script.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Inception"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.applications.vgg16 import VGG16, preprocess_input\n",
    "# from keras.applications.vgg19 import VGG19, preprocess_input\n",
    "from keras.applications.xception import Xception, preprocess_input\n",
    "# from keras.applications.resnet50 import ResNet50, preprocess_input\n",
    "from keras.applications.inception_v3 import InceptionV3, preprocess_input\n",
    "from keras import applications\n",
    "from keras.preprocessing import image\n",
    "from keras.models import Model\n",
    "from keras.models import model_from_json\n",
    "\n",
    "# other imports\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "import numpy as np\n",
    "import glob\n",
    "import h5py\n",
    "import os\n",
    "import json\n",
    "import pickle as cPickle\n",
    "import datetime\n",
    "import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "# organize imports\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "img_width, img_height = 128, 128\n",
    "train_data_dir = \"data/MSTAR_Chips/TRAIN_images\"\n",
    "validation_data_dir = \"data/MSTAR_Chips/TEST_images\"\n",
    "nb_train_samples = 1622\n",
    "nb_validation_samples = 1365 \n",
    "batch_size = 32\n",
    "epochs = 50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "img_width, img_height = 224,224"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model = applications.InceptionV3(weights='imagenet',include_top=False, input_shape = (img_width, img_height, 3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "____________________________________________________________________________________________________\n",
      "Layer (type)                     Output Shape          Param #     Connected to                     \n",
      "====================================================================================================\n",
      "input_1 (InputLayer)             (None, 224, 224, 3)   0                                            \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_1 (Conv2D)                (None, 111, 111, 32)  864         input_1[0][0]                    \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_1 (BatchNorm (None, 111, 111, 32)  96          conv2d_1[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "activation_1 (Activation)        (None, 111, 111, 32)  0           batch_normalization_1[0][0]      \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_2 (Conv2D)                (None, 109, 109, 32)  9216        activation_1[0][0]               \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_2 (BatchNorm (None, 109, 109, 32)  96          conv2d_2[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "activation_2 (Activation)        (None, 109, 109, 32)  0           batch_normalization_2[0][0]      \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_3 (Conv2D)                (None, 109, 109, 64)  18432       activation_2[0][0]               \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_3 (BatchNorm (None, 109, 109, 64)  192         conv2d_3[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "activation_3 (Activation)        (None, 109, 109, 64)  0           batch_normalization_3[0][0]      \n",
      "____________________________________________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2D)   (None, 54, 54, 64)    0           activation_3[0][0]               \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_4 (Conv2D)                (None, 54, 54, 80)    5120        max_pooling2d_1[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_4 (BatchNorm (None, 54, 54, 80)    240         conv2d_4[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "activation_4 (Activation)        (None, 54, 54, 80)    0           batch_normalization_4[0][0]      \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_5 (Conv2D)                (None, 52, 52, 192)   138240      activation_4[0][0]               \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_5 (BatchNorm (None, 52, 52, 192)   576         conv2d_5[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "activation_5 (Activation)        (None, 52, 52, 192)   0           batch_normalization_5[0][0]      \n",
      "____________________________________________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2D)   (None, 25, 25, 192)   0           activation_5[0][0]               \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_9 (Conv2D)                (None, 25, 25, 64)    12288       max_pooling2d_2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_9 (BatchNorm (None, 25, 25, 64)    192         conv2d_9[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "activation_9 (Activation)        (None, 25, 25, 64)    0           batch_normalization_9[0][0]      \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_7 (Conv2D)                (None, 25, 25, 48)    9216        max_pooling2d_2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_10 (Conv2D)               (None, 25, 25, 96)    55296       activation_9[0][0]               \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_7 (BatchNorm (None, 25, 25, 48)    144         conv2d_7[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_10 (BatchNor (None, 25, 25, 96)    288         conv2d_10[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_7 (Activation)        (None, 25, 25, 48)    0           batch_normalization_7[0][0]      \n",
      "____________________________________________________________________________________________________\n",
      "activation_10 (Activation)       (None, 25, 25, 96)    0           batch_normalization_10[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "average_pooling2d_1 (AveragePool (None, 25, 25, 192)   0           max_pooling2d_2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_6 (Conv2D)                (None, 25, 25, 64)    12288       max_pooling2d_2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_8 (Conv2D)                (None, 25, 25, 64)    76800       activation_7[0][0]               \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_11 (Conv2D)               (None, 25, 25, 96)    82944       activation_10[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_12 (Conv2D)               (None, 25, 25, 32)    6144        average_pooling2d_1[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_6 (BatchNorm (None, 25, 25, 64)    192         conv2d_6[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_8 (BatchNorm (None, 25, 25, 64)    192         conv2d_8[0][0]                   \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_11 (BatchNor (None, 25, 25, 96)    288         conv2d_11[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_12 (BatchNor (None, 25, 25, 32)    96          conv2d_12[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_6 (Activation)        (None, 25, 25, 64)    0           batch_normalization_6[0][0]      \n",
      "____________________________________________________________________________________________________\n",
      "activation_8 (Activation)        (None, 25, 25, 64)    0           batch_normalization_8[0][0]      \n",
      "____________________________________________________________________________________________________\n",
      "activation_11 (Activation)       (None, 25, 25, 96)    0           batch_normalization_11[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_12 (Activation)       (None, 25, 25, 32)    0           batch_normalization_12[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "mixed0 (Concatenate)             (None, 25, 25, 256)   0           activation_6[0][0]               \n",
      "                                                                   activation_8[0][0]               \n",
      "                                                                   activation_11[0][0]              \n",
      "                                                                   activation_12[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_16 (Conv2D)               (None, 25, 25, 64)    16384       mixed0[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_16 (BatchNor (None, 25, 25, 64)    192         conv2d_16[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_16 (Activation)       (None, 25, 25, 64)    0           batch_normalization_16[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_14 (Conv2D)               (None, 25, 25, 48)    12288       mixed0[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_17 (Conv2D)               (None, 25, 25, 96)    55296       activation_16[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_14 (BatchNor (None, 25, 25, 48)    144         conv2d_14[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_17 (BatchNor (None, 25, 25, 96)    288         conv2d_17[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_14 (Activation)       (None, 25, 25, 48)    0           batch_normalization_14[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_17 (Activation)       (None, 25, 25, 96)    0           batch_normalization_17[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "average_pooling2d_2 (AveragePool (None, 25, 25, 256)   0           mixed0[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_13 (Conv2D)               (None, 25, 25, 64)    16384       mixed0[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_15 (Conv2D)               (None, 25, 25, 64)    76800       activation_14[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_18 (Conv2D)               (None, 25, 25, 96)    82944       activation_17[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_19 (Conv2D)               (None, 25, 25, 64)    16384       average_pooling2d_2[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_13 (BatchNor (None, 25, 25, 64)    192         conv2d_13[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_15 (BatchNor (None, 25, 25, 64)    192         conv2d_15[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_18 (BatchNor (None, 25, 25, 96)    288         conv2d_18[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_19 (BatchNor (None, 25, 25, 64)    192         conv2d_19[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_13 (Activation)       (None, 25, 25, 64)    0           batch_normalization_13[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_15 (Activation)       (None, 25, 25, 64)    0           batch_normalization_15[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_18 (Activation)       (None, 25, 25, 96)    0           batch_normalization_18[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_19 (Activation)       (None, 25, 25, 64)    0           batch_normalization_19[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "mixed1 (Concatenate)             (None, 25, 25, 288)   0           activation_13[0][0]              \n",
