogscats/sample/\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The code above will work for any image recognition task, with any number of categories! All you have to do is to put your images into one folder per category, and run the code above.\n",
    "\n",
    "Let's take a look at how this works, step by step..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create a VGG model from scratch in Keras\n",
    "\n",
    "For the rest of this tutorial, we will not be using the Vgg16 class at all. Instead, we will recreate from scratch the functionality we just used. This is not necessary if all you want to do is use the existing model - but if you want to create your own models, you'll need to understand these details. It will also help you in the future when you debug any problems with your models, since you'll understand what's going on behind the scenes."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model setup\n",
    "\n",
    "We need to import all the modules we'll be using from numpy, scipy, and keras:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from numpy.random import random, permutation\n",
    "from scipy import misc, ndimage\n",
    "from scipy.ndimage.interpolation import zoom\n",
    "\n",
    "import keras\n",
    "from keras import backend as K\n",
    "from keras.utils.data_utils import get_file\n",
    "from keras.models import Sequential, Model\n",
    "from keras.layers.core import Flatten, Dense, Dropout, Lambda\n",
    "from keras.layers import Input\n",
    "from keras.layers.convolutional import Convolution2D, MaxPooling2D, ZeroPadding2D\n",
    "from keras.optimizers import SGD, RMSprop\n",
    "from keras.preprocessing import image"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's import the mappings from VGG ids to imagenet category ids and descriptions, for display purposes later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "FILES_PATH = 'http://www.platform.ai/models/'; CLASS_FILE='imagenet_class_index.json'\n",
    "# Keras' get_file() is a handy function that downloads files, and caches them for re-use later\n",
    "# fpath = get_file(CLASS_FILE, FILES_PATH+CLASS_FILE, cache_subdir='models')\n",
    "fpath = get_file(CLASS_FILE, CLASS_FILE, cache_subdir='models')\n",
    "with open(fpath) as f: class_dict = json.load(f)\n",
    "# Convert dictionary with string indexes into an array\n",
    "classes = [class_dict[str(i)][1] for i in range(len(class_dict))]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here's a few examples of the categories we just imported:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[u'tench', u'goldfish', u'great_white_shark', u'tiger_shark', u'hammerhead']"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classes[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model creation\n",
    "\n",
    "Creating the model involves creating the model architecture, and then loading the model weights into that architecture. We will start by defining the basic pieces of the VGG architecture.\n",
    "\n",
    "VGG has just one type of convolutional block, and one type of fully connected ('dense') block. Here's the convolutional block definition:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def ConvBlock(layers, model, filters):\n",
    "    for i in range(layers): \n",
    "        model.add(ZeroPadding2D((1,1)))\n",
    "        model.add(Convolution2D(filters, 3, 3, activation='relu'))\n",
    "    model.add(MaxPooling2D((2,2), strides=(2,2)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "...and here's the fully-connected definition."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def FCBlock(model):\n",
    "    model.add(Dense(4096, activation='relu'))\n",
    "    model.add(Dropout(0.5))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When the VGG model was trained in 2014, the creators subtracted the average of each of the three (R,G,B) channels first, so that the data for each channel had a mean of zero. Furthermore, their software that expected the channels to be in B,G,R order, whereas Python by default uses R,G,B. We need to preprocess our data to make these two changes, so that it is compatible with the VGG model:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Mean of each channel as provided by VGG researchers\n",
    "vgg_mean = np.array([123.68, 116.779, 103.939]).reshape((3,1,1))\n",
    "\n",
    "def vgg_preprocess(x):\n",
    "    x = x - vgg_mean     # subtract mean\n",
    "    return x[:, ::-1]    # reverse axis bgr->rgb\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we're ready to define the VGG model architecture - look at how simple it is, now that we have the basic blocks defined!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def VGG_16():\n",
    "    model = Sequential()\n",
    "    model.add(Lambda(vgg_preprocess,input_shape=(3,224,224)))\n",
    "\n",
    "    ConvBlock(2, model, 64)\n",
    "    ConvBlock(2, model, 128)\n",
    "    ConvBlock(3, model, 256)\n",
    "    ConvBlock(3, model, 512)\n",
    "    ConvBlock(3, model, 512)\n",
    "\n",
    "    model.add(Flatten())\n",
    "    FCBlock(model)\n",
    "    FCBlock(model)\n",
    "    model.add(Dense(1000, activation='softmax'))\n",
    "    return model\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We'll learn about what these different blocks do later in the course. For now, it's enough to know that:\n",
    "\n",
    "- Convolution layers are for finding patterns in images\n",
    "- Dense (fully connected) layers are for combining patterns across an image\n",
    "\n",
    "Now that we've defined the architecture, we can create the model like any python object:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/layers/core.py:621: UserWarning: `output_shape` argument not specified for layer lambda_1 and cannot be automatically inferred with the Theano backend. Defaulting to output shape `(None, 3, 224, 224)` (same as input shape). If the expected output shape is different, specify it via the `output_shape` argument.\n",
      "  .format(self.name, input_shape))\n"
     ]
    }
   ],
   "source": [
    "model = VGG_16()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As well as the architecture, we need the weights that the VGG creators trained. The weights are the part of the model that is learnt from the data, whereas the architecture is pre-defined based on the nature of the problem. \n",
    "\n",
