D_Feature_Extraction_NS_InceptionV3_1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "Entreno de un ConvNet InceptionV3 preentrenado con Imagenet y\n",
    "entrenando conjunto de datos pequeño\n",
    "Conjunto de datos de 4 clases\n",
    "Utilizaremos un conjunto de datos reducido:\n",
    "- Conjunto de entrenamiento: 1.000 muestras x clase \n",
    "- Conjunto de validación: 300 muestras x clase\n",
    "- Conjunto de test: 300 muestras x clase\n",
    "El conjunto de datos: balanced_dataset\n",
    "contiene tres carpetas, train, validation y test, cada una con 4 clases"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from keras.preprocessing.image  import ImageDataGenerator, img_to_array, load_img \n",
    "from keras.models  import Sequential\n",
    "from keras.layers  import Dropout,  Flatten,  Dense, GlobalAveragePooling2D, Input\n",
    "from keras.applications import InceptionV3\n",
    "from keras.optimizers import RMSprop\n",
    "from keras.utils.np_utils import to_categorical \n",
    "import time\n",
    "import os\n",
    "import numpy as np \n",
    "import math\n",
    "import  matplotlib.pyplot  as  plt\n",
    "\n",
    "width_image = 299\n",
    "height_image = 299\n",
    "batch_size = 64 \n",
    "epochs = 90\n",
    "nb_train = 4000\n",
    "nb_validation = 1200\n",
    "nb_test = 1200\n",
    "nb_classes = 4\n",
    "nb_FC = 1024\n",
    "channels_image = 3\n",
    "\n",
    "class_mode = 'categorical'\n",
    "train_dir  =  \"balanced_dataset/train\"\n",
    "validation_dir = \"balanced_dataset/validation\" \n",
    "test_dir  =  \"balanced_dataset/test\" \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.5/dist-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  from ._conv import register_converters as _register_converters\n",
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.applications import InceptionV3\n",
    "\n",
    "conv_base= InceptionV3(weights='imagenet', \n",
    "                 include_top = False, \n",
    "                 input_shape = (width_image, height_image, channels_image))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            (None, 299, 299, 3)  0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_1 (Conv2D)               (None, 149, 149, 32) 864         input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_1 (BatchNor (None, 149, 149, 32) 96          conv2d_1[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "activation_1 (Activation)       (None, 149, 149, 32) 0           batch_normalization_1[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_2 (Conv2D)               (None, 147, 147, 32) 9216        activation_1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_2 (BatchNor (None, 147, 147, 32) 96          conv2d_2[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "activation_2 (Activation)       (None, 147, 147, 32) 0           batch_normalization_2[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_3 (Conv2D)               (None, 147, 147, 64) 18432       activation_2[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_3 (BatchNor (None, 147, 147, 64) 192         conv2d_3[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "activation_3 (Activation)       (None, 147, 147, 64) 0           batch_normalization_3[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2D)  (None, 73, 73, 64)   0           activation_3[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_4 (Conv2D)               (None, 73, 73, 80)   5120        max_pooling2d_1[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_4 (BatchNor (None, 73, 73, 80)   240         conv2d_4[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "activation_4 (Activation)       (None, 73, 73, 80)   0           batch_normalization_4[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_5 (Conv2D)               (None, 71, 71, 192)  138240      activation_4[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_5 (BatchNor (None, 71, 71, 192)  576         conv2d_5[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "activation_5 (Activation)       (None, 71, 71, 192)  0           batch_normalization_5[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2D)  (None, 35, 35, 192)  0           activation_5[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_9 (Conv2D)               (None, 35, 35, 64)   12288       max_pooling2d_2[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_9 (BatchNor (None, 35, 35, 64)   192         conv2d_9[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "activation_9 (Activation)       (None, 35, 35, 64)   0           batch_normalization_9[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_7 (Conv2D)               (None, 35, 35, 48)   9216        max_pooling2d_2[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_10 (Conv2D)              (None, 35, 35, 96)   55296       activation_9[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_7 (BatchNor (None, 35, 35, 48)   144         conv2d_7[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_10 (BatchNo (None, 35, 35, 96)   288         conv2d_10[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_7 (Activation)       (None, 35, 35, 48)   0           batch_normalization_7[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "activation_10 (Activation)      (None, 35, 35, 96)   0           batch_normalization_10[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_1 (AveragePoo (None, 35, 35, 192)  0           max_pooling2d_2[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_6 (Conv2D)               (None, 35, 35, 64)   12288       max_pooling2d_2[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_8 (Conv2D)               (None, 35, 35, 64)   76800       activation_7[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_11 (Conv2D)              (None, 35, 35, 96)   82944       activation_10[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_12 (Conv2D)              (None, 35, 35, 32)   6144        average_pooling2d_1[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_6 (BatchNor (None, 35, 35, 64)   192         conv2d_6[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_8 (BatchNor (None, 35, 35, 64)   192         conv2d_8[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_11 (BatchNo (None, 35, 35, 96)   288         conv2d_11[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_12 (BatchNo (None, 35, 35, 32)   96          conv2d_12[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_6 (Activation)       (None, 35, 35, 64)   0           batch_normalization_6[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "activation_8 (Activation)       (None, 35, 35, 64)   0           batch_normalization_8[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "activation_11 (Activation)      (None, 35, 35, 96)   0           batch_normalization_11[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_12 (Activation)      (None, 35, 35, 32)   0           batch_normalization_12[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "mixed0 (Concatenate)            (None, 35, 35, 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, 35, 35, 64)   16384       mixed0[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_16 (BatchNo (None, 35, 35, 64)   192         