      "                                                                   activation_15[0][0]              \n",
      "                                                                   activation_18[0][0]              \n",
      "                                                                   activation_19[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_23 (Conv2D)               (None, 25, 25, 64)    18432       mixed1[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_23 (BatchNor (None, 25, 25, 64)    192         conv2d_23[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_23 (Activation)       (None, 25, 25, 64)    0           batch_normalization_23[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_21 (Conv2D)               (None, 25, 25, 48)    13824       mixed1[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_24 (Conv2D)               (None, 25, 25, 96)    55296       activation_23[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_21 (BatchNor (None, 25, 25, 48)    144         conv2d_21[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_24 (BatchNor (None, 25, 25, 96)    288         conv2d_24[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_21 (Activation)       (None, 25, 25, 48)    0           batch_normalization_21[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_24 (Activation)       (None, 25, 25, 96)    0           batch_normalization_24[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "average_pooling2d_3 (AveragePool (None, 25, 25, 288)   0           mixed1[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_20 (Conv2D)               (None, 25, 25, 64)    18432       mixed1[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_22 (Conv2D)               (None, 25, 25, 64)    76800       activation_21[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_25 (Conv2D)               (None, 25, 25, 96)    82944       activation_24[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_26 (Conv2D)               (None, 25, 25, 64)    18432       average_pooling2d_3[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_20 (BatchNor (None, 25, 25, 64)    192         conv2d_20[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_22 (BatchNor (None, 25, 25, 64)    192         conv2d_22[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_25 (BatchNor (None, 25, 25, 96)    288         conv2d_25[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_26 (BatchNor (None, 25, 25, 64)    192         conv2d_26[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_20 (Activation)       (None, 25, 25, 64)    0           batch_normalization_20[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_22 (Activation)       (None, 25, 25, 64)    0           batch_normalization_22[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_25 (Activation)       (None, 25, 25, 96)    0           batch_normalization_25[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_26 (Activation)       (None, 25, 25, 64)    0           batch_normalization_26[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "mixed2 (Concatenate)             (None, 25, 25, 288)   0           activation_20[0][0]              \n",
      "                                                                   activation_22[0][0]              \n",
      "                                                                   activation_25[0][0]              \n",
      "                                                                   activation_26[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_28 (Conv2D)               (None, 25, 25, 64)    18432       mixed2[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_28 (BatchNor (None, 25, 25, 64)    192         conv2d_28[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_28 (Activation)       (None, 25, 25, 64)    0           batch_normalization_28[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_29 (Conv2D)               (None, 25, 25, 96)    55296       activation_28[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_29 (BatchNor (None, 25, 25, 96)    288         conv2d_29[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_29 (Activation)       (None, 25, 25, 96)    0           batch_normalization_29[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_27 (Conv2D)               (None, 12, 12, 384)   995328      mixed2[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_30 (Conv2D)               (None, 12, 12, 96)    82944       activation_29[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_27 (BatchNor (None, 12, 12, 384)   1152        conv2d_27[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_30 (BatchNor (None, 12, 12, 96)    288         conv2d_30[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_27 (Activation)       (None, 12, 12, 384)   0           batch_normalization_27[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_30 (Activation)       (None, 12, 12, 96)    0           batch_normalization_30[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "max_pooling2d_3 (MaxPooling2D)   (None, 12, 12, 288)   0           mixed2[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "mixed3 (Concatenate)             (None, 12, 12, 768)   0           activation_27[0][0]              \n",
      "                                                                   activation_30[0][0]              \n",
      "                                                                   max_pooling2d_3[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_35 (Conv2D)               (None, 12, 12, 128)   98304       mixed3[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_35 (BatchNor (None, 12, 12, 128)   384         conv2d_35[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_35 (Activation)       (None, 12, 12, 128)   0           batch_normalization_35[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_36 (Conv2D)               (None, 12, 12, 128)   114688      activation_35[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_36 (BatchNor (None, 12, 12, 128)   384         conv2d_36[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_36 (Activation)       (None, 12, 12, 128)   0           batch_normalization_36[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_32 (Conv2D)               (None, 12, 12, 128)   98304       mixed3[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_37 (Conv2D)               (None, 12, 12, 128)   114688      activation_36[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_32 (BatchNor (None, 12, 12, 128)   384         conv2d_32[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_37 (BatchNor (None, 12, 12, 128)   384         conv2d_37[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_32 (Activation)       (None, 12, 12, 128)   0           batch_normalization_32[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_37 (Activation)       (None, 12, 12, 128)   0           batch_normalization_37[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_33 (Conv2D)               (None, 12, 12, 128)   114688      activation_32[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_38 (Conv2D)               (None, 12, 12, 128)   114688      activation_37[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_33 (BatchNor (None, 12, 12, 128)   384         conv2d_33[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_38 (BatchNor (None, 12, 12, 128)   384         conv2d_38[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_33 (Activation)       (None, 12, 12, 128)   0           batch_normalization_33[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_38 (Activation)       (None, 12, 12, 128)   0           batch_normalization_38[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "average_pooling2d_4 (AveragePool (None, 12, 12, 768)   0           mixed3[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_31 (Conv2D)               (None, 12, 12, 192)   147456      mixed3[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_34 (Conv2D)               (None, 12, 12, 192)   172032      activation_33[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_39 (Conv2D)               (None, 12, 12, 192)   172032      activation_38[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_40 (Conv2D)               (None, 12, 12, 192)   147456      average_pooling2d_4[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_31 (BatchNor (None, 12, 12, 192)   576         conv2d_31[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_34 (BatchNor (None, 12, 12, 192)   576         conv2d_34[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_39 (BatchNor (None, 12, 12, 192)   576         conv2d_39[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_40 (BatchNor (None, 12, 12, 192)   576         conv2d_40[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_31 (Activation)       (None, 12, 12, 192)   0           batch_normalization_31[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_34 (Activation)       (None, 12, 12, 192)   0           batch_normalization_34[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_39 (Activation)       (None, 12, 12, 192)   0           batch_normalization_39[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_40 (Activation)       (None, 12, 12, 192)   0           batch_normalization_40[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "mixed4 (Concatenate)             (None, 12, 12, 768)   0           activation_31[0][0]              \n",
      "                                                                   activation_34[0][0]              \n",
      "                                                                   activation_39[0][0]              \n",
