tflerning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_path = 'D:\\plantvillage_deeplearning_paper_dataset\\plantvillage_deeplearning_paper_dataset\\color'\n",
    "test_path = 'images/test/'\n",
    "batch_size = 32\n",
    "image_size = 128\n",
    "num_class = 38\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 43456 images belonging to 38 classes.\n",
      "Found 10849 images belonging to 38 classes.\n"
     ]
    }
   ],
   "source": [
    "\n",
    "train_datagen = ImageDataGenerator(validation_split=0.2,\n",
    "                                   shear_range=0.2,\n",
    "                                   zoom_range=0.2,\n",
    "                                   horizontal_flip=True\n",
    "                                  )\n",
    "\n",
    "\n",
    "train_generator = train_datagen.flow_from_directory(\n",
    "                        directory=train_path,\n",
    "                        target_size=(image_size,image_size),\n",
    "                        batch_size=batch_size,\n",
    "                        class_mode='categorical',\n",
    "                        color_mode='rgb',\n",
    "                        subset=\"training\",\n",
    "                        shuffle=True)\n",
    "\n",
    "validation_generator = train_datagen.flow_from_directory(\n",
    "                        directory=train_path,\n",
    "                        target_size=(image_size,image_size),\n",
    "                        batch_size=batch_size,\n",
    "                        class_mode='categorical',\n",
    "                        color_mode='rgb',\n",
    "                        subset=\"validation\",\n",
    "                        shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From D:\\anaconda\\lib\\site-packages\\tensorflow\\python\\framework\\op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n"
     ]
    }
   ],
   "source": [
    "import keras\n",
    "from keras.models import Model, load_model\n",
    "from keras.layers import Activation, Dropout, Flatten, Dense\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.applications.inception_v3 import InceptionV3\n",
    "\n",
    "#Load the VGG model\n",
    "base_model = InceptionV3(weights='imagenet', include_top=False, input_shape=(image_size, image_size, 3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_batch, y_batch = train_generator.next()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 24 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig=plt.figure()\n",
    "columns = 5\n",
    "rows = 5\n",
    "for i in range(1, columns*rows):\n",
    "    num = np.random.randint(batch_size)\n",
    "    image = x_batch[num].astype(np.int)\n",
    "    fig.add_subplot(rows, columns, i)\n",
    "    plt.imshow(image)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_2 (InputLayer)            (None, 128, 128, 3)  0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_95 (Conv2D)              (None, 63, 63, 32)   864         input_2[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_95 (BatchNo (None, 63, 63, 32)   96          conv2d_95[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_95 (Activation)      (None, 63, 63, 32)   0           batch_normalization_95[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_96 (Conv2D)              (None, 61, 61, 32)   9216        activation_95[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_96 (BatchNo (None, 61, 61, 32)   96          conv2d_96[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_96 (Activation)      (None, 61, 61, 32)   0           batch_normalization_96[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_97 (Conv2D)              (None, 61, 61, 64)   18432       activation_96[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_97 (BatchNo (None, 61, 61, 64)   192         conv2d_97[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_97 (Activation)      (None, 61, 61, 64)   0           batch_normalization_97[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_5 (MaxPooling2D)  (None, 30, 30, 64)   0           activation_97[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_98 (Conv2D)              (None, 30, 30, 80)   5120        max_pooling2d_5[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_98 (BatchNo (None, 30, 30, 80)   240         conv2d_98[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_98 (Activation)      (None, 30, 30, 80)   0           batch_normalization_98[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_99 (Conv2D)              (None, 28, 28, 192)  138240      activation_98[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_99 (BatchNo (None, 28, 28, 192)  576         conv2d_99[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_99 (Activation)      (None, 28, 28, 192)  0           batch_normalization_99[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_6 (MaxPooling2D)  (None, 13, 13, 192)  0           activation_99[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_103 (Conv2D)             (None, 13, 13, 64)   12288       max_pooling2d_6[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_103 (BatchN (None, 13, 13, 64)   192         conv2d_103[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_103 (Activation)     (None, 13, 13, 64)   0           batch_normalization_103[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_101 (Conv2D)             (None, 13, 13, 48)   9216        max_pooling2d_6[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_104 (Conv2D)             (None, 13, 13, 96)   55296       activation_103[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_101 (BatchN (None, 13, 13, 48)   144         conv2d_101[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_104 (BatchN (None, 13, 13, 96)   288         conv2d_104[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_101 (Activation)     (None, 13, 13, 48)   0           batch_normalization_101[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_104 (Activation)     (None, 13, 13, 96)   0           batch_normalization_104[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_10 (AveragePo (None, 13, 13, 192)  0           max_pooling2d_6[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_100 (Conv2D)             (None, 13, 13, 64)   12288       max_pooling2d_6[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_102 (Conv2D)             (None, 13, 13, 64)   76800       activation_101[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_105 (Conv2D)             (None, 13, 13, 96)   82944       activation_104[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_106 (Conv2D)             (None, 13, 13, 32)   6144        average_pooling2d_10[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_100 (BatchN (None, 13, 13, 64)   192         conv2d_100[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_102 (BatchN (None, 13, 13, 64)   192         conv2d_102[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_105 (BatchN (None, 13, 13, 96)   288         conv2d_105[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_106 (BatchN (None, 13, 13, 32)   96          conv2d_106[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_100 (Activation)     (None, 13, 13, 64)   0           batch_normalization_100[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_102 (Activation)     (None, 13, 13, 64)   0           batch_normalization_102[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_105 (Activation)     (None, 13, 13, 96)   0           batch_normalization_105[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_106 (Activation)     (None, 13, 13, 32)   0           batch_normalization_106[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "mixed0 (Concatenate)            (None, 13, 13, 256)  0           activation_100[0][0]             \n",
      "                                                                 activation_102[0][0]             \n",
      "                                                                 activation_105[0][0]             \n",
      "                                                                 activation_106[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_110 (Conv2D)             (None, 13, 13, 64)   16384       mixed0[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_110 (BatchN (None, 13, 13, 64)   192         conv2d_110[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_110 (Activation)     (None, 13, 13, 64)   0           batch_normalization_110[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_108 (Conv2D)             (None, 13, 13, 48)   12288       mixed0[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_111 (Conv2D)             (None, 13, 13, 96)   55296       activation_110[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_108 (BatchN (None, 13, 13, 48)   144         conv2d_108[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_111 (BatchN (None, 13, 13, 96)   288         conv2d_111[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_108 (Activation)     (None, 13, 13, 48)   0           batch_normalization_108[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_111 (Activation)     (None, 13, 13, 96)   0           batch_normalization_111[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_11 (AveragePo (None, 13, 13, 256)  0           mixed0[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_107 (Conv2D)             (None, 13, 13, 64)   16384       mixed0[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_109 (Conv2D)             (None, 13, 13, 64)   76800       activation_108[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_112 (Conv2D)             (None, 13, 13, 96)   82944       activation_111[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_113 (Conv2D)             (None, 13, 13, 64)   16384       average_pooling2d_11[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_107 (BatchN (None, 13, 13, 64)   192         conv2d_107[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_109 (BatchN (None, 13, 13, 64)   192         conv2d_109[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_112 (BatchN (None, 13, 13, 96)   288         conv2d_112[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_113 (BatchN (None, 13, 13, 64)   192         conv2d_113[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_107 (Activation)     (None, 13, 13, 64)   0           batch_normalization_107[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_109 (Activation)     (None, 13, 13, 64)   0           batch_normalization_109[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_112 (Activation)     (None, 13, 13, 96)   0           batch_normalization_112[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_113 (Activation)     (None, 13, 13, 64)   0           batch_normalization_113[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "mixed1 (Concatenate)            (None, 13, 13, 288)  0           activation_107[0][0]             \n",
      "                                                                 activation_109[0][0]             \n",
      "                                                                 activation_112[0][0]             \n",