conv2d_16[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_16 (Activation)      (None, 35, 35, 64)   0           batch_normalization_16[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_14 (Conv2D)              (None, 35, 35, 48)   12288       mixed0[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_17 (Conv2D)              (None, 35, 35, 96)   55296       activation_16[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_14 (BatchNo (None, 35, 35, 48)   144         conv2d_14[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_17 (BatchNo (None, 35, 35, 96)   288         conv2d_17[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_14 (Activation)      (None, 35, 35, 48)   0           batch_normalization_14[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_17 (Activation)      (None, 35, 35, 96)   0           batch_normalization_17[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_2 (AveragePoo (None, 35, 35, 256)  0           mixed0[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_13 (Conv2D)              (None, 35, 35, 64)   16384       mixed0[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_15 (Conv2D)              (None, 35, 35, 64)   76800       activation_14[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_18 (Conv2D)              (None, 35, 35, 96)   82944       activation_17[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_19 (Conv2D)              (None, 35, 35, 64)   16384       average_pooling2d_2[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_13 (BatchNo (None, 35, 35, 64)   192         conv2d_13[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_15 (BatchNo (None, 35, 35, 64)   192         conv2d_15[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_18 (BatchNo (None, 35, 35, 96)   288         conv2d_18[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_19 (BatchNo (None, 35, 35, 64)   192         conv2d_19[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_13 (Activation)      (None, 35, 35, 64)   0           batch_normalization_13[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_15 (Activation)      (None, 35, 35, 64)   0           batch_normalization_15[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_18 (Activation)      (None, 35, 35, 96)   0           batch_normalization_18[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_19 (Activation)      (None, 35, 35, 64)   0           batch_normalization_19[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "mixed1 (Concatenate)            (None, 35, 35, 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, 35, 35, 64)   18432       mixed1[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_23 (BatchNo (None, 35, 35, 64)   192         conv2d_23[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_23 (Activation)      (None, 35, 35, 64)   0           batch_normalization_23[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_21 (Conv2D)              (None, 35, 35, 48)   13824       mixed1[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_24 (Conv2D)              (None, 35, 35, 96)   55296       activation_23[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_21 (BatchNo (None, 35, 35, 48)   144         conv2d_21[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_24 (BatchNo (None, 35, 35, 96)   288         conv2d_24[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_21 (Activation)      (None, 35, 35, 48)   0           batch_normalization_21[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_24 (Activation)      (None, 35, 35, 96)   0           batch_normalization_24[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_3 (AveragePoo (None, 35, 35, 288)  0           mixed1[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_20 (Conv2D)              (None, 35, 35, 64)   18432       mixed1[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_22 (Conv2D)              (None, 35, 35, 64)   76800       activation_21[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_25 (Conv2D)              (None, 35, 35, 96)   82944       activation_24[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_26 (Conv2D)              (None, 35, 35, 64)   18432       average_pooling2d_3[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_20 (BatchNo (None, 35, 35, 64)   192         conv2d_20[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_22 (BatchNo (None, 35, 35, 64)   192         conv2d_22[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_25 (BatchNo (None, 35, 35, 96)   288         conv2d_25[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_26 (BatchNo (None, 35, 35, 64)   192         conv2d_26[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_20 (Activation)      (None, 35, 35, 64)   0           batch_normalization_20[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_22 (Activation)      (None, 35, 35, 64)   0           batch_normalization_22[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_25 (Activation)      (None, 35, 35, 96)   0           batch_normalization_25[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_26 (Activation)      (None, 35, 35, 64)   0           batch_normalization_26[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "mixed2 (Concatenate)            (None, 35, 35, 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, 35, 35, 64)   18432       mixed2[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_28 (BatchNo (None, 35, 35, 64)   192         conv2d_28[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_28 (Activation)      (None, 35, 35, 64)   0           batch_normalization_28[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_29 (Conv2D)              (None, 35, 35, 96)   55296       activation_28[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_29 (BatchNo (None, 35, 35, 96)   288         conv2d_29[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_29 (Activation)      (None, 35, 35, 96)   0           batch_normalization_29[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_27 (Conv2D)              (None, 17, 17, 384)  995328      mixed2[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_30 (Conv2D)              (None, 17, 17, 96)   82944       activation_29[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_27 (BatchNo (None, 17, 17, 384)  1152        conv2d_27[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_30 (BatchNo (None, 17, 17, 96)   288         conv2d_30[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_27 (Activation)      (None, 17, 17, 384)  0           batch_normalization_27[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_30 (Activation)      (None, 17, 17, 96)   0           batch_normalization_30[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_3 (MaxPooling2D)  (None, 17, 17, 288)  0           mixed2[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "mixed3 (Concatenate)            (None, 17, 17, 768)  0           activation_27[0][0]              \n",
      "                                                                 activation_30[0][0]              \n",