      "                                                                   activation_40[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_45 (Conv2D)               (None, 12, 12, 160)   122880      mixed4[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_45 (BatchNor (None, 12, 12, 160)   480         conv2d_45[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_45 (Activation)       (None, 12, 12, 160)   0           batch_normalization_45[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_46 (Conv2D)               (None, 12, 12, 160)   179200      activation_45[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_46 (BatchNor (None, 12, 12, 160)   480         conv2d_46[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_46 (Activation)       (None, 12, 12, 160)   0           batch_normalization_46[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_42 (Conv2D)               (None, 12, 12, 160)   122880      mixed4[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_47 (Conv2D)               (None, 12, 12, 160)   179200      activation_46[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_42 (BatchNor (None, 12, 12, 160)   480         conv2d_42[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_47 (BatchNor (None, 12, 12, 160)   480         conv2d_47[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_42 (Activation)       (None, 12, 12, 160)   0           batch_normalization_42[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_47 (Activation)       (None, 12, 12, 160)   0           batch_normalization_47[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_43 (Conv2D)               (None, 12, 12, 160)   179200      activation_42[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_48 (Conv2D)               (None, 12, 12, 160)   179200      activation_47[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_43 (BatchNor (None, 12, 12, 160)   480         conv2d_43[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_48 (BatchNor (None, 12, 12, 160)   480         conv2d_48[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_43 (Activation)       (None, 12, 12, 160)   0           batch_normalization_43[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_48 (Activation)       (None, 12, 12, 160)   0           batch_normalization_48[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "average_pooling2d_5 (AveragePool (None, 12, 12, 768)   0           mixed4[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_41 (Conv2D)               (None, 12, 12, 192)   147456      mixed4[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_44 (Conv2D)               (None, 12, 12, 192)   215040      activation_43[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_49 (Conv2D)               (None, 12, 12, 192)   215040      activation_48[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_50 (Conv2D)               (None, 12, 12, 192)   147456      average_pooling2d_5[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_41 (BatchNor (None, 12, 12, 192)   576         conv2d_41[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_44 (BatchNor (None, 12, 12, 192)   576         conv2d_44[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_49 (BatchNor (None, 12, 12, 192)   576         conv2d_49[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_50 (BatchNor (None, 12, 12, 192)   576         conv2d_50[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_41 (Activation)       (None, 12, 12, 192)   0           batch_normalization_41[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_44 (Activation)       (None, 12, 12, 192)   0           batch_normalization_44[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_49 (Activation)       (None, 12, 12, 192)   0           batch_normalization_49[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_50 (Activation)       (None, 12, 12, 192)   0           batch_normalization_50[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "mixed5 (Concatenate)             (None, 12, 12, 768)   0           activation_41[0][0]              \n",
      "                                                                   activation_44[0][0]              \n",
      "                                                                   activation_49[0][0]              \n",
      "                                                                   activation_50[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_55 (Conv2D)               (None, 12, 12, 160)   122880      mixed5[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_55 (BatchNor (None, 12, 12, 160)   480         conv2d_55[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_55 (Activation)       (None, 12, 12, 160)   0           batch_normalization_55[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_56 (Conv2D)               (None, 12, 12, 160)   179200      activation_55[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_56 (BatchNor (None, 12, 12, 160)   480         conv2d_56[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_56 (Activation)       (None, 12, 12, 160)   0           batch_normalization_56[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_52 (Conv2D)               (None, 12, 12, 160)   122880      mixed5[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_57 (Conv2D)               (None, 12, 12, 160)   179200      activation_56[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_52 (BatchNor (None, 12, 12, 160)   480         conv2d_52[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_57 (BatchNor (None, 12, 12, 160)   480         conv2d_57[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_52 (Activation)       (None, 12, 12, 160)   0           batch_normalization_52[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_57 (Activation)       (None, 12, 12, 160)   0           batch_normalization_57[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_53 (Conv2D)               (None, 12, 12, 160)   179200      activation_52[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_58 (Conv2D)               (None, 12, 12, 160)   179200      activation_57[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_53 (BatchNor (None, 12, 12, 160)   480         conv2d_53[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_58 (BatchNor (None, 12, 12, 160)   480         conv2d_58[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_53 (Activation)       (None, 12, 12, 160)   0           batch_normalization_53[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_58 (Activation)       (None, 12, 12, 160)   0           batch_normalization_58[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "average_pooling2d_6 (AveragePool (None, 12, 12, 768)   0           mixed5[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_51 (Conv2D)               (None, 12, 12, 192)   147456      mixed5[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_54 (Conv2D)               (None, 12, 12, 192)   215040      activation_53[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_59 (Conv2D)               (None, 12, 12, 192)   215040      activation_58[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_60 (Conv2D)               (None, 12, 12, 192)   147456      average_pooling2d_6[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_51 (BatchNor (None, 12, 12, 192)   576         conv2d_51[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_54 (BatchNor (None, 12, 12, 192)   576         conv2d_54[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_59 (BatchNor (None, 12, 12, 192)   576         conv2d_59[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_60 (BatchNor (None, 12, 12, 192)   576         conv2d_60[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_51 (Activation)       (None, 12, 12, 192)   0           batch_normalization_51[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_54 (Activation)       (None, 12, 12, 192)   0           batch_normalization_54[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_59 (Activation)       (None, 12, 12, 192)   0           batch_normalization_59[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_60 (Activation)       (None, 12, 12, 192)   0           batch_normalization_60[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "mixed6 (Concatenate)             (None, 12, 12, 768)   0           activation_51[0][0]              \n",
      "                                                                   activation_54[0][0]              \n",
      "                                                                   activation_59[0][0]              \n",
      "                                                                   activation_60[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_65 (Conv2D)               (None, 12, 12, 192)   147456      mixed6[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_65 (BatchNor (None, 12, 12, 192)   576         conv2d_65[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_65 (Activation)       (None, 12, 12, 192)   0           batch_normalization_65[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_66 (Conv2D)               (None, 12, 12, 192)   258048      activation_65[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_66 (BatchNor (None, 12, 12, 192)   576         conv2d_66[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_66 (Activation)       (None, 12, 12, 192)   0           batch_normalization_66[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_62 (Conv2D)               (None, 12, 12, 192)   147456      mixed6[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_67 (Conv2D)               (None, 12, 12, 192)   258048      activation_66[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_62 (BatchNor (None, 12, 12, 192)   576         conv2d_62[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_67 (BatchNor (None, 12, 12, 192)   576         conv2d_67[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_62 (Activation)       (None, 12, 12, 192)   0           batch_normalization_62[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_67 (Activation)       (None, 12, 12, 192)   0           batch_normalization_67[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_63 (Conv2D)               (None, 12, 12, 192)   258048      activation_62[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_68 (Conv2D)               (None, 12, 12, 192)   258048      activation_67[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_63 (BatchNor (None, 12, 12, 192)   576         conv2d_63[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_68 (BatchNor (None, 12, 12, 192)   576         