    "Downloading pre-trained weights is much preferred to training the model ourselves, since otherwise we would have to download the entire Imagenet archive, and train the model for many days! It's very helpful when researchers release their weights, as they did here.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# fpath = get_file('vgg16.h5', FILES_PATH+'vgg16.h5', cache_subdir='models')\n",
    "# fpath = get_file('vgg16.h5','vgg16.h5', cache_subdir='models')\n",
    "model.load_weights('vgg16.h5')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Getting imagenet predictions\n",
    "\n",
    "The setup of the imagenet model is now complete, so all we have to do is grab a batch of images and call *predict()* on them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "batch_size = 4"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Keras provides functionality to create batches of data from directories containing images; all we have to do is to define the size to resize the images to, what type of labels to create, whether to randomly shuffle the images, and how many images to include in each batch. We use this little wrapper to define some helpful defaults appropriate for imagenet data:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def get_batches(dirname, gen=image.ImageDataGenerator(), shuffle=True, \n",
    "                batch_size=batch_size, class_mode='categorical'):\n",
    "    return gen.flow_from_directory(dirname, target_size=(224,224), \n",
    "                class_mode=class_mode, shuffle=shuffle, batch_size=batch_size)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From here we can use exactly the same steps as before to look at predictions from the model.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "u'/Users/Chandra/z3_Projects/1_DeepLearning/FastAI/courses/deeplearning1/nbs'"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "% pwd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "path = \"/Users/Chandra/z3_Projects/1_DeepLearning/FastAI/courses/deeplearning1/nbs/MyExercises/dogscats/sample/\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/Users/Chandra/z3_Projects/1_DeepLearning/FastAI/courses/deeplearning1/nbs/MyExercises/dogscats/sample/train'"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path+'train'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 16 images belonging to 2 classes.\n",
      "Found 8 images belonging to 2 classes.\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsMAAADACAYAAADhh27FAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11e76a750>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "batches = get_batches(path+'train', batch_size=batch_size)\n",
    "val_batches = get_batches(path+'valid', batch_size=batch_size)\n",
    "imgs,labels = next(batches)\n",
    "\n",
    "# This shows the 'ground truth'\n",
    "plots(imgs, titles=labels)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The VGG model returns 1,000 probabilities for each image, representing the probability that the model assigns to each possible imagenet category for each image. By finding the index with the largest probability (with *np.argmax()*) we can find the predicted label."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def pred_batch(imgs):\n",
    "    preds = model.predict(imgs)\n",
    "    idxs = np.argmax(preds, axis=1)\n",
    "\n",
    "    print('Shape: {}'.format(preds.shape))\n",
    "    print('First 5 classes: {}'.format(classes[:5]))\n",
    "    print('First 5 probabilities: {}\\n'.format(preds[0, :5]))\n",
    "    print('Predictions prob/class: ')\n",
    "    \n",
    "    for i in range(len(idxs)):\n",
    "        idx = idxs[i]\n",
    "        print ('  {:.4f}/{}'.format(preds[i, idx], classes[idx]))\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape: (4, 1000)\n",
      "First 5 classes: [u'tench', u'goldfish', u'great_white_shark', u'tiger_shark', u'hammerhead']\n",
      "First 5 probabilities: [  2.9194e-09   6.2409e-09   5.5053e-10   4.7807e-10   4.3210e-09]\n",
      "\n",
      "Predictions prob/class: \n",
      "  0.4821/Bouvier_des_Flandres\n",
      "  0.1044/hamster\n",
      "  0.2421/toy_terrier\n",
      "  0.1634/Rottweiler\n"
     ]
    }
   ],
   "source": [
    "pred_batch(imgs)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "____________________________________________________________________________________________________\n",
      "Layer (type)                     Output Shape          Param #     Connected to                     \n",
      "====================================================================================================\n",
      "lambda_1 (Lambda)                (None, 3, 224, 224)   0           lambda_input_1[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_1 (ZeroPadding2D)  (None, 3, 226, 226)   0           lambda_1[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_1 (Convolution2D)  (None, 64, 224, 224)  1792        zeropadding2d_1[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_2 (ZeroPadding2D)  (None, 64, 226, 226)  0           convolution2d_1[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_2 (Convolution2D)  (None, 64, 224, 224)  36928       zeropadding2d_2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_1 (MaxPooling2D)    (None, 64, 112, 112)  0           convolution2d_2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_3 (ZeroPadding2D)  (None, 64, 114, 114)  0           maxpooling2d_1[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_3 (Convolution2D)  (None, 128, 112, 112) 73856       zeropadding2d_3[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_4 (ZeroPadding2D)  (None, 128, 114, 114) 0           convolution2d_3[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_4 (Convolution2D)  (None, 128, 112, 112) 147584      zeropadding2d_4[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_2 (MaxPooling2D)    (None, 128, 56, 56)   0           convolution2d_4[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_5 (ZeroPadding2D)  (None, 128, 58, 58)   0           maxpooling2d_2[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_5 (Convolution2D)  (None, 256, 56, 56)   295168      zeropadding2d_5[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_6 (ZeroPadding2D)  (None, 256, 58, 58)   0           convolution2d_5[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_6 (Convolution2D)  (None, 256, 56, 56)   590080      