      "                                                                 max_pooling2d_3[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_35 (Conv2D)              (None, 17, 17, 128)  98304       mixed3[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_35 (BatchNo (None, 17, 17, 128)  384         conv2d_35[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_35 (Activation)      (None, 17, 17, 128)  0           batch_normalization_35[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_36 (Conv2D)              (None, 17, 17, 128)  114688      activation_35[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_36 (BatchNo (None, 17, 17, 128)  384         conv2d_36[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_36 (Activation)      (None, 17, 17, 128)  0           batch_normalization_36[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_32 (Conv2D)              (None, 17, 17, 128)  98304       mixed3[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_37 (Conv2D)              (None, 17, 17, 128)  114688      activation_36[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_32 (BatchNo (None, 17, 17, 128)  384         conv2d_32[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_37 (BatchNo (None, 17, 17, 128)  384         conv2d_37[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_32 (Activation)      (None, 17, 17, 128)  0           batch_normalization_32[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_37 (Activation)      (None, 17, 17, 128)  0           batch_normalization_37[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_33 (Conv2D)              (None, 17, 17, 128)  114688      activation_32[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_38 (Conv2D)              (None, 17, 17, 128)  114688      activation_37[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_33 (BatchNo (None, 17, 17, 128)  384         conv2d_33[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_38 (BatchNo (None, 17, 17, 128)  384         conv2d_38[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_33 (Activation)      (None, 17, 17, 128)  0           batch_normalization_33[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_38 (Activation)      (None, 17, 17, 128)  0           batch_normalization_38[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_4 (AveragePoo (None, 17, 17, 768)  0           mixed3[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_31 (Conv2D)              (None, 17, 17, 192)  147456      mixed3[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_34 (Conv2D)              (None, 17, 17, 192)  172032      activation_33[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_39 (Conv2D)              (None, 17, 17, 192)  172032      activation_38[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_40 (Conv2D)              (None, 17, 17, 192)  147456      average_pooling2d_4[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_31 (BatchNo (None, 17, 17, 192)  576         conv2d_31[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_34 (BatchNo (None, 17, 17, 192)  576         conv2d_34[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_39 (BatchNo (None, 17, 17, 192)  576         conv2d_39[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_40 (BatchNo (None, 17, 17, 192)  576         conv2d_40[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_31 (Activation)      (None, 17, 17, 192)  0           batch_normalization_31[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_34 (Activation)      (None, 17, 17, 192)  0           batch_normalization_34[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_39 (Activation)      (None, 17, 17, 192)  0           batch_normalization_39[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_40 (Activation)      (None, 17, 17, 192)  0           batch_normalization_40[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "mixed4 (Concatenate)            (None, 17, 17, 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, 17, 17, 160)  122880      mixed4[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_45 (BatchNo (None, 17, 17, 160)  480         conv2d_45[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_45 (Activation)      (None, 17, 17, 160)  0           batch_normalization_45[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_46 (Conv2D)              (None, 17, 17, 160)  179200      activation_45[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_46 (BatchNo (None, 17, 17, 160)  480         conv2d_46[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_46 (Activation)      (None, 17, 17, 160)  0           batch_normalization_46[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_42 (Conv2D)              (None, 17, 17, 160)  122880      mixed4[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_47 (Conv2D)              (None, 17, 17, 160)  179200      activation_46[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_42 (BatchNo (None, 17, 17, 160)  480         conv2d_42[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_47 (BatchNo (None, 17, 17, 160)  480         conv2d_47[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_42 (Activation)      (None, 17, 17, 160)  0           batch_normalization_42[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_47 (Activation)      (None, 17, 17, 160)  0           batch_normalization_47[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_43 (Conv2D)              (None, 17, 17, 160)  179200      activation_42[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_48 (Conv2D)              (None, 17, 17, 160)  179200      activation_47[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_43 (BatchNo (None, 17, 17, 160)  480         conv2d_43[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_48 (BatchNo (None, 17, 17, 160)  480         conv2d_48[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_43 (Activation)      (None, 17, 17, 160)  0           batch_normalization_43[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_48 (Activation)      (None, 17, 17, 160)  0           batch_normalization_48[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_5 (AveragePoo (None, 17, 17, 768)  0           mixed4[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_41 (Conv2D)              (None, 17, 17, 192)  147456      mixed4[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_44 (Conv2D)              (None, 17, 17, 192)  215040      activation_43[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_49 (Conv2D)              (None, 17, 17, 192)  215040      activation_48[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_50 (Conv2D)              (None, 17, 17, 192)  147456      average_pooling2d_5[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_41 (BatchNo (None, 17, 17, 192)  576         conv2d_41[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_44 (BatchNo (None, 17, 17, 192)  576         conv2d_44[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_49 (BatchNo (None, 17, 17, 192)  576         conv2d_49[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_50 (BatchNo (None, 17, 17, 192)  576         conv2d_50[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_41 (Activation)      (None, 17, 17, 192)  0           batch_normalization_41[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_44 (Activation)      (None, 17, 17, 192)  0           batch_normalization_44[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_49 (Activation)      (None, 17, 17, 192)  0           