conv2d_68[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_63 (Activation)       (None, 12, 12, 192)   0           batch_normalization_63[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_68 (Activation)       (None, 12, 12, 192)   0           batch_normalization_68[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "average_pooling2d_7 (AveragePool (None, 12, 12, 768)   0           mixed6[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_61 (Conv2D)               (None, 12, 12, 192)   147456      mixed6[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_64 (Conv2D)               (None, 12, 12, 192)   258048      activation_63[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_69 (Conv2D)               (None, 12, 12, 192)   258048      activation_68[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_70 (Conv2D)               (None, 12, 12, 192)   147456      average_pooling2d_7[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_61 (BatchNor (None, 12, 12, 192)   576         conv2d_61[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_64 (BatchNor (None, 12, 12, 192)   576         conv2d_64[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_69 (BatchNor (None, 12, 12, 192)   576         conv2d_69[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_70 (BatchNor (None, 12, 12, 192)   576         conv2d_70[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_61 (Activation)       (None, 12, 12, 192)   0           batch_normalization_61[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_64 (Activation)       (None, 12, 12, 192)   0           batch_normalization_64[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_69 (Activation)       (None, 12, 12, 192)   0           batch_normalization_69[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_70 (Activation)       (None, 12, 12, 192)   0           batch_normalization_70[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "mixed7 (Concatenate)             (None, 12, 12, 768)   0           activation_61[0][0]              \n",
      "                                                                   activation_64[0][0]              \n",
      "                                                                   activation_69[0][0]              \n",
      "                                                                   activation_70[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_73 (Conv2D)               (None, 12, 12, 192)   147456      mixed7[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_73 (BatchNor (None, 12, 12, 192)   576         conv2d_73[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_73 (Activation)       (None, 12, 12, 192)   0           batch_normalization_73[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_74 (Conv2D)               (None, 12, 12, 192)   258048      activation_73[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_74 (BatchNor (None, 12, 12, 192)   576         conv2d_74[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_74 (Activation)       (None, 12, 12, 192)   0           batch_normalization_74[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_71 (Conv2D)               (None, 12, 12, 192)   147456      mixed7[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_75 (Conv2D)               (None, 12, 12, 192)   258048      activation_74[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_71 (BatchNor (None, 12, 12, 192)   576         conv2d_71[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_75 (BatchNor (None, 12, 12, 192)   576         conv2d_75[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_71 (Activation)       (None, 12, 12, 192)   0           batch_normalization_71[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_75 (Activation)       (None, 12, 12, 192)   0           batch_normalization_75[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_72 (Conv2D)               (None, 5, 5, 320)     552960      activation_71[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_76 (Conv2D)               (None, 5, 5, 192)     331776      activation_75[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_72 (BatchNor (None, 5, 5, 320)     960         conv2d_72[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_76 (BatchNor (None, 5, 5, 192)     576         conv2d_76[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_72 (Activation)       (None, 5, 5, 320)     0           batch_normalization_72[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_76 (Activation)       (None, 5, 5, 192)     0           batch_normalization_76[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "max_pooling2d_4 (MaxPooling2D)   (None, 5, 5, 768)     0           mixed7[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "mixed8 (Concatenate)             (None, 5, 5, 1280)    0           activation_72[0][0]              \n",
      "                                                                   activation_76[0][0]              \n",
      "                                                                   max_pooling2d_4[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_81 (Conv2D)               (None, 5, 5, 448)     573440      mixed8[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_81 (BatchNor (None, 5, 5, 448)     1344        conv2d_81[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_81 (Activation)       (None, 5, 5, 448)     0           batch_normalization_81[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_78 (Conv2D)               (None, 5, 5, 384)     491520      mixed8[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_82 (Conv2D)               (None, 5, 5, 384)     1548288     activation_81[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_78 (BatchNor (None, 5, 5, 384)     1152        conv2d_78[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_82 (BatchNor (None, 5, 5, 384)     1152        conv2d_82[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_78 (Activation)       (None, 5, 5, 384)     0           batch_normalization_78[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_82 (Activation)       (None, 5, 5, 384)     0           batch_normalization_82[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_79 (Conv2D)               (None, 5, 5, 384)     442368      activation_78[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_80 (Conv2D)               (None, 5, 5, 384)     442368      activation_78[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_83 (Conv2D)               (None, 5, 5, 384)     442368      activation_82[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_84 (Conv2D)               (None, 5, 5, 384)     442368      activation_82[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "average_pooling2d_8 (AveragePool (None, 5, 5, 1280)    0           mixed8[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_77 (Conv2D)               (None, 5, 5, 320)     409600      mixed8[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_79 (BatchNor (None, 5, 5, 384)     1152        conv2d_79[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_80 (BatchNor (None, 5, 5, 384)     1152        conv2d_80[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_83 (BatchNor (None, 5, 5, 384)     1152        conv2d_83[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_84 (BatchNor (None, 5, 5, 384)     1152        conv2d_84[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_85 (Conv2D)               (None, 5, 5, 192)     245760      average_pooling2d_8[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_77 (BatchNor (None, 5, 5, 320)     960         conv2d_77[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_79 (Activation)       (None, 5, 5, 384)     0           batch_normalization_79[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_80 (Activation)       (None, 5, 5, 384)     0           batch_normalization_80[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_83 (Activation)       (None, 5, 5, 384)     0           batch_normalization_83[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_84 (Activation)       (None, 5, 5, 384)     0           batch_normalization_84[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_85 (BatchNor (None, 5, 5, 192)     576         conv2d_85[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_77 (Activation)       (None, 5, 5, 320)     0           batch_normalization_77[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "mixed9_0 (Concatenate)           (None, 5, 5, 768)     0           activation_79[0][0]              \n",
      "                                                                   activation_80[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "concatenate_1 (Concatenate)      (None, 5, 5, 768)     0           activation_83[0][0]              \n",
      "                                                                   activation_84[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "activation_85 (Activation)       (None, 5, 5, 192)     0           batch_normalization_85[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "mixed9 (Concatenate)             (None, 5, 5, 2048)    0           activation_77[0][0]              \n",
      "                                                                   mixed9_0[0][0]                   \n",
      "                                                                   concatenate_1[0][0]              \n",