      "                                                                 activation_113[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_117 (Conv2D)             (None, 13, 13, 64)   18432       mixed1[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_117 (BatchN (None, 13, 13, 64)   192         conv2d_117[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_117 (Activation)     (None, 13, 13, 64)   0           batch_normalization_117[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_115 (Conv2D)             (None, 13, 13, 48)   13824       mixed1[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_118 (Conv2D)             (None, 13, 13, 96)   55296       activation_117[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_115 (BatchN (None, 13, 13, 48)   144         conv2d_115[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_118 (BatchN (None, 13, 13, 96)   288         conv2d_118[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_115 (Activation)     (None, 13, 13, 48)   0           batch_normalization_115[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_118 (Activation)     (None, 13, 13, 96)   0           batch_normalization_118[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_12 (AveragePo (None, 13, 13, 288)  0           mixed1[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_114 (Conv2D)             (None, 13, 13, 64)   18432       mixed1[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_116 (Conv2D)             (None, 13, 13, 64)   76800       activation_115[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_119 (Conv2D)             (None, 13, 13, 96)   82944       activation_118[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_120 (Conv2D)             (None, 13, 13, 64)   18432       average_pooling2d_12[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_114 (BatchN (None, 13, 13, 64)   192         conv2d_114[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_116 (BatchN (None, 13, 13, 64)   192         conv2d_116[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_119 (BatchN (None, 13, 13, 96)   288         conv2d_119[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_120 (BatchN (None, 13, 13, 64)   192         conv2d_120[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_114 (Activation)     (None, 13, 13, 64)   0           batch_normalization_114[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_116 (Activation)     (None, 13, 13, 64)   0           batch_normalization_116[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_119 (Activation)     (None, 13, 13, 96)   0           batch_normalization_119[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_120 (Activation)     (None, 13, 13, 64)   0           batch_normalization_120[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "mixed2 (Concatenate)            (None, 13, 13, 288)  0           activation_114[0][0]             \n",
      "                                                                 activation_116[0][0]             \n",
      "                                                                 activation_119[0][0]             \n",
      "                                                                 activation_120[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_122 (Conv2D)             (None, 13, 13, 64)   18432       mixed2[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_122 (BatchN (None, 13, 13, 64)   192         conv2d_122[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_122 (Activation)     (None, 13, 13, 64)   0           batch_normalization_122[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_123 (Conv2D)             (None, 13, 13, 96)   55296       activation_122[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_123 (BatchN (None, 13, 13, 96)   288         conv2d_123[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_123 (Activation)     (None, 13, 13, 96)   0           batch_normalization_123[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_121 (Conv2D)             (None, 6, 6, 384)    995328      mixed2[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_124 (Conv2D)             (None, 6, 6, 96)     82944       activation_123[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_121 (BatchN (None, 6, 6, 384)    1152        conv2d_121[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_124 (BatchN (None, 6, 6, 96)     288         conv2d_124[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_121 (Activation)     (None, 6, 6, 384)    0           batch_normalization_121[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_124 (Activation)     (None, 6, 6, 96)     0           batch_normalization_124[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_7 (MaxPooling2D)  (None, 6, 6, 288)    0           mixed2[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "mixed3 (Concatenate)            (None, 6, 6, 768)    0           activation_121[0][0]             \n",
      "                                                                 activation_124[0][0]             \n",
      "                                                                 max_pooling2d_7[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_129 (Conv2D)             (None, 6, 6, 128)    98304       mixed3[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_129 (BatchN (None, 6, 6, 128)    384         conv2d_129[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_129 (Activation)     (None, 6, 6, 128)    0           batch_normalization_129[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_130 (Conv2D)             (None, 6, 6, 128)    114688      activation_129[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_130 (BatchN (None, 6, 6, 128)    384         conv2d_130[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_130 (Activation)     (None, 6, 6, 128)    0           batch_normalization_130[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_126 (Conv2D)             (None, 6, 6, 128)    98304       mixed3[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_131 (Conv2D)             (None, 6, 6, 128)    114688      activation_130[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_126 (BatchN (None, 6, 6, 128)    384         conv2d_126[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_131 (BatchN (None, 6, 6, 128)    384         conv2d_131[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_126 (Activation)     (None, 6, 6, 128)    0           batch_normalization_126[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_131 (Activation)     (None, 6, 6, 128)    0           batch_normalization_131[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_127 (Conv2D)             (None, 6, 6, 128)    114688      activation_126[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_132 (Conv2D)             (None, 6, 6, 128)    114688      activation_131[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_127 (BatchN (None, 6, 6, 128)    384         conv2d_127[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_132 (BatchN (None, 6, 6, 128)    384         conv2d_132[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_127 (Activation)     (None, 6, 6, 128)    0           batch_normalization_127[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_132 (Activation)     (None, 6, 6, 128)    0           batch_normalization_132[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_13 (AveragePo (None, 6, 6, 768)    0           mixed3[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_125 (Conv2D)             (None, 6, 6, 192)    147456      mixed3[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_128 (Conv2D)             (None, 6, 6, 192)    172032      activation_127[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_133 (Conv2D)             (None, 6, 6, 192)    172032      activation_132[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_134 (Conv2D)             (None, 6, 6, 192)    147456      average_pooling2d_13[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_125 (BatchN (None, 6, 6, 192)    576         conv2d_125[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_128 (BatchN (None, 6, 6, 192)    576         conv2d_128[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_133 (BatchN (None, 6, 6, 192)    576         conv2d_133[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_134 (BatchN (None, 6, 6, 192)    576         conv2d_134[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_125 (Activation)     (None, 6, 6, 192)    0           batch_normalization_125[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_128 (Activation)     (None, 6, 6, 192)    0           batch_normalization_128[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_133 (Activation)     (None, 6, 6, 192)    0           batch_normalization_133[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_134 (Activation)     (None, 6, 6, 192)    0           batch_normalization_134[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "mixed4 (Concatenate)            (None, 6, 6, 768)    0           activation_125[0][0]             \n",
      "                                                                 activation_128[0][0]             \n",
      "                                                                 activation_133[0][0]             \n",
      "                                                                 activation_134[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_139 (Conv2D)             (None, 6, 6, 160)    122880      mixed4[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_139 (BatchN (None, 6, 6, 160)    480         conv2d_139[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_139 (Activation)     (None, 6, 6, 160)    0           batch_normalization_139[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_140 (Conv2D)             (None, 6, 6, 160)    179200      activation_139[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_140 (BatchN (None, 6, 6, 160)    480         conv2d_140[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_140 (Activation)     (None, 6, 6, 160)    0           batch_normalization_140[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_136 (Conv2D)             (None, 6, 6, 160)    122880      mixed4[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_141 (Conv2D)             (None, 6, 6, 160)    179200      activation_140[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_136 (BatchN (None, 6, 6, 160)    480         conv2d_136[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_141 (BatchN (None, 6, 6, 160)    480         conv2d_141[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_136 (Activation)     (None, 6, 6, 160)    0           batch_normalization_136[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_141 (Activation)     (None, 6, 6, 160)    0           batch_normalization_141[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_137 (Conv2D)             (None, 6, 6, 160)    179200      activation_136[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_142 (Conv2D)             (None, 6, 6, 160)    179200      activation_141[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_137 (BatchN (None, 6, 6, 160)    480         conv2d_137[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_142 (BatchN (None, 6, 6, 160)    480         conv2d_142[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_137 (Activation)     (None, 6, 6, 160)    0           batch_normalization_137[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_142 (Activation)     (None, 6, 6, 160)    0           batch_normalization_142[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_14 (AveragePo (None, 6, 6, 768)    0           mixed4[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_135 (Conv2D)             (None, 6, 6, 192)    147456      mixed4[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_138 (Conv2D)             (None, 6, 6, 192)    215040      activation_137[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_143 (Conv2D)             (None, 6, 6, 192)    215040      activation_142[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_144 (Conv2D)             (None, 6, 6, 192)    147456      average_pooling2d_14[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_135 (BatchN (None, 6, 6, 192)    576         conv2d_135[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_138 (BatchN (None, 6, 6, 192)    576         conv2d_138[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_143 (BatchN (None, 6, 6, 192)    576         conv2d_143[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_144 (BatchN (None, 6, 6, 192)    576         conv2d_144[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_135 (Activation)     (None, 6, 6, 192)    0           batch_normalization_135[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_138 (Activation)     (None, 6, 6, 192)    0           batch_normalization_138[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_143 (Activation)     (None, 6, 6, 192)    0           batch_normalization_143[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_144 (Activation)     (None, 6, 6, 192)    0           batch_normalization_144[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "mixed5 (Concatenate)            (None, 6, 6, 768)    0           activation_135[0][0]             \n",