batch_normalization_49[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_50 (Activation)      (None, 17, 17, 192)  0           batch_normalization_50[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "mixed5 (Concatenate)            (None, 17, 17, 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, 17, 17, 160)  122880      mixed5[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_55 (BatchNo (None, 17, 17, 160)  480         conv2d_55[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_55 (Activation)      (None, 17, 17, 160)  0           batch_normalization_55[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_56 (Conv2D)              (None, 17, 17, 160)  179200      activation_55[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_56 (BatchNo (None, 17, 17, 160)  480         conv2d_56[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_56 (Activation)      (None, 17, 17, 160)  0           batch_normalization_56[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_52 (Conv2D)              (None, 17, 17, 160)  122880      mixed5[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_57 (Conv2D)              (None, 17, 17, 160)  179200      activation_56[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_52 (BatchNo (None, 17, 17, 160)  480         conv2d_52[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_57 (BatchNo (None, 17, 17, 160)  480         conv2d_57[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_52 (Activation)      (None, 17, 17, 160)  0           batch_normalization_52[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_57 (Activation)      (None, 17, 17, 160)  0           batch_normalization_57[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_53 (Conv2D)              (None, 17, 17, 160)  179200      activation_52[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_58 (Conv2D)              (None, 17, 17, 160)  179200      activation_57[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_53 (BatchNo (None, 17, 17, 160)  480         conv2d_53[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_58 (BatchNo (None, 17, 17, 160)  480         conv2d_58[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_53 (Activation)      (None, 17, 17, 160)  0           batch_normalization_53[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_58 (Activation)      (None, 17, 17, 160)  0           batch_normalization_58[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_6 (AveragePoo (None, 17, 17, 768)  0           mixed5[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_51 (Conv2D)              (None, 17, 17, 192)  147456      mixed5[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_54 (Conv2D)              (None, 17, 17, 192)  215040      activation_53[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_59 (Conv2D)              (None, 17, 17, 192)  215040      activation_58[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_60 (Conv2D)              (None, 17, 17, 192)  147456      average_pooling2d_6[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_51 (BatchNo (None, 17, 17, 192)  576         conv2d_51[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_54 (BatchNo (None, 17, 17, 192)  576         conv2d_54[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_59 (BatchNo (None, 17, 17, 192)  576         conv2d_59[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_60 (BatchNo (None, 17, 17, 192)  576         conv2d_60[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_51 (Activation)      (None, 17, 17, 192)  0           batch_normalization_51[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_54 (Activation)      (None, 17, 17, 192)  0           batch_normalization_54[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_59 (Activation)      (None, 17, 17, 192)  0           batch_normalization_59[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_60 (Activation)      (None, 17, 17, 192)  0           batch_normalization_60[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "mixed6 (Concatenate)            (None, 17, 17, 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, 17, 17, 192)  147456      mixed6[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_65 (BatchNo (None, 17, 17, 192)  576         conv2d_65[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_65 (Activation)      (None, 17, 17, 192)  0           batch_normalization_65[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_66 (Conv2D)              (None, 17, 17, 192)  258048      activation_65[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_66 (BatchNo (None, 17, 17, 192)  576         conv2d_66[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_66 (Activation)      (None, 17, 17, 192)  0           batch_normalization_66[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_62 (Conv2D)              (None, 17, 17, 192)  147456      mixed6[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_67 (Conv2D)              (None, 17, 17, 192)  258048      activation_66[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_62 (BatchNo (None, 17, 17, 192)  576         conv2d_62[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_67 (BatchNo (None, 17, 17, 192)  576         conv2d_67[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_62 (Activation)      (None, 17, 17, 192)  0           batch_normalization_62[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_67 (Activation)      (None, 17, 17, 192)  0           batch_normalization_67[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_63 (Conv2D)              (None, 17, 17, 192)  258048      activation_62[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_68 (Conv2D)              (None, 17, 17, 192)  258048      activation_67[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_63 (BatchNo (None, 17, 17, 192)  576         conv2d_63[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_68 (BatchNo (None, 17, 17, 192)  576         conv2d_68[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_63 (Activation)      (None, 17, 17, 192)  0           batch_normalization_63[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_68 (Activation)      (None, 17, 17, 192)  0           batch_normalization_68[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_7 (AveragePoo (None, 17, 17, 768)  0           mixed6[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_61 (Conv2D)              (None, 17, 17, 192)  147456      mixed6[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_64 (Conv2D)              (None, 17, 17, 192)  258048      activation_63[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_69 (Conv2D)              (None, 17, 17, 192)  258048      activation_68[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_70 (Conv2D)              (None, 17, 17, 192)  147456      average_pooling2d_7[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_61 (BatchNo (None, 17, 17, 192)  576         conv2d_61[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_64 (BatchNo (None, 17, 17, 192)  576         conv2d_64[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_69 (BatchNo (None, 17, 17, 192)  576         conv2d_69[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_70 (BatchNo (None, 17, 17, 192)  576         