      "                                                                   activation_85[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_90 (Conv2D)               (None, 5, 5, 448)     917504      mixed9[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_90 (BatchNor (None, 5, 5, 448)     1344        conv2d_90[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_90 (Activation)       (None, 5, 5, 448)     0           batch_normalization_90[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_87 (Conv2D)               (None, 5, 5, 384)     786432      mixed9[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_91 (Conv2D)               (None, 5, 5, 384)     1548288     activation_90[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_87 (BatchNor (None, 5, 5, 384)     1152        conv2d_87[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_91 (BatchNor (None, 5, 5, 384)     1152        conv2d_91[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_87 (Activation)       (None, 5, 5, 384)     0           batch_normalization_87[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_91 (Activation)       (None, 5, 5, 384)     0           batch_normalization_91[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_88 (Conv2D)               (None, 5, 5, 384)     442368      activation_87[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_89 (Conv2D)               (None, 5, 5, 384)     442368      activation_87[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_92 (Conv2D)               (None, 5, 5, 384)     442368      activation_91[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_93 (Conv2D)               (None, 5, 5, 384)     442368      activation_91[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "average_pooling2d_9 (AveragePool (None, 5, 5, 2048)    0           mixed9[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_86 (Conv2D)               (None, 5, 5, 320)     655360      mixed9[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_88 (BatchNor (None, 5, 5, 384)     1152        conv2d_88[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_89 (BatchNor (None, 5, 5, 384)     1152        conv2d_89[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_92 (BatchNor (None, 5, 5, 384)     1152        conv2d_92[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_93 (BatchNor (None, 5, 5, 384)     1152        conv2d_93[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_94 (Conv2D)               (None, 5, 5, 192)     393216      average_pooling2d_9[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_86 (BatchNor (None, 5, 5, 320)     960         conv2d_86[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_88 (Activation)       (None, 5, 5, 384)     0           batch_normalization_88[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_89 (Activation)       (None, 5, 5, 384)     0           batch_normalization_89[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_92 (Activation)       (None, 5, 5, 384)     0           batch_normalization_92[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "activation_93 (Activation)       (None, 5, 5, 384)     0           batch_normalization_93[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_94 (BatchNor (None, 5, 5, 192)     576         conv2d_94[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "activation_86 (Activation)       (None, 5, 5, 320)     0           batch_normalization_86[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "mixed9_1 (Concatenate)           (None, 5, 5, 768)     0           activation_88[0][0]              \n",
      "                                                                   activation_89[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "concatenate_2 (Concatenate)      (None, 5, 5, 768)     0           activation_92[0][0]              \n",
      "                                                                   activation_93[0][0]              \n",
      "____________________________________________________________________________________________________\n",
      "activation_94 (Activation)       (None, 5, 5, 192)     0           batch_normalization_94[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "mixed10 (Concatenate)            (None, 5, 5, 2048)    0           activation_86[0][0]              \n",
      "                                                                   mixed9_1[0][0]                   \n",
      "                                                                   concatenate_2[0][0]              \n",
      "                                                                   activation_94[0][0]              \n",
      "====================================================================================================\n",
      "Total params: 21,802,784\n",
      "Trainable params: 21,768,352\n",
      "Non-trainable params: 34,432\n",
      "____________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Anaconda3\\lib\\site-packages\\ipykernel\\__main__.py:1: UserWarning: Update your `Model` call to the Keras 2 API: `Model(inputs=Tensor(\"in..., outputs=Tensor(\"mi...)`\n",
      "  if __name__ == '__main__':\n"
     ]
    }
   ],
   "source": [
    "feat_ext1=Model(input=model.input, output=model.get_layer('mixed10').output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data/MSTAR_Chips/TRAIN_images/bmp2_tank\n",
      "data/MSTAR_Chips/TRAIN_images/btr70_transport\n",
      "data/MSTAR_Chips/TRAIN_images/t72_tank\n"
     ]
    }
   ],
   "source": [
    "# path to training dataset\n",
    "train_labels = os.listdir(train_data_dir)\n",
    "\n",
    "\n",
    "# variables to hold features and labels\n",
    "inception_features = []\n",
    "inception_test_features = []\n",
    "inception_labels   = []\n",
    "inception_test_labels =[]\n",
    "image_size=(224,224)\n",
    "# loop over all the labels in the folder\n",
    "for label in train_labels:\n",
    "\tcur_path = train_data_dir + \"/\" + label\n",
    "\tprint(cur_path)\n",
    "\tfor image_path in glob.glob(cur_path + \"/*.jpg\"):\n",
    "\t\timg = image.load_img(image_path, target_size=image_size)\n",
    "\t\tx = image.img_to_array(img)\n",
    "\t\tx = np.expand_dims(x, axis=0)\n",
    "\t\tx = preprocess_input(x)\n",
    "\t\tfeature = feat_ext1.predict(x)\n",
    "\t\tflat = feature.flatten()\n",
    "\t\tinception_features.append(flat)\n",
    "\t\tinception_labels.append(label)\n",
    "\n",
    "# loop over all the labels in the folder\n",
    "for label in train_labels:\n",
    "\tcur_path = validation_data_dir + \"/\" + label\n",
    "\tfor image_path in glob.glob(cur_path + \"/*.jpg\"):\n",
    "\t\timg = image.load_img(image_path, target_size=image_size)\n",
    "\t\tx = image.img_to_array(img)\n",
    "\t\tx = np.expand_dims(x, axis=0)\n",
    "\t\tx = preprocess_input(x)\n",
    "\t\tfeature = feat_ext1.predict(x)\n",
    "\t\tflat = feature.flatten()\n",
    "\t\tinception_test_features.append(flat)\n",
    "\t\tinception_test_labels.append(label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# encode the labels using LabelEncoder\n",
    "targetNames = np.unique(labels)\n",
    "le = LabelEncoder()\n",
    "inception_le_labels = le.fit_transform(inception_labels)\n",
    "inception_test_labels=le.transform(inception_test_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "seed=2017"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] creating model...\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
       "          penalty='l2', random_state=2017, solver='liblinear', tol=0.0001,\n",
       "          verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# use logistic regression as the model\n",
    "print(\"[INFO] creating model...\")\n",
    "model = LogisticRegression(random_state=seed)\n",
    "model.fit(inception_features, inception_le_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 precision    recall  f1-score   support\n",
      "\n",
      "      bmp2_tank       0.84      0.90      0.87       549\n",
      "btr70_transport       0.78      0.83      0.80       183\n",
      "       t72_tank       0.95      0.88      0.91       633\n",
      "\n",
      "    avg / total       0.88      0.88      0.88      1365\n",
      "\n",
      "Accuracy : 0.878388278388\n"
     ]
    }
   ],
   "source": [
    "# evaluate the model of test data\n",
    "preds = model.predict(inception_test_features)\n",
    "print(classification_report(le.inverse_transform(preds),le.inverse_transform(inception_test_labels)))\n",
    "print(\"Accuracy :\",accuracy_score(le.inverse_transform(preds),le.inverse_transform(inception_test_labels)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## VGG16"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model_ = applications.VGG16(weights='imagenet',include_top=False, input_shape = (img_width, img_height, 3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Anaconda3\\lib\\site-packages\\ipykernel\\__main__.py:1: UserWarning: Update your `Model` call to the Keras 2 API: `Model(inputs=Tensor(\"in..., outputs=Tensor(\"bl...)`\n",
      "  if __name__ == '__main__':\n"
     ]
    }
   ],
   "source": [
    "feat_ext_vgg=Model(input=model_.input, output=model_.get_layer('block5_pool').output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data/MSTAR_Chips/TRAIN_images/bmp2_tank\n",
      "data/MSTAR_Chips/TRAIN_images/btr70_transport\n",
      "data/MSTAR_Chips/TRAIN_images/t72_tank\n"
     ]
    }
   ],
   "source": [
    "# path to training dataset\n",
    "train_labels = os.listdir(train_data_dir)\n",
    "\n",
    "\n",
    "# variables to hold features and labels\n",
    "vgg_features = []\n",
    "vgg_test_features = []\n",
    "vgg_labels   = []\n",
    "vgg_test_labels =[]\n",
    "image_size=(128,128)\n",
    "# loop over all the labels in the folder\n",
    "for label in train_labels:\n",
    "\tcur_path = train_data_dir + \"/\" + label\n",
    "\tprint(cur_path)\n",
    "\tfor image_path in glob.glob(cur_path + \"/*.jpg\"):\n",
    "\t\timg = image.load_img(image_path, target_size=image_size)\n",
    "\t\tx = image.img_to_array(img)\n",
    "\t\tx = np.expand_dims(x, axis=0)\n",
    "\t\tx = preprocess_input(x)\n",
    "\t\tfeature = feat_ext_vgg.predict(x)\n",
    "\t\tflat = feature.flatten()\n",
    "\t\tvgg_features.append(flat)\n",
    "\t\tvgg_labels.append(label)\n",
    "\n",
    "# loop over all the labels in the folder\n",
    "for label in train_labels:\n",
    "\tcur_path = validation_data_dir + \"/\" + label\n",
    "\tfor image_path in glob.glob(cur_path + \"/*.jpg\"):\n",
    "\t\timg = image.load_img(image_path, target_size=image_size)\n",
    "\t\tx = image.img_to_array(img)\n",
    "\t\tx = np.expand_dims(x, axis=0)\n",
    "\t\tx = preprocess_input(x)\n",
    "\t\tfeature = feat_ext_vgg.predict(x)\n",
    "\t\tflat = feature.flatten()\n",
    "\t\tvgg_test_features.append(flat)\n",
    "\t\tvgg_test_labels.append(label)\n",
    "\t\t"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# encode the labels using LabelEncoder\n",
    "targetNames = np.unique(labels)\n",
    "le = LabelEncoder()\n",
    "vgg_le_labels = le.fit_transform(vgg_labels)\n",
    "vgg_test_labels=le.transform(vgg_test_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] creating model...\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
       "          penalty='l2', random_state=2017, solver='liblinear', tol=0.0001,\n",