      "                                                                 activation_138[0][0]             \n",
      "                                                                 activation_143[0][0]             \n",
      "                                                                 activation_144[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_149 (Conv2D)             (None, 6, 6, 160)    122880      mixed5[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_149 (BatchN (None, 6, 6, 160)    480         conv2d_149[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_149 (Activation)     (None, 6, 6, 160)    0           batch_normalization_149[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_150 (Conv2D)             (None, 6, 6, 160)    179200      activation_149[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_150 (BatchN (None, 6, 6, 160)    480         conv2d_150[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_150 (Activation)     (None, 6, 6, 160)    0           batch_normalization_150[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_146 (Conv2D)             (None, 6, 6, 160)    122880      mixed5[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_151 (Conv2D)             (None, 6, 6, 160)    179200      activation_150[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_146 (BatchN (None, 6, 6, 160)    480         conv2d_146[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_151 (BatchN (None, 6, 6, 160)    480         conv2d_151[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_146 (Activation)     (None, 6, 6, 160)    0           batch_normalization_146[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_151 (Activation)     (None, 6, 6, 160)    0           batch_normalization_151[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_147 (Conv2D)             (None, 6, 6, 160)    179200      activation_146[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_152 (Conv2D)             (None, 6, 6, 160)    179200      activation_151[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_147 (BatchN (None, 6, 6, 160)    480         conv2d_147[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_152 (BatchN (None, 6, 6, 160)    480         conv2d_152[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_147 (Activation)     (None, 6, 6, 160)    0           batch_normalization_147[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_152 (Activation)     (None, 6, 6, 160)    0           batch_normalization_152[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_15 (AveragePo (None, 6, 6, 768)    0           mixed5[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_145 (Conv2D)             (None, 6, 6, 192)    147456      mixed5[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_148 (Conv2D)             (None, 6, 6, 192)    215040      activation_147[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_153 (Conv2D)             (None, 6, 6, 192)    215040      activation_152[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_154 (Conv2D)             (None, 6, 6, 192)    147456      average_pooling2d_15[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_145 (BatchN (None, 6, 6, 192)    576         conv2d_145[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_148 (BatchN (None, 6, 6, 192)    576         conv2d_148[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_153 (BatchN (None, 6, 6, 192)    576         conv2d_153[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_154 (BatchN (None, 6, 6, 192)    576         conv2d_154[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_145 (Activation)     (None, 6, 6, 192)    0           batch_normalization_145[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_148 (Activation)     (None, 6, 6, 192)    0           batch_normalization_148[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_153 (Activation)     (None, 6, 6, 192)    0           batch_normalization_153[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_154 (Activation)     (None, 6, 6, 192)    0           batch_normalization_154[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "mixed6 (Concatenate)            (None, 6, 6, 768)    0           activation_145[0][0]             \n",
      "                                                                 activation_148[0][0]             \n",
      "                                                                 activation_153[0][0]             \n",
      "                                                                 activation_154[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_159 (Conv2D)             (None, 6, 6, 192)    147456      mixed6[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_159 (BatchN (None, 6, 6, 192)    576         conv2d_159[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_159 (Activation)     (None, 6, 6, 192)    0           batch_normalization_159[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_160 (Conv2D)             (None, 6, 6, 192)    258048      activation_159[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_160 (BatchN (None, 6, 6, 192)    576         conv2d_160[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_160 (Activation)     (None, 6, 6, 192)    0           batch_normalization_160[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_156 (Conv2D)             (None, 6, 6, 192)    147456      mixed6[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_161 (Conv2D)             (None, 6, 6, 192)    258048      activation_160[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_156 (BatchN (None, 6, 6, 192)    576         conv2d_156[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_161 (BatchN (None, 6, 6, 192)    576         conv2d_161[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_156 (Activation)     (None, 6, 6, 192)    0           batch_normalization_156[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_161 (Activation)     (None, 6, 6, 192)    0           batch_normalization_161[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_157 (Conv2D)             (None, 6, 6, 192)    258048      activation_156[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_162 (Conv2D)             (None, 6, 6, 192)    258048      activation_161[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_157 (BatchN (None, 6, 6, 192)    576         conv2d_157[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_162 (BatchN (None, 6, 6, 192)    576         conv2d_162[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_157 (Activation)     (None, 6, 6, 192)    0           batch_normalization_157[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_162 (Activation)     (None, 6, 6, 192)    0           batch_normalization_162[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_16 (AveragePo (None, 6, 6, 768)    0           mixed6[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_155 (Conv2D)             (None, 6, 6, 192)    147456      mixed6[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_158 (Conv2D)             (None, 6, 6, 192)    258048      activation_157[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_163 (Conv2D)             (None, 6, 6, 192)    258048      activation_162[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_164 (Conv2D)             (None, 6, 6, 192)    147456      average_pooling2d_16[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_155 (BatchN (None, 6, 6, 192)    576         conv2d_155[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_158 (BatchN (None, 6, 6, 192)    576         conv2d_158[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_163 (BatchN (None, 6, 6, 192)    576         conv2d_163[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_164 (BatchN (None, 6, 6, 192)    576         conv2d_164[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_155 (Activation)     (None, 6, 6, 192)    0           batch_normalization_155[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_158 (Activation)     (None, 6, 6, 192)    0           batch_normalization_158[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_163 (Activation)     (None, 6, 6, 192)    0           batch_normalization_163[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_164 (Activation)     (None, 6, 6, 192)    0           batch_normalization_164[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "mixed7 (Concatenate)            (None, 6, 6, 768)    0           activation_155[0][0]             \n",
      "                                                                 activation_158[0][0]             \n",
      "                                                                 activation_163[0][0]             \n",
      "                                                                 activation_164[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_167 (Conv2D)             (None, 6, 6, 192)    147456      mixed7[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_167 (BatchN (None, 6, 6, 192)    576         conv2d_167[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_167 (Activation)     (None, 6, 6, 192)    0           batch_normalization_167[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_168 (Conv2D)             (None, 6, 6, 192)    258048      activation_167[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_168 (BatchN (None, 6, 6, 192)    576         conv2d_168[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_168 (Activation)     (None, 6, 6, 192)    0           batch_normalization_168[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_165 (Conv2D)             (None, 6, 6, 192)    147456      mixed7[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_169 (Conv2D)             (None, 6, 6, 192)    258048      activation_168[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_165 (BatchN (None, 6, 6, 192)    576         conv2d_165[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_169 (BatchN (None, 6, 6, 192)    576         conv2d_169[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_165 (Activation)     (None, 6, 6, 192)    0           batch_normalization_165[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_169 (Activation)     (None, 6, 6, 192)    0           batch_normalization_169[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_166 (Conv2D)             (None, 2, 2, 320)    552960      activation_165[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_170 (Conv2D)             (None, 2, 2, 192)    331776      activation_169[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_166 (BatchN (None, 2, 2, 320)    960         conv2d_166[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_170 (BatchN (None, 2, 2, 192)    576         conv2d_170[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_166 (Activation)     (None, 2, 2, 320)    0           batch_normalization_166[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_170 (Activation)     (None, 2, 2, 192)    0           batch_normalization_170[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_8 (MaxPooling2D)  (None, 2, 2, 768)    0           mixed7[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "mixed8 (Concatenate)            (None, 2, 2, 1280)   0           activation_166[0][0]             \n",
      "                                                                 activation_170[0][0]             \n",