conv2d_70[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_61 (Activation)      (None, 17, 17, 192)  0           batch_normalization_61[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_64 (Activation)      (None, 17, 17, 192)  0           batch_normalization_64[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_69 (Activation)      (None, 17, 17, 192)  0           batch_normalization_69[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_70 (Activation)      (None, 17, 17, 192)  0           batch_normalization_70[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "mixed7 (Concatenate)            (None, 17, 17, 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, 17, 17, 192)  147456      mixed7[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_73 (BatchNo (None, 17, 17, 192)  576         conv2d_73[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_73 (Activation)      (None, 17, 17, 192)  0           batch_normalization_73[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_74 (Conv2D)              (None, 17, 17, 192)  258048      activation_73[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_74 (BatchNo (None, 17, 17, 192)  576         conv2d_74[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_74 (Activation)      (None, 17, 17, 192)  0           batch_normalization_74[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_71 (Conv2D)              (None, 17, 17, 192)  147456      mixed7[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_75 (Conv2D)              (None, 17, 17, 192)  258048      activation_74[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_71 (BatchNo (None, 17, 17, 192)  576         conv2d_71[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_75 (BatchNo (None, 17, 17, 192)  576         conv2d_75[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_71 (Activation)      (None, 17, 17, 192)  0           batch_normalization_71[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_75 (Activation)      (None, 17, 17, 192)  0           batch_normalization_75[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_72 (Conv2D)              (None, 8, 8, 320)    552960      activation_71[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_76 (Conv2D)              (None, 8, 8, 192)    331776      activation_75[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_72 (BatchNo (None, 8, 8, 320)    960         conv2d_72[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_76 (BatchNo (None, 8, 8, 192)    576         conv2d_76[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_72 (Activation)      (None, 8, 8, 320)    0           batch_normalization_72[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_76 (Activation)      (None, 8, 8, 192)    0           batch_normalization_76[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_4 (MaxPooling2D)  (None, 8, 8, 768)    0           mixed7[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "mixed8 (Concatenate)            (None, 8, 8, 1280)   0           activation_72[0][0]              \n",
      "                                                                 activation_76[0][0]              \n",
      "                                                                 max_pooling2d_4[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_81 (Conv2D)              (None, 8, 8, 448)    573440      mixed8[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_81 (BatchNo (None, 8, 8, 448)    1344        conv2d_81[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_81 (Activation)      (None, 8, 8, 448)    0           batch_normalization_81[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_78 (Conv2D)              (None, 8, 8, 384)    491520      mixed8[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_82 (Conv2D)              (None, 8, 8, 384)    1548288     activation_81[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_78 (BatchNo (None, 8, 8, 384)    1152        conv2d_78[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_82 (BatchNo (None, 8, 8, 384)    1152        conv2d_82[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_78 (Activation)      (None, 8, 8, 384)    0           batch_normalization_78[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_82 (Activation)      (None, 8, 8, 384)    0           batch_normalization_82[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_79 (Conv2D)              (None, 8, 8, 384)    442368      activation_78[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_80 (Conv2D)              (None, 8, 8, 384)    442368      activation_78[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_83 (Conv2D)              (None, 8, 8, 384)    442368      activation_82[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_84 (Conv2D)              (None, 8, 8, 384)    442368      activation_82[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_8 (AveragePoo (None, 8, 8, 1280)   0           mixed8[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_77 (Conv2D)              (None, 8, 8, 320)    409600      mixed8[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_79 (BatchNo (None, 8, 8, 384)    1152        conv2d_79[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_80 (BatchNo (None, 8, 8, 384)    1152        conv2d_80[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_83 (BatchNo (None, 8, 8, 384)    1152        conv2d_83[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_84 (BatchNo (None, 8, 8, 384)    1152        conv2d_84[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_85 (Conv2D)              (None, 8, 8, 192)    245760      average_pooling2d_8[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_77 (BatchNo (None, 8, 8, 320)    960         conv2d_77[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_79 (Activation)      (None, 8, 8, 384)    0           batch_normalization_79[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_80 (Activation)      (None, 8, 8, 384)    0           batch_normalization_80[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_83 (Activation)      (None, 8, 8, 384)    0           batch_normalization_83[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_84 (Activation)      (None, 8, 8, 384)    0           batch_normalization_84[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_85 (BatchNo (None, 8, 8, 192)    576         conv2d_85[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_77 (Activation)      (None, 8, 8, 320)    0           batch_normalization_77[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "mixed9_0 (Concatenate)          (None, 8, 8, 768)    0           activation_79[0][0]              \n",
      "                                                                 activation_80[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "concatenate_1 (Concatenate)     (None, 8, 8, 768)    0           activation_83[0][0]              \n",