zeropadding2d_6[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_7 (ZeroPadding2D)  (None, 256, 58, 58)   0           convolution2d_6[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_7 (Convolution2D)  (None, 256, 56, 56)   590080      zeropadding2d_7[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_3 (MaxPooling2D)    (None, 256, 28, 28)   0           convolution2d_7[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_8 (ZeroPadding2D)  (None, 256, 30, 30)   0           maxpooling2d_3[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_8 (Convolution2D)  (None, 512, 28, 28)   1180160     zeropadding2d_8[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_9 (ZeroPadding2D)  (None, 512, 30, 30)   0           convolution2d_8[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_9 (Convolution2D)  (None, 512, 28, 28)   2359808     zeropadding2d_9[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_10 (ZeroPadding2D) (None, 512, 30, 30)   0           convolution2d_9[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_10 (Convolution2D) (None, 512, 28, 28)   2359808     zeropadding2d_10[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_4 (MaxPooling2D)    (None, 512, 14, 14)   0           convolution2d_10[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_11 (ZeroPadding2D) (None, 512, 16, 16)   0           maxpooling2d_4[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_11 (Convolution2D) (None, 512, 14, 14)   2359808     zeropadding2d_11[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_12 (ZeroPadding2D) (None, 512, 16, 16)   0           convolution2d_11[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_12 (Convolution2D) (None, 512, 14, 14)   2359808     zeropadding2d_12[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_13 (ZeroPadding2D) (None, 512, 16, 16)   0           convolution2d_12[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_13 (Convolution2D) (None, 512, 14, 14)   2359808     zeropadding2d_13[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_5 (MaxPooling2D)    (None, 512, 7, 7)     0           convolution2d_13[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "flatten_1 (Flatten)              (None, 25088)         0           maxpooling2d_5[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "dense_1 (Dense)                  (None, 4096)          102764544   flatten_1[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "dropout_1 (Dropout)              (None, 4096)          0           dense_1[0][0]                    \n",
      "____________________________________________________________________________________________________\n",
      "dense_2 (Dense)                  (None, 4096)          16781312    dropout_1[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "dropout_2 (Dropout)              (None, 4096)          0           dense_2[0][0]                    \n",
      "____________________________________________________________________________________________________\n",
      "dense_3 (Dense)                  (None, 1000)          4097000     dropout_2[0][0]                  \n",
      "====================================================================================================\n",
      "Total params: 138,357,544\n",
      "Trainable params: 138,357,544\n",
      "Non-trainable params: 0\n",
      "____________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##  Fine Tuning the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/layers/core.py:621: UserWarning: `output_shape` argument not specified for layer lambda_3 and cannot be automatically inferred with the Theano backend. Defaulting to output shape `(None, 3, 224, 224)` (same as input shape). If the expected output shape is different, specify it via the `output_shape` argument.\n",
      "  .format(self.name, input_shape))\n"
     ]
    }
   ],
   "source": [
    "from keras.optimizers import SGD, RMSprop, Adam\n",
    "\n",
    "model = VGG_16()\n",
    "model.pop()\n",
    "for layer in model.layers: layer.trainable=False\n",
    "model.add(Dense(batches.nb_class, activation='softmax'))    \n",
    "lr=0.001\n",
    "model.compile(optimizer=Adam(lr=lr),loss='categorical_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1\n",
      "16/16 [==============================] - 11s - loss: 0.7663 - acc: 0.6250 - val_loss: 0.6978 - val_acc: 0.5000\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x1205ae6d0>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nb_epoch=1\n",
    "model.fit_generator(batches, samples_per_epoch=batches.nb_sample, nb_epoch=nb_epoch,validation_data=val_batches, nb_val_samples=val_batches.nb_sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "u'/Users/Chandra/z3_Projects/1_DeepLearning/FastAI/courses/deeplearning1/nbs'"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "% pwd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "____________________________________________________________________________________________________\n",
      "Layer (type)                     Output Shape          Param #     Connected to                     \n",
      "====================================================================================================\n",
      "lambda_3 (Lambda)                (None, 3, 224, 224)   0           lambda_input_3[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_27 (ZeroPadding2D) (None, 3, 226, 226)   0           lambda_3[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_27 (Convolution2D) (None, 64, 224, 224)  1792        zeropadding2d_27[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_28 (ZeroPadding2D) (None, 64, 226, 226)  0           convolution2d_27[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_28 (Convolution2D) (None, 64, 224, 224)  36928       zeropadding2d_28[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_11 (MaxPooling2D)   (None, 64, 112, 112)  0           convolution2d_28[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_29 (ZeroPadding2D) (None, 64, 114, 114)  0           maxpooling2d_11[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_29 (Convolution2D) (None, 128, 112, 112) 73856       zeropadding2d_29[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_30 (ZeroPadding2D) (None, 128, 114, 114) 0           convolution2d_29[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_30 (Convolution2D) (None, 128, 112, 112) 147584      zeropadding2d_30[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_12 (MaxPooling2D)   (None, 128, 56, 56)   0           convolution2d_30[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_31 (ZeroPadding2D) (None, 128, 58, 58)   0           maxpooling2d_12[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_31 (Convolution2D) (None, 256, 56, 56)   295168      zeropadding2d_31[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_32 (ZeroPadding2D) (None, 256, 58, 58)   0           convolution2d_31[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_32 (Convolution2D) (None, 256, 56, 56)   590080      zeropadding2d_32[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_33 (ZeroPadding2D) (None, 256, 58, 58)   0           convolution2d_32[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_33 (Convolution2D) (None, 256, 56, 56)   590080      zeropadding2d_33[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_13 (MaxPooling2D)   (None, 256, 28, 28)   0           convolution2d_33[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_34 (ZeroPadding2D) (None, 256, 30, 30)   0           maxpooling2d_13[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_34 (Convolution2D) (None, 512, 28, 28)   1180160     zeropadding2d_34[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_35 (ZeroPadding2D) (None, 512, 30, 30)   0           convolution2d_34[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_35 (Convolution2D) (None, 512, 28, 28)   2359808     zeropadding2d_35[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_36 (ZeroPadding2D) (None, 512, 30, 30)   0           convolution2d_35[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_36 (Convolution2D) (None, 512, 28, 28)   2359808     zeropadding2d_36[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_14 (MaxPooling2D)   (None, 512, 14, 14)   0           convolution2d_36[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_37 (ZeroPadding2D) (None, 512, 16, 16)   0           maxpooling2d_14[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_37 (Convolution2D) (None, 512, 14, 14)   2359808     zeropadding2d_37[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_38 (ZeroPadding2D) (None, 512, 16, 16)   0           convolution2d_37[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_38 (Convolution2D) (None, 512, 14, 14)   2359808     zeropadding2d_38[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_39 (ZeroPadding2D) (None, 512, 16, 16)   0           convolution2d_38[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_39 (Convolution2D) (None, 512, 14, 14)   2359808     zeropadding2d_39[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_15 (MaxPooling2D)   (None, 512, 7, 7)     0           convolution2d_39[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "flatten_3 (Flatten)              (None, 25088)         0           maxpooling2d_15[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "dense_8 (Dense)                  (None, 4096)          102764544   flatten_3[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "dropout_5 (Dropout)              (None, 4096)          0           dense_8[0][0]                    \n",
      "____________________________________________________________________________________________________\n",
      "dense_9 (Dense)                  (None, 4096)          16781312    dropout_5[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "dropout_6 (Dropout)              (None, 4096)          0           dense_9[0][0]                    \n",
      "____________________________________________________________________________________________________\n",
      "dense_11 (Dense)                 (None, 2)             8194        dropout_6[0][0]                  \n",
      "====================================================================================================\n",
      "Total params: 134,268,738\n",
      "Trainable params: 8,194\n",
      "Non-trainable params: 134,260,544\n",
      "____________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from keras.models import load_model\n",
    "model.save('my_model_copy2.h5')  # creates a HDF5 file 'my_model.h5'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "model.save_weights('my_model_weights_copy2.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "vgg_mean = np.array([123.68, 116.779, 103.939]).reshape((3,1,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "global name 'vgg_mean' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-40-1280da7d7331>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mnew_model\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mload_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'my_model_copy2.h5'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mcustom_objects\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m'vgg_mean'\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mvgg_mean\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/models.pyc\u001b[0m in \u001b[0;36mload_model\u001b[0;34m(filepath, custom_objects)\u001b[0m\n\u001b[1;32m    140\u001b[0m         \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'No model found in config file.'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    141\u001b[0m     \u001b[0mmodel_config\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel_config\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdecode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'utf-8'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 142\u001b[0;31m     \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel_from_config\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel_config\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcustom_objects\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcustom_objects\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    143\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    144\u001b[0m     \u001b[0;31m# set weights\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/models.pyc\u001b[0m in \u001b[0;36mmodel_from_config\u001b[0;34m(config, custom_objects)\u001b[0m\n\u001b[1;32m    191\u001b[0m                         \u001b[0;34m'Maybe you meant to use '\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    192\u001b[0m                         '`Sequential.from_config(config)`?')