      "                                                                 max_pooling2d_8[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_175 (Conv2D)             (None, 2, 2, 448)    573440      mixed8[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_175 (BatchN (None, 2, 2, 448)    1344        conv2d_175[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_175 (Activation)     (None, 2, 2, 448)    0           batch_normalization_175[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_172 (Conv2D)             (None, 2, 2, 384)    491520      mixed8[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_176 (Conv2D)             (None, 2, 2, 384)    1548288     activation_175[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_172 (BatchN (None, 2, 2, 384)    1152        conv2d_172[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_176 (BatchN (None, 2, 2, 384)    1152        conv2d_176[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_172 (Activation)     (None, 2, 2, 384)    0           batch_normalization_172[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_176 (Activation)     (None, 2, 2, 384)    0           batch_normalization_176[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_173 (Conv2D)             (None, 2, 2, 384)    442368      activation_172[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_174 (Conv2D)             (None, 2, 2, 384)    442368      activation_172[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_177 (Conv2D)             (None, 2, 2, 384)    442368      activation_176[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_178 (Conv2D)             (None, 2, 2, 384)    442368      activation_176[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_17 (AveragePo (None, 2, 2, 1280)   0           mixed8[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_171 (Conv2D)             (None, 2, 2, 320)    409600      mixed8[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_173 (BatchN (None, 2, 2, 384)    1152        conv2d_173[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_174 (BatchN (None, 2, 2, 384)    1152        conv2d_174[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_177 (BatchN (None, 2, 2, 384)    1152        conv2d_177[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_178 (BatchN (None, 2, 2, 384)    1152        conv2d_178[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_179 (Conv2D)             (None, 2, 2, 192)    245760      average_pooling2d_17[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_171 (BatchN (None, 2, 2, 320)    960         conv2d_171[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_173 (Activation)     (None, 2, 2, 384)    0           batch_normalization_173[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_174 (Activation)     (None, 2, 2, 384)    0           batch_normalization_174[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_177 (Activation)     (None, 2, 2, 384)    0           batch_normalization_177[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_178 (Activation)     (None, 2, 2, 384)    0           batch_normalization_178[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_179 (BatchN (None, 2, 2, 192)    576         conv2d_179[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_171 (Activation)     (None, 2, 2, 320)    0           batch_normalization_171[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "mixed9_0 (Concatenate)          (None, 2, 2, 768)    0           activation_173[0][0]             \n",
      "                                                                 activation_174[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "concatenate_3 (Concatenate)     (None, 2, 2, 768)    0           activation_177[0][0]             \n",
      "                                                                 activation_178[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_179 (Activation)     (None, 2, 2, 192)    0           batch_normalization_179[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "mixed9 (Concatenate)            (None, 2, 2, 2048)   0           activation_171[0][0]             \n",
      "                                                                 mixed9_0[0][0]                   \n",
      "                                                                 concatenate_3[0][0]              \n",
      "                                                                 activation_179[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_184 (Conv2D)             (None, 2, 2, 448)    917504      mixed9[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_184 (BatchN (None, 2, 2, 448)    1344        conv2d_184[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_184 (Activation)     (None, 2, 2, 448)    0           batch_normalization_184[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_181 (Conv2D)             (None, 2, 2, 384)    786432      mixed9[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_185 (Conv2D)             (None, 2, 2, 384)    1548288     activation_184[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_181 (BatchN (None, 2, 2, 384)    1152        conv2d_181[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_185 (BatchN (None, 2, 2, 384)    1152        conv2d_185[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_181 (Activation)     (None, 2, 2, 384)    0           batch_normalization_181[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_185 (Activation)     (None, 2, 2, 384)    0           batch_normalization_185[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_182 (Conv2D)             (None, 2, 2, 384)    442368      activation_181[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_183 (Conv2D)             (None, 2, 2, 384)    442368      activation_181[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_186 (Conv2D)             (None, 2, 2, 384)    442368      activation_185[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_187 (Conv2D)             (None, 2, 2, 384)    442368      