      "                                                                 activation_84[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_85 (Activation)      (None, 8, 8, 192)    0           batch_normalization_85[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "mixed9 (Concatenate)            (None, 8, 8, 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, 8, 8, 448)    917504      mixed9[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_90 (BatchNo (None, 8, 8, 448)    1344        conv2d_90[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_90 (Activation)      (None, 8, 8, 448)    0           batch_normalization_90[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_87 (Conv2D)              (None, 8, 8, 384)    786432      mixed9[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_91 (Conv2D)              (None, 8, 8, 384)    1548288     activation_90[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_87 (BatchNo (None, 8, 8, 384)    1152        conv2d_87[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_91 (BatchNo (None, 8, 8, 384)    1152        conv2d_91[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_87 (Activation)      (None, 8, 8, 384)    0           batch_normalization_87[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_91 (Activation)      (None, 8, 8, 384)    0           batch_normalization_91[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_88 (Conv2D)              (None, 8, 8, 384)    442368      activation_87[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_89 (Conv2D)              (None, 8, 8, 384)    442368      activation_87[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_92 (Conv2D)              (None, 8, 8, 384)    442368      activation_91[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_93 (Conv2D)              (None, 8, 8, 384)    442368      activation_91[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "average_pooling2d_9 (AveragePoo (None, 8, 8, 2048)   0           mixed9[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_86 (Conv2D)              (None, 8, 8, 320)    655360      mixed9[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_88 (BatchNo (None, 8, 8, 384)    1152        conv2d_88[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_89 (BatchNo (None, 8, 8, 384)    1152        conv2d_89[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_92 (BatchNo (None, 8, 8, 384)    1152        conv2d_92[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_93 (BatchNo (None, 8, 8, 384)    1152        conv2d_93[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_94 (Conv2D)              (None, 8, 8, 192)    393216      average_pooling2d_9[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_86 (BatchNo (None, 8, 8, 320)    960         conv2d_86[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_88 (Activation)      (None, 8, 8, 384)    0           batch_normalization_88[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_89 (Activation)      (None, 8, 8, 384)    0           batch_normalization_89[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_92 (Activation)      (None, 8, 8, 384)    0           batch_normalization_92[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "activation_93 (Activation)      (None, 8, 8, 384)    0           batch_normalization_93[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_94 (BatchNo (None, 8, 8, 192)    576         conv2d_94[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_86 (Activation)      (None, 8, 8, 320)    0           batch_normalization_86[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "mixed9_1 (Concatenate)          (None, 8, 8, 768)    0           activation_88[0][0]              \n",
      "                                                                 activation_89[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "concatenate_2 (Concatenate)     (None, 8, 8, 768)    0           activation_92[0][0]              \n",
      "                                                                 activation_93[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "activation_94 (Activation)      (None, 8, 8, 192)    0           batch_normalization_94[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "mixed10 (Concatenate)           (None, 8, 8, 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": [
    "conv_base.summary()"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "FEATURE EXTRACTION (Extracción de características)\n",
    "\n",
    "Ahora hay dos formas de proceder:\n",
    "1- Ejecutar la base convolucional sobre el conjunto de datos, registrar la salida en una matriz Numpy en disco, y luego usar esa información como entrada para un clasificador independiente y densamente conectado. Esta solución es rápida y económica computacionalmente, pero no permite utilizar Data Augmentation.\n",
    "La primera parte es la base convolucional de un modelo preentrenado, segundo se entrena la parte convolucional con los nuevos datos de entrenamiento, depués añadimos un nuevo clasificador densamente conectado en la parte superior (final) y volvemos a entrenar. Si el modelo con el que se entrenó el modelo preentrenado es muy difernte del que queremos aplicar, es conveniente no utilizar toda la base convolucional (eliminar las últimas) ya que éstas són cada vez más especificas al conjunto de datos con el que se entrenó.\n",
    "2- Amplicar el modelo que tenemos (conv_base) agregando capas densas en la parte superior y ejecutando todo de punta a punta en los datos de entrada. Esto permite Data Augmentation, porque cada imagen de entrada pasa por la base convolucional. Esta técnica es mucho mas costosa y requiere GPU"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "1. Extracción de características sin Data Augmentation --> CPU\n",
    "\n",
    "Ejecutamos instancias de ImageDataGenerator para extraer imagenes como matrices Numpy y sus etiquetas. Extraeremos características de estas imágenes llamando al método del modelo conv_base\n",
    "\n",
    "- La ultima salida de la capa convolucional es de (8, 8, 2048 )\n",
    "out_x, out_y, conv_len"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 4000 images belonging to 4 classes.\n",
      "Found 1200 images belonging to 4 classes.\n",
      "Found 1200 images belonging to 4 classes.\n"
     ]
    }
   ],
   "source": [
    "#Extracción de características del modelo Preentrenado y nuestro dataset\n",
    "import numpy as np\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.utils import to_categorical\n",
    "\n",
    "#Tamaño de salida de la última capa convoluciónal del modelo preentrenado\n",
    "#lo vemos en el conv_base.summary()\n",
    "out_x = 8\n",
    "out_y = 8\n",
    "conv_len = 2048\n",
    "\n",
    "datagen = ImageDataGenerator(rescale = 1./255)\n",
    "\n",
    "def extract_features(directory, sample_count):\n",
    "    features = np.zeros(shape=(sample_count, out_x, out_x, conv_len))\n",
    "    labels = to_categorical(np.zeros(shape=(sample_count)),nb_classes) #INDICAR NUMERO DE CLASES\n",
    "    generator = datagen.flow_from_directory(\n",
    "        directory,\n",
    "        target_size = (height_image, width_image),\n",
    "        batch_size = batch_size,\n",
    "        #classes = 4,\n",
    "        class_mode = 'categorical')\n",
    "    i = 0\n",
    "    for inputs_batch, labels_batch in generator:\n",
    "        features_batch = conv_base.predict(inputs_batch) ##Asociamos al modelo preentrenado\n",
    "        features[i * batch_size : (i + 1) * batch_size] = features_batch\n",
    "        labels[i * batch_size : (i + 1) * batch_size] = labels_batch\n",