\n\u001b[0;32m--> 193\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mlayer_from_config\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcustom_objects\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcustom_objects\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    194\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    195\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/utils/layer_utils.pyc\u001b[0m in \u001b[0;36mlayer_from_config\u001b[0;34m(config, custom_objects)\u001b[0m\n\u001b[1;32m     41\u001b[0m                                        custom_objects=custom_objects)\n\u001b[1;32m     42\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 43\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mlayer_class\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfrom_config\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'config'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     44\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     45\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/models.pyc\u001b[0m in \u001b[0;36mfrom_config\u001b[0;34m(cls, config, layer_cache)\u001b[0m\n\u001b[1;32m   1084\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1085\u001b[0m             \u001b[0mlayer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mget_or_create_layer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfirst_layer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1086\u001b[0;31m             \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlayer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1087\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1088\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mconf\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/models.pyc\u001b[0m in \u001b[0;36madd\u001b[0;34m(self, layer)\u001b[0m\n\u001b[1;32m    297\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    298\u001b[0m                     \u001b[0minput_dtype\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 299\u001b[0;31m                 \u001b[0mlayer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcreate_input_layer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch_input_shape\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput_dtype\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    300\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    301\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlayer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minbound_nodes\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/engine/topology.pyc\u001b[0m in \u001b[0;36mcreate_input_layer\u001b[0;34m(self, batch_input_shape, input_dtype, name)\u001b[0m\n\u001b[1;32m    399\u001b[0m         \u001b[0;31m# and create the node connecting the current layer\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    400\u001b[0m         \u001b[0;31m# to the input layer we just created.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 401\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    402\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    403\u001b[0m     def add_weight(self, shape, initializer, name=None,\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/engine/topology.pyc\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, x, mask)\u001b[0m\n\u001b[1;32m    570\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0minbound_layers\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    571\u001b[0m             \u001b[0;31m# This will call layer.build() if necessary.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 572\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd_inbound_node\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minbound_layers\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_indices\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtensor_indices\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    573\u001b[0m             \u001b[0;31m# Outputs were already computed when calling self.add_inbound_node.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    574\u001b[0m             \u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minbound_nodes\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput_tensors\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/engine/topology.pyc\u001b[0m in \u001b[0;36madd_inbound_node\u001b[0;34m(self, inbound_layers, node_indices, tensor_indices)\u001b[0m\n\u001b[1;32m    633\u001b[0m         \u001b[0;31m# creating the node automatically updates self.inbound_nodes\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    634\u001b[0m         \u001b[0;31m# as well as outbound_nodes on inbound layers.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 635\u001b[0;31m         \u001b[0mNode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcreate_node\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minbound_layers\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_indices\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtensor_indices\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    636\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    637\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mget_output_shape_for\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput_shape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/engine/topology.pyc\u001b[0m in \u001b[0;36mcreate_node\u001b[0;34m(cls, outbound_layer, inbound_layers, node_indices, tensor_indices)\u001b[0m\n\u001b[1;32m    164\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    165\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_tensors\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 166\u001b[0;31m             \u001b[0moutput_tensors\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mto_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutbound_layer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_tensors\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minput_masks\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    167\u001b[0m             \u001b[0moutput_masks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mto_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutbound_layer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcompute_mask\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_tensors\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput_masks\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    168\u001b[0m             \u001b[0;31m# TODO: try to auto-infer shape\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/layers/core.pyc\u001b[0m in \u001b[0;36mcall\u001b[0;34m(self, x, mask)\u001b[0m\n\u001b[1;32m    