       "          verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# use logistic regression as the model\n",
    "print(\"[INFO] creating model...\")\n",
    "model = LogisticRegression(random_state=seed)\n",
    "model.fit(vgg_features, vgg_le_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 precision    recall  f1-score   support\n",
      "\n",
      "      bmp2_tank       0.87      0.92      0.89       550\n",
      "btr70_transport       0.82      0.82      0.82       196\n",
      "       t72_tank       0.97      0.91      0.94       619\n",
      "\n",
      "    avg / total       0.91      0.90      0.90      1365\n",
      "\n",
      "Accuracy : 0.903296703297\n"
     ]
    }
   ],
   "source": [
    "# evaluate the model of test data\n",
    "preds = model.predict(vgg_test_features)\n",
    "print(classification_report(le.inverse_transform(preds),le.inverse_transform(vgg_test_labels)))\n",
    "print(\"Accuracy :\",accuracy_score(le.inverse_transform(preds),le.inverse_transform(vgg_test_labels)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1622"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Combine Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "img_width, img_height = 224,224\n",
    "train_data_dir = \"data/MSTAR_Chips/TRAIN_images\"\n",
    "validation_data_dir = \"data/MSTAR_Chips/TEST_images\"\n",
    "nb_train_samples = 1622\n",
    "nb_validation_samples = 1365 \n",
    "batch_size = 32\n",
    "epochs = 50"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/rishikesh/anaconda/envs/tensorflow/lib/python3.5/site-packages/ipykernel_launcher.py:2: UserWarning: Update your `Model` call to the Keras 2 API: `Model(outputs=Tensor(\"mi..., inputs=Tensor(\"in...)`\n",
      "  \n"
     ]
    }
   ],
   "source": [
    "model = applications.InceptionV3(weights='imagenet',include_top=False, input_shape = (img_width, img_height, 3))\n",
    "feat_ext1=Model(input=model.input, output=model.get_layer('mixed10').output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/rishikesh/anaconda/envs/tensorflow/lib/python3.5/site-packages/ipykernel_launcher.py:2: UserWarning: Update your `Model` call to the Keras 2 API: `Model(inputs=Tensor(\"in..., outputs=Tensor(\"bl...)`\n",
      "  \n"
     ]
    }
   ],
   "source": [
    "model_ = applications.VGG16(weights='imagenet',include_top=False, input_shape = (img_width, img_height, 3))\n",
    "feat_ext_vgg=Model(input=model_.input, output=model_.get_layer('block5_pool').output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sq_model =applications.Xception(weights='imagenet',include_top=False, input_shape = (img_width, img_height, 3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "____________________________________________________________________________________________________\n",
      "Layer (type)                     Output Shape          Param #     Connected to                     \n",
      "====================================================================================================\n",
      "input_3 (InputLayer)             (None, 224, 224, 3)   0                                            \n",
      "____________________________________________________________________________________________________\n",
      "block1_conv1 (Conv2D)            (None, 111, 111, 32)  864         input_3[0][0]                    \n",
      "____________________________________________________________________________________________________\n",
      "block1_conv1_bn (BatchNormalizat (None, 111, 111, 32)  128         block1_conv1[0][0]               \n",
      "____________________________________________________________________________________________________\n",
      "block1_conv1_act (Activation)    (None, 111, 111, 32)  0           block1_conv1_bn[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block1_conv2 (Conv2D)            (None, 109, 109, 64)  18432       block1_conv1_act[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block1_conv2_bn (BatchNormalizat (None, 109, 109, 64)  256         block1_conv2[0][0]               \n",
      "____________________________________________________________________________________________________\n",
      "block1_conv2_act (Activation)    (None, 109, 109, 64)  0           block1_conv2_bn[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block2_sepconv1 (SeparableConv2D (None, 109, 109, 128) 8768        block1_conv2_act[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block2_sepconv1_bn (BatchNormali (None, 109, 109, 128) 512         block2_sepconv1[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block2_sepconv2_act (Activation) (None, 109, 109, 128) 0           block2_sepconv1_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block2_sepconv2 (SeparableConv2D (None, 109, 109, 128) 17536       block2_sepconv2_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block2_sepconv2_bn (BatchNormali (None, 109, 109, 128) 512         block2_sepconv2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_95 (Conv2D)               (None, 55, 55, 128)   8192        block1_conv2_act[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block2_pool (MaxPooling2D)       (None, 55, 55, 128)   0           block2_sepconv2_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_95 (BatchNor (None, 55, 55, 128)   512         conv2d_95[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "add_1 (Add)                      (None, 55, 55, 128)   0           block2_pool[0][0]                \n",
      "                                                                   batch_normalization_95[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "block3_sepconv1_act (Activation) (None, 55, 55, 128)   0           add_1[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block3_sepconv1 (SeparableConv2D (None, 55, 55, 256)   33920       block3_sepconv1_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block3_sepconv1_bn (BatchNormali (None, 55, 55, 256)   1024        block3_sepconv1[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block3_sepconv2_act (Activation) (None, 55, 55, 256)   0           block3_sepconv1_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block3_sepconv2 (SeparableConv2D (None, 55, 55, 256)   67840       block3_sepconv2_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block3_sepconv2_bn (BatchNormali (None, 55, 55, 256)   1024        block3_sepconv2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_96 (Conv2D)               (None, 28, 28, 256)   32768       add_1[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block3_pool (MaxPooling2D)       (None, 28, 28, 256)   0           block3_sepconv2_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_96 (BatchNor (None, 28, 28, 256)   1024        conv2d_96[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "add_2 (Add)                      (None, 28, 28, 256)   0           block3_pool[0][0]                \n",
      "                                                                   batch_normalization_96[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "block4_sepconv1_act (Activation) (None, 28, 28, 256)   0           add_2[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block4_sepconv1 (SeparableConv2D (None, 28, 28, 728)   188672      block4_sepconv1_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block4_sepconv1_bn (BatchNormali (None, 28, 28, 728)   2912        block4_sepconv1[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block4_sepconv2_act (Activation) (None, 28, 28, 728)   0           block4_sepconv1_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block4_sepconv2 (SeparableConv2D (None, 28, 28, 728)   536536      block4_sepconv2_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block4_sepconv2_bn (BatchNormali (None, 28, 28, 728)   2912        block4_sepconv2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_97 (Conv2D)               (None, 14, 14, 728)   186368      add_2[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block4_pool (MaxPooling2D)       (None, 14, 14, 728)   0           block4_sepconv2_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_97 (BatchNor (None, 14, 14, 728)   2912        conv2d_97[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "add_3 (Add)                      (None, 14, 14, 728)   0           block4_pool[0][0]                \n",
      "                                                                   batch_normalization_97[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "block5_sepconv1_act (Activation) (None, 14, 14, 728)   0           add_3[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block5_sepconv1 (SeparableConv2D (None, 14, 14, 728)   536536      block5_sepconv1_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block5_sepconv1_bn (BatchNormali (None, 14, 14, 728)   2912        block5_sepconv1[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block5_sepconv2_act (Activation) (None, 14, 14, 728)   0           block5_sepconv1_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block5_sepconv2 (SeparableConv2D (None, 14, 14, 728)   536536      block5_sepconv2_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block5_sepconv2_bn (BatchNormali (None, 14, 14, 728)   2912        block5_sepconv2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block5_sepconv3_act (Activation) (None, 14, 14, 728)   0           block5_sepconv2_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block5_sepconv3 (SeparableConv2D (None, 14, 14, 728)   536536      block5_sepconv3_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block5_sepconv3_bn (BatchNormali (None, 14, 14, 728)   2912        block5_sepconv3[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "add_4 (Add)                      (None, 14, 14, 728)   0           block5_sepconv3_bn[0][0]         \n",