    "        i += 1\n",
    "        if i * batch_size >= sample_count:\n",
    "            break\n",
    "    return features, labels\n",
    "\n",
    "#Train: 1000 muestras x clase, con 4 clases, 1000 x 4 = 4000\n",
    "train_features, train_labels = extract_features(train_dir, nb_train)\n",
    "#validation 500 muestras x clase con 4 clases, 300 x 4 = 1200\n",
    "validation_features, validation_labels = extract_features(validation_dir, nb_validation) \n",
    "#test 500 muestras x clase con 4 clases, 300 x 4 = 1200\n",
    "test_features, test_labels = extract_features(test_dir, nb_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "np.save('InceptionV3/InceptionV3_FE_NS_train_1.npy',  train_features)\n",
    "np.save('InceptionV3/InceptionV3_FE_NS_validation_1.npy',  validation_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4000, 8, 8, 2048)\n",
      "(1200, 8, 8, 2048)\n",
      "(1200, 8, 8, 2048)\n"
     ]
    }
   ],
   "source": [
    "print(train_features.shape)\n",
    "print(validation_features.shape)\n",
    "print(test_features.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Save Features\n",
    "train_data = np.load('InceptionV3/InceptionV3_FE_NS_train_1.npy') \n",
    "validation_data  =  np.load('InceptionV3/InceptionV3_FE_NS_validation_1.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4000, 51200)\n",
      "(1200, 51200)\n",
      "(1200, 51200)\n"
     ]
    }
   ],
   "source": [
    "#Las características extraídas están como muetras en forma (muestra, 4, 4, 512).-->(muestras, out_x, out_y, conv_len)\n",
    "#Debemos aplanarlas (muestras, 8192) para alimentar un clasificador densamente conectado 4x4x512 = 8192 si entrada 150x150\n",
    "#Debemos aplanarlas (muestras, 25088) para alimentar un clasificador densamente conectado 7x7x512 = 25088 si entrada 224x224\n",
    "\n",
    "\n",
    "train_features = np.reshape(train_features, (nb_train, out_x * out_y * conv_len))\n",
    "validation_features = np.reshape(validation_features, (nb_validation, out_x * out_y * conv_len))\n",
    "test_features = np.reshape(test_features, (nb_test, out_x * out_y * conv_len))\n",
    "print(train_features.shape)\n",
    "print(validation_features.shape)\n",
    "print(test_features.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from keras.models  import Model\n",
    "# add a global spatial average pooling layer\n",
    "inputTensor = Input((8,8,2048)) #Use (None,None,2048) if bottlenecks vary in size    \n",
    "x = GlobalAveragePooling2D()(inputTensor)\n",
    "\n",
    "# let's add a fully-connected layer\n",
    "x = Dense(1024, activation='relu')(x)\n",
    "\n",
    "# and a logistic layer -- let's say we have 2 classes\n",
    "predictions = Dense(nb_classes, activation='softmax')(x)\n",
    "\n",
    "# What is the correct input? Obviously not base_model.input.\n",
    "model = Model(inputTensor, predictions)\n",
    "\n",
    "# first: train only the top layers (which were randomly initialized)\n",
    "# i.e. freeze all convolutional InceptionV3 layers\n",
    "for layer in conv_base.layers:\n",
    "    layer.trainable = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Compile models\n",
    "model.compile(\n",
    "    optimizer = RMSprop(lr=2e-5),\n",
    "    loss='categorical_crossentropy',  \n",
    "    metrics=['accuracy'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 4000 samples, validate on 1200 samples\n",
      "Epoch 1/90\n",
      "4000/4000 [==============================] - 5s 1ms/step - loss: 1.1671 - acc: 0.5067 - val_loss: 1.0343 - val_acc: 0.5925\n",
      "Epoch 2/90\n",
      "4000/4000 [==============================] - 4s 967us/step - loss: 0.9771 - acc: 0.6175 - val_loss: 0.9299 - val_acc: 0.6308\n",
      "Epoch 3/90\n",
      "4000/4000 [==============================] - 4s 944us/step - loss: 0.8862 - acc: 0.6607 - val_loss: 0.8576 - val_acc: 0.6758\n",
      "Epoch 4/90\n",
      "4000/4000 [==============================] - 3s 821us/step - loss: 0.8285 - acc: 0.6880 - val_loss: 0.8067 - val_acc: 0.6992\n",
      "Epoch 5/90\n",
      "4000/4000 [==============================] - 3s 856us/step - loss: 0.7836 - acc: 0.7075 - val_loss: 0.7653 - val_acc: 0.7083\n",
      "Epoch 6/90\n",
      "4000/4000 [==============================] - 4s 892us/step - loss: 0.7459 - acc: 0.7242 - val_loss: 0.7577 - val_acc: 0.6983\n",
      "Epoch 7/90\n",
      "4000/4000 [==============================] - 3s 863us/step - loss: 0.7167 - acc: 0.7298 - val_loss: 0.7234 - val_acc: 0.7225\n",
      "Epoch 8/90\n",
      "4000/4000 [==============================] - 4s 892us/step - loss: 0.6919 - acc: 0.7428 - val_loss: 0.6955 - val_acc: 0.7417\n",
      "Epoch 9/90\n",
      "4000/4000 [==============================] - 4s 951us/step - loss: 0.6709 - acc: 0.7493 - val_loss: 0.6820 - val_acc: 0.7350\n",
      "Epoch 10/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.6517 - acc: 0.7540 - val_loss: 0.6837 - val_acc: 0.7333\n",
      "Epoch 11/90\n",
      "4000/4000 [==============================] - 4s 941us/step - loss: 0.6316 - acc: 0.7692 - val_loss: 0.6565 - val_acc: 0.7567\n",
      "Epoch 12/90\n",
      "4000/4000 [==============================] - 3s 834us/step - loss: 0.6183 - acc: 0.7732 - val_loss: 0.6419 - val_acc: 0.7625\n",
      "Epoch 13/90\n",
      "4000/4000 [==============================] - 4s 898us/step - loss: 0.6022 - acc: 0.7812 - val_loss: 0.6431 - val_acc: 0.7600\n",
      "Epoch 14/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.5885 - acc: 0.7890 - val_loss: 0.6194 - val_acc: 0.7725\n",
      "Epoch 15/90\n",
      "4000/4000 [==============================] - 4s 992us/step - loss: 0.5760 - acc: 0.7867 - val_loss: 0.6278 - val_acc: 0.7633\n",
      "Epoch 16/90\n",
      "4000/4000 [==============================] - 3s 859us/step - loss: 0.5643 - acc: 0.7950 - val_loss: 0.6152 - val_acc: 0.7683\n",
      "Epoch 17/90\n",
      "4000/4000 [==============================] - 4s 880us/step - loss: 0.5516 - acc: 0.7975 - val_loss: 0.5979 - val_acc: 0.7808\n",
      "Epoch 18/90\n",
      "4000/4000 [==============================] - 4s 997us/step - loss: 0.5419 - acc: 0.8055 - val_loss: 0.6122 - val_acc: 0.7708\n",
      "Epoch 19/90\n",
      "4000/4000 [==============================] - 4s 890us/step - loss: 0.5338 - acc: 0.8110 - val_loss: 0.5914 - val_acc: 0.7725\n",
      "Epoch 20/90\n",
      "4000/4000 [==============================] - 4s 937us/step - loss: 0.5260 - acc: 0.8115 - val_loss: 0.5746 - val_acc: 0.7808\n",
      "Epoch 21/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.5153 - acc: 0.8187 - val_loss: 0.5698 - val_acc: 0.7883\n",
      "Epoch 22/90\n",
      "4000/4000 [==============================] - 4s 974us/step - loss: 0.5055 - acc: 0.8220 - val_loss: 0.5661 - val_acc: 0.7875\n",
      "Epoch 23/90\n",
      "4000/4000 [==============================] - 4s 877us/step - loss: 0.4970 - acc: 0.8275 - val_loss: 0.5745 - val_acc: 0.7842\n",
      "Epoch 24/90\n",
      "4000/4000 [==============================] - 4s 903us/step - loss: 0.4892 - acc: 0.8273 - val_loss: 0.5517 - val_acc: 0.7942\n",
      "Epoch 25/90\n",
      "4000/4000 [==============================] - 3s 849us/step - loss: 0.4824 - acc: 0.8307 - val_loss: 0.5602 - val_acc: 0.7858\n",