638\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0;34m'mask'\u001b[0m \u001b[0;32min\u001b[0m \u001b[0marg_spec\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    639\u001b[0m             \u001b[0marguments\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'mask'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmask\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 640\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfunction\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0marguments\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    641\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    642\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mget_config\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/layers/core.pyc\u001b[0m in \u001b[0;36mvgg_preprocess\u001b[0;34m(x)\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mvgg_preprocess\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m     \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mvgg_mean\u001b[0m     \u001b[0;31m# subtract mean\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      6\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m    \u001b[0;31m# reverse axis bgr->rgb\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNameError\u001b[0m: global name 'vgg_mean' is not defined"
     ]
    }
   ],
   "source": [
    "new_model = load_model('my_model_copy2.h5',custom_objects={'vgg_mean': vgg_mean})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved model to disk\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# serialize model to JSON\n",
    "model_json = model.to_json()\n",
    "with open(\"model.json\", \"w\") as json_file:\n",
    "    json_file.write(model_json)\n",
    "# serialize weights to HDF5\n",
    "model.save_weights(\"model.h5\")\n",
    "print(\"Saved model to disk\")\n",
    " "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "global name 'vgg_mean' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-49-c15fe059e350>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mloaded_model_json\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson_file\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mjson_file\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m \u001b[0mloaded_model\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel_from_json\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloaded_model_json\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mcustom_objects\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m'vgg_mean'\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mvgg_mean\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/models.pyc\u001b[0m in \u001b[0;36mmodel_from_json\u001b[0;34m(json_string, custom_objects)\u001b[0m\n\u001b[1;32m    211\u001b[0m     \u001b[0;32mfrom\u001b[0m \u001b[0mkeras\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mutils\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlayer_utils\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mlayer_from_config\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    212\u001b[0m     \u001b[0mconfig\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloads\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mjson_string\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 213\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mlayer_from_config\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcustom_objects\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcustom_objects\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    214\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    215\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/utils/layer_utils.pyc\u001b[0m in \u001b[0;36mlayer_from_config\u001b[0;34m(config, custom_objects)\u001b[0m\n\u001b[1;32m     41\u001b[0m                                        custom_objects=custom_objects)\n\u001b[1;32m     42\u001b[0m     \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 43\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mlayer_class\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfrom_config\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'config'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     44\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     45\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/models.pyc\u001b[0m in \u001b[0;36mfrom_config\u001b[0;34m(cls, config, layer_cache)\u001b[0m\n\u001b[1;32m   1084\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1085\u001b[0m             \u001b[0mlayer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mget_or_create_layer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfirst_layer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1086\u001b[0;31m             \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlayer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1087\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1088\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mconf\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mconfig\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/models.pyc\u001b[0m in \u001b[0;36madd\u001b[0;34m(self, layer)\u001b[0m\n\u001b[1;32m    297\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    298\u001b[0m                     \u001b[0minput_dtype\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 299\u001b[0;31m                 \u001b[0mlayer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcreate_input_layer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbatch_input_shape\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput_dtype\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    300\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    301\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlayer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minbound_nodes\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/engine/topology.pyc\u001b[0m in \u001b[0;36mcreate_input_layer\u001b[0;34m(self, batch_input_shape, input_dtype, name)\u001b[0m\n\u001b[1;32m    399\u001b[0m         \u001b[0;31m# and create the node connecting the current layer\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    400\u001b[0m         \u001b[0;31m# to the input layer we just created.