activation_185[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_18 (AveragePo (None, 2, 2, 2048)   0           mixed9[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_180 (Conv2D)             (None, 2, 2, 320)    655360      mixed9[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_182 (BatchN (None, 2, 2, 384)    1152        conv2d_182[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_183 (BatchN (None, 2, 2, 384)    1152        conv2d_183[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_186 (BatchN (None, 2, 2, 384)    1152        conv2d_186[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_187 (BatchN (None, 2, 2, 384)    1152        conv2d_187[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_188 (Conv2D)             (None, 2, 2, 192)    393216      average_pooling2d_18[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_180 (BatchN (None, 2, 2, 320)    960         conv2d_180[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_182 (Activation)     (None, 2, 2, 384)    0           batch_normalization_182[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_183 (Activation)     (None, 2, 2, 384)    0           batch_normalization_183[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_186 (Activation)     (None, 2, 2, 384)    0           batch_normalization_186[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "activation_187 (Activation)     (None, 2, 2, 384)    0           batch_normalization_187[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_188 (BatchN (None, 2, 2, 192)    576         conv2d_188[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "activation_180 (Activation)     (None, 2, 2, 320)    0           batch_normalization_180[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "mixed9_1 (Concatenate)          (None, 2, 2, 768)    0           activation_182[0][0]             \n",
      "                                                                 activation_183[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "concatenate_4 (Concatenate)     (None, 2, 2, 768)    0           activation_186[0][0]             \n",
      "                                                                 activation_187[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "activation_188 (Activation)     (None, 2, 2, 192)    0           batch_normalization_188[0][0]    \n",
      "__________________________________________________________________________________________________\n",
      "mixed10 (Concatenate)           (None, 2, 2, 2048)   0           activation_180[0][0]             \n",
      "                                                                 mixed9_1[0][0]                   \n",
      "                                                                 concatenate_4[0][0]              \n",
      "                                                                 activation_188[0][0]             \n",
      "==================================================================================================\n",
      "Total params: 21,802,784\n",
      "Trainable params: 21,768,352\n",
      "Non-trainable params: 34,432\n",
      "__________________________________________________________________________________________________\n",
      "None\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "inception_v3 (Model)         (None, 2, 2, 2048)        21802784  \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 8192)              0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 1024)              8389632   \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1024)              1049600   \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 38)                38950     \n",
      "=================================================================\n",
      "Total params: 31,280,966\n",
      "Trainable params: 31,246,534\n",
      "Non-trainable params: 34,432\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "#Load the VGG model\n",
    "base_model = InceptionV3(weights='imagenet', include_top=False, input_shape=(image_size, image_size, 3))\n",
    "\n",
    "print(base_model.summary())\n",
    "\n",
    "    # Freeze the layers \n",
    "# for layer in base_model.layers:\n",
    "#     layer.trainable = True\n",
    " \n",
    "# # Create the model\n",
    "model = keras.models.Sequential()\n",
    "\n",
    "# # Add the vgg convolutional base model\n",
    "model.add(base_model)\n",
    " \n",
    "# # Add new layers\n",
    "model.add(Flatten())\n",
    "model.add(Dense(1024, activation='relu'))\n",
    "model.add(Dense(1024, activation='relu'))\n",
    "model.add(Dense(num_class, activation='softmax'))\n",
    " \n",
    "# # Show a summary of the model. Check the number of trainable parameters    \n",
    "print(model.summary())\n",
    "model.compile(loss='categorical_crossentropy',\n",
    "          optimizer='adam',\n",
    "          metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From D:\\anaconda\\lib\\site-packages\\tensorflow\\python\\ops\\math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Epoch 1/40\n",
      " 102/1358 [=>............................] - ETA: 34:51 - loss: 3.9112 - acc: 0.1042"
     ]
    }
   ],
   "source": [
    "model.fit_generator(\n",
    "    train_generator,\n",
    "    steps_per_epoch = train_generator.samples // batch_size,\n",
    "    validation_data = validation_generator, \n",
    "    validation_steps = validation_generator.samples // batch_size,\n",
    "    epochs = 40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import backend as K\n",
    "K.tensorflow_backend._get_available_gpus()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model saved\n"
     ]
    }
   ],
   "source": [
    "model.save('scratch_model.h5')\n",
    "print(\"Model saved\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!conda install h5py\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.metrics[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}