      "                                                                   add_3[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block6_sepconv1_act (Activation) (None, 14, 14, 728)   0           add_4[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block6_sepconv1 (SeparableConv2D (None, 14, 14, 728)   536536      block6_sepconv1_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block6_sepconv1_bn (BatchNormali (None, 14, 14, 728)   2912        block6_sepconv1[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block6_sepconv2_act (Activation) (None, 14, 14, 728)   0           block6_sepconv1_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block6_sepconv2 (SeparableConv2D (None, 14, 14, 728)   536536      block6_sepconv2_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block6_sepconv2_bn (BatchNormali (None, 14, 14, 728)   2912        block6_sepconv2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block6_sepconv3_act (Activation) (None, 14, 14, 728)   0           block6_sepconv2_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block6_sepconv3 (SeparableConv2D (None, 14, 14, 728)   536536      block6_sepconv3_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block6_sepconv3_bn (BatchNormali (None, 14, 14, 728)   2912        block6_sepconv3[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "add_5 (Add)                      (None, 14, 14, 728)   0           block6_sepconv3_bn[0][0]         \n",
      "                                                                   add_4[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block7_sepconv1_act (Activation) (None, 14, 14, 728)   0           add_5[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block7_sepconv1 (SeparableConv2D (None, 14, 14, 728)   536536      block7_sepconv1_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block7_sepconv1_bn (BatchNormali (None, 14, 14, 728)   2912        block7_sepconv1[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block7_sepconv2_act (Activation) (None, 14, 14, 728)   0           block7_sepconv1_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block7_sepconv2 (SeparableConv2D (None, 14, 14, 728)   536536      block7_sepconv2_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block7_sepconv2_bn (BatchNormali (None, 14, 14, 728)   2912        block7_sepconv2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block7_sepconv3_act (Activation) (None, 14, 14, 728)   0           block7_sepconv2_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block7_sepconv3 (SeparableConv2D (None, 14, 14, 728)   536536      block7_sepconv3_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block7_sepconv3_bn (BatchNormali (None, 14, 14, 728)   2912        block7_sepconv3[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "add_6 (Add)                      (None, 14, 14, 728)   0           block7_sepconv3_bn[0][0]         \n",
      "                                                                   add_5[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block8_sepconv1_act (Activation) (None, 14, 14, 728)   0           add_6[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block8_sepconv1 (SeparableConv2D (None, 14, 14, 728)   536536      block8_sepconv1_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block8_sepconv1_bn (BatchNormali (None, 14, 14, 728)   2912        block8_sepconv1[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block8_sepconv2_act (Activation) (None, 14, 14, 728)   0           block8_sepconv1_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block8_sepconv2 (SeparableConv2D (None, 14, 14, 728)   536536      block8_sepconv2_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block8_sepconv2_bn (BatchNormali (None, 14, 14, 728)   2912        block8_sepconv2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block8_sepconv3_act (Activation) (None, 14, 14, 728)   0           block8_sepconv2_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block8_sepconv3 (SeparableConv2D (None, 14, 14, 728)   536536      block8_sepconv3_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block8_sepconv3_bn (BatchNormali (None, 14, 14, 728)   2912        block8_sepconv3[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "add_7 (Add)                      (None, 14, 14, 728)   0           block8_sepconv3_bn[0][0]         \n",
      "                                                                   add_6[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block9_sepconv1_act (Activation) (None, 14, 14, 728)   0           add_7[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block9_sepconv1 (SeparableConv2D (None, 14, 14, 728)   536536      block9_sepconv1_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block9_sepconv1_bn (BatchNormali (None, 14, 14, 728)   2912        block9_sepconv1[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block9_sepconv2_act (Activation) (None, 14, 14, 728)   0           block9_sepconv1_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block9_sepconv2 (SeparableConv2D (None, 14, 14, 728)   536536      block9_sepconv2_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block9_sepconv2_bn (BatchNormali (None, 14, 14, 728)   2912        block9_sepconv2[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "block9_sepconv3_act (Activation) (None, 14, 14, 728)   0           block9_sepconv2_bn[0][0]         \n",
      "____________________________________________________________________________________________________\n",
      "block9_sepconv3 (SeparableConv2D (None, 14, 14, 728)   536536      block9_sepconv3_act[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block9_sepconv3_bn (BatchNormali (None, 14, 14, 728)   2912        block9_sepconv3[0][0]            \n",
      "____________________________________________________________________________________________________\n",
      "add_8 (Add)                      (None, 14, 14, 728)   0           block9_sepconv3_bn[0][0]         \n",
      "                                                                   add_7[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block10_sepconv1_act (Activation (None, 14, 14, 728)   0           add_8[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block10_sepconv1 (SeparableConv2 (None, 14, 14, 728)   536536      block10_sepconv1_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block10_sepconv1_bn (BatchNormal (None, 14, 14, 728)   2912        block10_sepconv1[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block10_sepconv2_act (Activation (None, 14, 14, 728)   0           block10_sepconv1_bn[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block10_sepconv2 (SeparableConv2 (None, 14, 14, 728)   536536      block10_sepconv2_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block10_sepconv2_bn (BatchNormal (None, 14, 14, 728)   2912        block10_sepconv2[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block10_sepconv3_act (Activation (None, 14, 14, 728)   0           block10_sepconv2_bn[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block10_sepconv3 (SeparableConv2 (None, 14, 14, 728)   536536      block10_sepconv3_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block10_sepconv3_bn (BatchNormal (None, 14, 14, 728)   2912        block10_sepconv3[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "add_9 (Add)                      (None, 14, 14, 728)   0           block10_sepconv3_bn[0][0]        \n",
      "                                                                   add_8[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block11_sepconv1_act (Activation (None, 14, 14, 728)   0           add_9[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block11_sepconv1 (SeparableConv2 (None, 14, 14, 728)   536536      block11_sepconv1_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block11_sepconv1_bn (BatchNormal (None, 14, 14, 728)   2912        block11_sepconv1[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block11_sepconv2_act (Activation (None, 14, 14, 728)   0           block11_sepconv1_bn[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block11_sepconv2 (SeparableConv2 (None, 14, 14, 728)   536536      block11_sepconv2_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block11_sepconv2_bn (BatchNormal (None, 14, 14, 728)   2912        block11_sepconv2[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block11_sepconv3_act (Activation (None, 14, 14, 728)   0           block11_sepconv2_bn[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block11_sepconv3 (SeparableConv2 (None, 14, 14, 728)   536536      block11_sepconv3_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block11_sepconv3_bn (BatchNormal (None, 14, 14, 728)   2912        block11_sepconv3[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "add_10 (Add)                     (None, 14, 14, 728)   0           block11_sepconv3_bn[0][0]        \n",
      "                                                                   add_9[0][0]                      \n",
      "____________________________________________________________________________________________________\n",
      "block12_sepconv1_act (Activation (None, 14, 14, 728)   0           add_10[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "block12_sepconv1 (SeparableConv2 (None, 14, 14, 728)   536536      block12_sepconv1_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block12_sepconv1_bn (BatchNormal (None, 14, 14, 728)   2912        block12_sepconv1[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block12_sepconv2_act (Activation (None, 14, 14, 728)   0           block12_sepconv1_bn[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block12_sepconv2 (SeparableConv2 (None, 14, 14, 728)   536536      block12_sepconv2_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block12_sepconv2_bn (BatchNormal (None, 14, 14, 728)   2912        block12_sepconv2[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block12_sepconv3_act (Activation (None, 14, 14, 728)   0           block12_sepconv2_bn[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block12_sepconv3 (SeparableConv2 (None, 14, 14, 728)   536536      block12_sepconv3_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block12_sepconv3_bn (BatchNormal (None, 14, 14, 728)   2912        block12_sepconv3[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "add_11 (Add)                     (None, 14, 14, 728)   0           block12_sepconv3_bn[0][0]        \n",