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 401\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    402\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    403\u001b[0m     def add_weight(self, shape, initializer, name=None,\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/engine/topology.pyc\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, x, mask)\u001b[0m\n\u001b[1;32m    570\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0minbound_layers\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    571\u001b[0m             \u001b[0;31m# This will call layer.build() if necessary.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 572\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd_inbound_node\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minbound_layers\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_indices\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtensor_indices\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    573\u001b[0m             \u001b[0;31m# Outputs were already computed when calling self.add_inbound_node.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    574\u001b[0m             \u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minbound_nodes\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutput_tensors\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/engine/topology.pyc\u001b[0m in \u001b[0;36madd_inbound_node\u001b[0;34m(self, inbound_layers, node_indices, tensor_indices)\u001b[0m\n\u001b[1;32m    633\u001b[0m         \u001b[0;31m# creating the node automatically updates self.inbound_nodes\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    634\u001b[0m         \u001b[0;31m# as well as outbound_nodes on inbound layers.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 635\u001b[0;31m         \u001b[0mNode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcreate_node\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minbound_layers\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode_indices\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtensor_indices\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    636\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    637\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mget_output_shape_for\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput_shape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/engine/topology.pyc\u001b[0m in \u001b[0;36mcreate_node\u001b[0;34m(cls, outbound_layer, inbound_layers, node_indices, tensor_indices)\u001b[0m\n\u001b[1;32m    164\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    165\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_tensors\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 166\u001b[0;31m             \u001b[0moutput_tensors\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mto_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutbound_layer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_tensors\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmask\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minput_masks\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    167\u001b[0m             \u001b[0moutput_masks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mto_list\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutbound_layer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcompute_mask\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput_tensors\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput_masks\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    168\u001b[0m             \u001b[0;31m# TODO: try to auto-infer shape\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/layers/core.pyc\u001b[0m in \u001b[0;36mcall\u001b[0;34m(self, x, mask)\u001b[0m\n\u001b[1;32m    638\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0;34m'mask'\u001b[0m \u001b[0;32min\u001b[0m \u001b[0marg_spec\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    639\u001b[0m             \u001b[0marguments\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'mask'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmask\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 640\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfunction\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0marguments\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    641\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    642\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mget_config\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/layers/core.pyc\u001b[0m in \u001b[0;36mvgg_preprocess\u001b[0;34m(x)\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mvgg_preprocess\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m     \u001b[0mx\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mvgg_mean\u001b[0m     \u001b[0;31m# subtract mean\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      6\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m:\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m    \u001b[0;31m# reverse axis bgr->rgb\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNameError\u001b[0m: global name 'vgg_mean' is not defined"
     ]
    }
   ],
   "source": [
    "# load json and create model\n",
    "from keras.models import model_from_json\n",
    "json_file = open('model.json', 'r')\n",
    "loaded_model_json = json_file.read()\n",
    "json_file.close()\n",
    "loaded_model = model_from_json(loaded_model_json,custom_objects={'vgg_mean': vgg_mean})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/Chandra/anaconda/lib/python2.7/site-packages/keras/layers/core.py:621: UserWarning: `output_shape` argument not specified for layer lambda_2 and cannot be automatically inferred with the Theano backend. Defaulting to output shape `(None, 3, 224, 224)` (same as input shape). If the expected output shape is different, specify it via the `output_shape` argument.\n",
      "  .format(self.name, input_shape))\n"
     ]
    }
   ],
   "source": [
    "model = VGG_16()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model.load_weights(\"model.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "____________________________________________________________________________________________________\n",
      "Layer (type)                     Output Shape          Param #     Connected to                     \n",
      "====================================================================================================\n",
      "lambda_2 (Lambda)                (None, 3, 224, 224)   0           lambda_input_4[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_14 (ZeroPadding2D) (None, 3, 226, 226)   0           lambda_2[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_14 (Convolution2D) (None, 64, 224, 224)  1792        zeropadding2d_14[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_15 (ZeroPadding2D) (None, 64, 226, 226)  0           convolution2d_14[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_15 (Convolution2D) (None, 64, 224, 224)  36928       zeropadding2d_15[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_6 (MaxPooling2D)    (None, 64, 112, 112)  0           convolution2d_15[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_16 (ZeroPadding2D) (None, 64, 114, 114)  0           maxpooling2d_6[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_16 (Convolution2D) (None, 128, 112, 112) 73856       zeropadding2d_16[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_17 (ZeroPadding2D) (None, 128, 114, 114) 0           convolution2d_16[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_17 (Convolution2D) (None, 128, 112, 112) 147584      zeropadding2d_17[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_7 (MaxPooling2D)    (None, 128, 56, 56)   0           convolution2d_17[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_18 (ZeroPadding2D) (None, 128, 58, 58)   0           maxpooling2d_7[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_18 (Convolution2D) (None, 256, 56, 56)   295168      zeropadding2d_18[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_19 (ZeroPadding2D) (None, 256, 58, 58)   0           convolution2d_18[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_19 (Convolution2D) (None, 256, 56, 56)   590080      zeropadding2d_19[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_20 (ZeroPadding2D) (None, 256, 58, 58)   0           convolution2d_19[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_20 (Convolution2D) (None, 256, 56, 56)   590080      zeropadding2d_20[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_8 (MaxPooling2D)    (None, 256, 28, 28)   0           convolution2d_20[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_21 (ZeroPadding2D) (None, 256, 30, 30)   0           maxpooling2d_8[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_21 (Convolution2D) (None, 512, 28, 28)   1180160     zeropadding2d_21[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_22 (ZeroPadding2D) (None, 512, 30, 30)   0           convolution2d_21[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_22 (Convolution2D) (None, 512, 28, 28)   2359808     zeropadding2d_22[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_23 (ZeroPadding2D) (None, 512, 30, 30)   0           convolution2d_22[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_23 (Convolution2D) (None, 512, 28, 28)   2359808     zeropadding2d_23[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_9 (MaxPooling2D)    (None, 512, 14, 14)   0           convolution2d_23[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_24 (ZeroPadding2D) (None, 512, 16, 16)   0           maxpooling2d_9[0][0]             \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_24 (Convolution2D) (None, 512, 14, 14)   2359808     zeropadding2d_24[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_25 (ZeroPadding2D) (None, 512, 16, 16)   0           convolution2d_24[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_25 (Convolution2D) (None, 512, 14, 14)   2359808     zeropadding2d_25[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "zeropadding2d_26 (ZeroPadding2D) (None, 512, 16, 16)   0           convolution2d_25[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "convolution2d_26 (Convolution2D) (None, 512, 14, 14)   2359808     zeropadding2d_26[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "maxpooling2d_10 (MaxPooling2D)   (None, 512, 7, 7)     0           convolution2d_26[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "flatten_2 (Flatten)              (None, 25088)         0           maxpooling2d_10[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "dense_5 (Dense)                  (None, 4096)          102764544   flatten_2[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "dropout_3 (Dropout)              (None, 4096)          0           dense_5[0][0]                    \n",
      "____________________________________________________________________________________________________\n",
      "dense_6 (Dense)                  (None, 4096)          16781312    dropout_3[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "dropout_4 (Dropout)              (None, 4096)          0           dense_6[0][0]                    \n",
      "____________________________________________________________________________________________________\n",
      "dense_7 (Dense)                  (None, 1000)          8194        dropout_4[0][0]                  \n",
      "====================================================================================================\n",
      "Total params: 134,268,738\n",
      "Trainable params: 134,268,738\n",
      "Non-trainable params: 0\n",
      "____________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [default]",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  },
  "nav_menu": {},
  "nbpresent": {
   "slides": {
    "28b43202-5690-4169-9aca-6b9dabfeb3ec": {
     "id": "28b43202-5690-4169-9aca-6b9dabfeb3ec",
     "prev": null,
     "regions": {
      "3bba644a-cf4d-4a49-9fbd-e2554428cf9f": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "f3d3a388-7e2a-4151-9b50-c20498fceacc",
        "part": "whole"
       },
       "id": "3bba644a-cf4d-4a49-9fbd-e2554428cf9f"
      }
     }
    },
    "8104def2-4b68-44a0-8f1b-b03bf3b2a079": {
     "id": "8104def2-4b68-44a0-8f1b-b03bf3b2a079",
     "prev": "28b43202-5690-4169-9aca-6b9dabfeb3ec",
     "regions": {
      "7dded777-1ddf-4100-99ae-25cf1c15b575": {
       "attrs": {
        "height": 0.8,
        "width": 0.8,
        "x": 0.1,
        "y": 0.1
       },
       "content": {
        "cell": "fe47bd48-3414-4657-92e7-8b8d6cb0df00",
        "part": "whole"
       },
       "id": "7dded777-1ddf-4100-99ae-25cf1c15b575"
      }
     }
    }
   },
   "themes": {}
  },
  "toc": {
   "navigate_menu": true,
   "number_sections": true,
   "sideBar": true,
   "threshold": 6,
   "toc_cell": false,
   "toc_section_display": "block",
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}