      "                                                                   add_10[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "block13_sepconv1_act (Activation (None, 14, 14, 728)   0           add_11[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "block13_sepconv1 (SeparableConv2 (None, 14, 14, 728)   536536      block13_sepconv1_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block13_sepconv1_bn (BatchNormal (None, 14, 14, 728)   2912        block13_sepconv1[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block13_sepconv2_act (Activation (None, 14, 14, 728)   0           block13_sepconv1_bn[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block13_sepconv2 (SeparableConv2 (None, 14, 14, 1024)  752024      block13_sepconv2_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block13_sepconv2_bn (BatchNormal (None, 14, 14, 1024)  4096        block13_sepconv2[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "conv2d_98 (Conv2D)               (None, 7, 7, 1024)    745472      add_11[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "block13_pool (MaxPooling2D)      (None, 7, 7, 1024)    0           block13_sepconv2_bn[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "batch_normalization_98 (BatchNor (None, 7, 7, 1024)    4096        conv2d_98[0][0]                  \n",
      "____________________________________________________________________________________________________\n",
      "add_12 (Add)                     (None, 7, 7, 1024)    0           block13_pool[0][0]               \n",
      "                                                                   batch_normalization_98[0][0]     \n",
      "____________________________________________________________________________________________________\n",
      "block14_sepconv1 (SeparableConv2 (None, 7, 7, 1536)    1582080     add_12[0][0]                     \n",
      "____________________________________________________________________________________________________\n",
      "block14_sepconv1_bn (BatchNormal (None, 7, 7, 1536)    6144        block14_sepconv1[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block14_sepconv1_act (Activation (None, 7, 7, 1536)    0           block14_sepconv1_bn[0][0]        \n",
      "____________________________________________________________________________________________________\n",
      "block14_sepconv2 (SeparableConv2 (None, 7, 7, 2048)    3159552     block14_sepconv1_act[0][0]       \n",
      "____________________________________________________________________________________________________\n",
      "block14_sepconv2_bn (BatchNormal (None, 7, 7, 2048)    8192        block14_sepconv2[0][0]           \n",
      "____________________________________________________________________________________________________\n",
      "block14_sepconv2_act (Activation (None, 7, 7, 2048)    0           block14_sepconv2_bn[0][0]        \n",
      "====================================================================================================\n",
      "Total params: 20,861,480\n",
      "Trainable params: 20,806,952\n",
      "Non-trainable params: 54,528\n",
      "____________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "sq_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/rishikesh/anaconda/envs/tensorflow/lib/python3.5/site-packages/ipykernel_launcher.py:1: UserWarning: Update your `Model` call to the Keras 2 API: `Model(inputs=Tensor(\"in..., outputs=Tensor(\"bl...)`\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    }
   ],
   "source": [
    "feat_ext_sq=Model(input=sq_model.input, output=sq_model.get_layer('block14_sepconv2_act').output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data/MSTAR_Chips/TRAIN_images/bmp2_tank\n",
      "data/MSTAR_Chips/TRAIN_images/btr70_transport\n",
      "data/MSTAR_Chips/TRAIN_images/t72_tank\n"
     ]
    }
   ],
   "source": [
    "# path to training dataset\n",
    "train_labels = os.listdir(train_data_dir)\n",
    "\n",
    "\n",
    "# variables to hold inception features and labels\n",
    "inception_features = []\n",
    "inception_test_features = []\n",
    "\n",
    "# variables to hold vgg features and labels\n",
    "vgg_features = []\n",
    "vgg_test_features = []\n",
    "\n",
    "# variables to hold vgg features and labels\n",
    "sq_features = []\n",
    "sq_test_features = []\n",
    "\n",
    "\n",
    "labels   = []\n",
    "test_labels =[]\n",
    "\n",
    "image_size=(224,224)\n",
    "# loop over all the labels in the folder\n",
    "for label in train_labels:\n",
    "\tcur_path = train_data_dir + \"/\" + label\n",
    "\tprint(cur_path)\n",
    "\tfor image_path in glob.glob(cur_path + \"/*.jpg\"):\n",
    "\t\timg = image.load_img(image_path, target_size=image_size)\n",
    "\t\tx = image.img_to_array(img)\n",
    "\t\tx = np.expand_dims(x, axis=0)\n",
    "\t\tx = preprocess_input(x)\n",
    "\t\tfeature = feat_ext1.predict(x)\n",
    "\t\tflat = feature.flatten()\n",
    "\t\tinception_features.append(flat)\n",
    "\t\tfeature = feat_ext_vgg.predict(x)\n",
    "\t\tflat = feature.flatten()\n",
    "\t\tvgg_features.append(flat)\n",
    "\t\tfeature = feat_ext_sq.predict(x)\n",
    "\t\tflat = feature.flatten()\n",
    "\t\tsq_features.append(flat)\n",
    "\t\tlabels.append(label)\n",
    "\n",
    "# loop over all the labels in the folder\n",
    "for label in train_labels:\n",
    "\tcur_path = validation_data_dir + \"/\" + label\n",
    "\tfor image_path in glob.glob(cur_path + \"/*.jpg\"):\n",
    "\t\timg = image.load_img(image_path, target_size=image_size)\n",
    "\t\tx = image.img_to_array(img)\n",
    "\t\tx = np.expand_dims(x, axis=0)\n",
    "\t\tx = preprocess_input(x)\n",
    "\t\tfeature = feat_ext1.predict(x)\n",
    "\t\tflat = feature.flatten()\n",
    "\t\tinception_test_features.append(flat)\n",
    "\t\tfeature = feat_ext_vgg.predict(x)\n",
    "\t\tflat = feature.flatten()\n",
    "\t\tvgg_test_features.append(flat)\n",
    "\t\tfeature = feat_ext_sq.predict(x)\n",
    "\t\tflat = feature.flatten()\n",
    "\t\tsq_test_features.append(flat)\n",
    "\t\ttest_labels.append(label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# encode the labels using LabelEncoder\n",
    "targetNames = np.unique(labels)\n",
    "le = LabelEncoder()\n",
    "le_labels = le.fit_transform(labels)\n",
    "test_labels=le.transform(test_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1622, 25088)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.asarray(vgg_features).shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1622, 51200)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.asarray(inception_features).shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1622, 100352)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.asarray(sq_features).shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "features = np.concatenate( [np.asarray(vgg_features), np.asarray(inception_features), np.asarray(sq_features)], axis=1 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "test_features = np.concatenate( [np.asarray(vgg_test_features), np.asarray(inception_test_features),np.asarray(sq_test_features)], axis=1 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "seed=2017"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/rishikesh/anaconda/envs/tensorflow/lib/python3.5/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "import xgboost as xgb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Prepare the inputs for the model\n",
    "train_X = np.asmatrix(features)\n",
    "test_X = np.asmatrix(test_features)\n",
    "#train_y = train_df['Survived']\n",
    "\n",
    "# You can experiment with many other options here, using the same .fit() and .predict()\n",
    "# methods; see http://scikit-learn.org\n",
    "# This example uses the current build of XGBoost, from https://github.com/dmlc/xgboost\n",
    "gbm = xgb.XGBClassifier(max_depth=3, n_estimators=300, learning_rate=0.05).fit(train_X, le_labels)\n",
    "predictions = gbm.predict(test_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] creating model...\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
       "          penalty='l2', random_state=2017, solver='liblinear', tol=0.0001,\n",
       "          verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# use logistic regression as the model\n",
    "print(\"[INFO] creating model...\")\n",
    "model = LogisticRegression(random_state=seed)\n",
    "model.fit(features, le_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 precision    recall  f1-score   support\n",
      "\n",
      "      bmp2_tank       0.87      0.92      0.90       555\n",
      "btr70_transport       0.79      0.85      0.82       181\n",
      "       t72_tank       0.97      0.90      0.94       629\n",
      "\n",
      "    avg / total       0.91      0.90      0.91      1365\n",
      "\n",
      "Accuracy : 0.904029304029\n"
     ]
    }
   ],
   "source": [
    "# evaluate the model of test data\n",
    "preds = model.predict(test_features)\n",
    "print(classification_report(le.inverse_transform(preds),le.inverse_transform(test_labels)))\n",
    "print(\"Accuracy :\",accuracy_score(le.inverse_transform(preds),le.inverse_transform(test_labels)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.decomposition import PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1622, 176640)"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1365, 176640)"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_features.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X=np.concatenate( [features,test_features], axis=0 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2987, 176640)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pca = PCA(n_components=10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PCA(copy=True, iterated_power='auto', n_components=10000, random_state=None,\n",
       "  svd_solver='auto', tol=0.0, whiten=False)"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca.fit(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] creating model...\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
       "          penalty='l2', random_state=2017, solver='liblinear', tol=0.0001,\n",
       "          verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# use logistic regression as the model\n",
    "print(\"[INFO] creating model...\")\n",
    "clf = LogisticRegression(random_state=seed)\n",
    "clf.fit(pca.transform(features), le_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# evaluate the model of test data\n",
    "preds = clf.predict(pca.transform(test_features))\n",
    "print(classification_report(le.inverse_transform(preds),le.inverse_transform(test_labels)))\n",
    "print(\"Accuracy :\",accuracy_score(le.inverse_transform(preds),le.inverse_transform(test_labels)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}