      "Epoch 26/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.4740 - acc: 0.8350 - val_loss: 0.5660 - val_acc: 0.7833\n",
      "Epoch 27/90\n",
      "4000/4000 [==============================] - 3s 853us/step - loss: 0.4679 - acc: 0.8363 - val_loss: 0.5371 - val_acc: 0.8017\n",
      "Epoch 28/90\n",
      "4000/4000 [==============================] - 3s 837us/step - loss: 0.4602 - acc: 0.8410 - val_loss: 0.5331 - val_acc: 0.8042\n",
      "Epoch 29/90\n",
      "4000/4000 [==============================] - 3s 841us/step - loss: 0.4546 - acc: 0.8475 - val_loss: 0.5305 - val_acc: 0.8008\n",
      "Epoch 30/90\n",
      "4000/4000 [==============================] - 3s 848us/step - loss: 0.4504 - acc: 0.8423 - val_loss: 0.5253 - val_acc: 0.8058\n",
      "Epoch 31/90\n",
      "4000/4000 [==============================] - 3s 859us/step - loss: 0.4413 - acc: 0.8508 - val_loss: 0.5216 - val_acc: 0.8058\n",
      "Epoch 32/90\n",
      "4000/4000 [==============================] - 4s 958us/step - loss: 0.4366 - acc: 0.8495 - val_loss: 0.5440 - val_acc: 0.7925\n",
      "Epoch 33/90\n",
      "4000/4000 [==============================] - 4s 930us/step - loss: 0.4315 - acc: 0.8550 - val_loss: 0.5252 - val_acc: 0.8025\n",
      "Epoch 34/90\n",
      "4000/4000 [==============================] - 3s 867us/step - loss: 0.4266 - acc: 0.8558 - val_loss: 0.5177 - val_acc: 0.8050\n",
      "Epoch 35/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.4201 - acc: 0.8593 - val_loss: 0.5096 - val_acc: 0.8083\n",
      "Epoch 36/90\n",
      "4000/4000 [==============================] - 4s 915us/step - loss: 0.4134 - acc: 0.8608 - val_loss: 0.5267 - val_acc: 0.7983\n",
      "Epoch 37/90\n",
      "4000/4000 [==============================] - 4s 935us/step - loss: 0.4101 - acc: 0.8590 - val_loss: 0.5100 - val_acc: 0.8092\n",
      "Epoch 38/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.4057 - acc: 0.8632 - val_loss: 0.5074 - val_acc: 0.8075\n",
      "Epoch 39/90\n",
      "4000/4000 [==============================] - 4s 941us/step - loss: 0.4001 - acc: 0.8668 - val_loss: 0.4997 - val_acc: 0.8133\n",
      "Epoch 40/90\n",
      "4000/4000 [==============================] - 4s 923us/step - loss: 0.3961 - acc: 0.8683 - val_loss: 0.5007 - val_acc: 0.8108\n",
      "Epoch 41/90\n",
      "4000/4000 [==============================] - 4s 879us/step - loss: 0.3913 - acc: 0.8695 - val_loss: 0.4975 - val_acc: 0.8083\n",
      "Epoch 42/90\n",
      "4000/4000 [==============================] - 4s 886us/step - loss: 0.3863 - acc: 0.8732 - val_loss: 0.4921 - val_acc: 0.8158\n",
      "Epoch 43/90\n",
      "4000/4000 [==============================] - 4s 980us/step - loss: 0.3843 - acc: 0.8713 - val_loss: 0.4910 - val_acc: 0.8183\n",
      "Epoch 44/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.3772 - acc: 0.8762 - val_loss: 0.4914 - val_acc: 0.8150\n",
      "Epoch 45/90\n",
      "4000/4000 [==============================] - 3s 849us/step - loss: 0.3752 - acc: 0.8798 - val_loss: 0.4909 - val_acc: 0.8192\n",
      "Epoch 46/90\n",
      "4000/4000 [==============================] - 4s 891us/step - loss: 0.3695 - acc: 0.8790 - val_loss: 0.5021 - val_acc: 0.8075\n",
      "Epoch 47/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.3659 - acc: 0.8808 - val_loss: 0.4830 - val_acc: 0.8058\n",
      "Epoch 48/90\n",
      "4000/4000 [==============================] - 4s 998us/step - loss: 0.3612 - acc: 0.8832 - val_loss: 0.4917 - val_acc: 0.8167\n",
      "Epoch 49/90\n",
      "4000/4000 [==============================] - 4s 892us/step - loss: 0.3587 - acc: 0.8795 - val_loss: 0.4850 - val_acc: 0.8200\n",
      "Epoch 50/90\n",
      "4000/4000 [==============================] - 4s 933us/step - loss: 0.3544 - acc: 0.8880 - val_loss: 0.4749 - val_acc: 0.8192\n",
      "Epoch 51/90\n",
      "4000/4000 [==============================] - 4s 999us/step - loss: 0.3497 - acc: 0.8877 - val_loss: 0.4762 - val_acc: 0.8250\n",
      "Epoch 52/90\n",
      "4000/4000 [==============================] - 4s 877us/step - loss: 0.3458 - acc: 0.8875 - val_loss: 0.4886 - val_acc: 0.8183\n",
      "Epoch 53/90\n",
      "4000/4000 [==============================] - 3s 851us/step - loss: 0.3444 - acc: 0.8905 - val_loss: 0.4745 - val_acc: 0.8183\n",
      "Epoch 54/90\n",
      "4000/4000 [==============================] - 4s 941us/step - loss: 0.3395 - acc: 0.8920 - val_loss: 0.4707 - val_acc: 0.8250\n",
      "Epoch 55/90\n",
      "4000/4000 [==============================] - 4s 994us/step - loss: 0.3356 - acc: 0.8937 - val_loss: 0.4660 - val_acc: 0.8217\n",
      "Epoch 56/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.3307 - acc: 0.8958 - val_loss: 0.4725 - val_acc: 0.8150\n",
      "Epoch 57/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.3286 - acc: 0.8985 - val_loss: 0.4909 - val_acc: 0.8150\n",
      "Epoch 58/90\n",
      "4000/4000 [==============================] - 4s 956us/step - loss: 0.3260 - acc: 0.8958 - val_loss: 0.4731 - val_acc: 0.8175\n",
      "Epoch 59/90\n",
      "4000/4000 [==============================] - 4s 935us/step - loss: 0.3220 - acc: 0.9002 - val_loss: 0.4645 - val_acc: 0.8267\n",
      "Epoch 60/90\n",
      "4000/4000 [==============================] - 4s 970us/step - loss: 0.3191 - acc: 0.9012 - val_loss: 0.4630 - val_acc: 0.8242\n",
      "Epoch 61/90\n",
      "4000/4000 [==============================] - 4s 892us/step - loss: 0.3150 - acc: 0.9040 - val_loss: 0.4563 - val_acc: 0.8283\n",
      "Epoch 62/90\n",
      "4000/4000 [==============================] - 4s 977us/step - loss: 0.3134 - acc: 0.9050 - val_loss: 0.4613 - val_acc: 0.8267\n",
      "Epoch 63/90\n",
      "4000/4000 [==============================] - 3s 874us/step - loss: 0.3091 - acc: 0.9060 - val_loss: 0.4526 - val_acc: 0.8358\n",
      "Epoch 64/90\n",
      "4000/4000 [==============================] - 4s 917us/step - loss: 0.3060 - acc: 0.9045 - val_loss: 0.4617 - val_acc: 0.8275\n",
      "Epoch 65/90\n",
      "4000/4000 [==============================] - 4s 968us/step - loss: 0.3029 - acc: 0.9105 - val_loss: 0.4526 - val_acc: 0.8367\n",
      "Epoch 66/90\n",
      "4000/4000 [==============================] - 4s 998us/step - loss: 0.2999 - acc: 0.9098 - val_loss: 0.4540 - val_acc: 0.8283\n",
      "Epoch 67/90\n",
      "4000/4000 [==============================] - 4s 905us/step - loss: 0.2955 - acc: 0.9130 - val_loss: 0.4586 - val_acc: 0.8317\n",
      "Epoch 68/90\n",
      "4000/4000 [==============================] - 4s 985us/step - loss: 0.2936 - acc: 0.9145 - val_loss: 0.4637 - val_acc: 0.8167\n",
      "Epoch 69/90\n",
      "4000/4000 [==============================] - 4s 907us/step - loss: 0.2907 - acc: 0.9153 - val_loss: 0.4530 - val_acc: 0.8308\n",
      "Epoch 70/90\n",
      "4000/4000 [==============================] - 4s 951us/step - loss: 0.2884 - acc: 0.9150 - val_loss: 0.4470 - val_acc: 0.8392\n",
      "Epoch 71/90\n",
      "4000/4000 [==============================] - 4s 987us/step - loss: 0.2844 - acc: 0.9183 - val_loss: 0.4495 - val_acc: 0.8292\n",
      "Epoch 72/90\n",
      "4000/4000 [==============================] - 4s 987us/step - loss: 0.2839 - acc: 0.9193 - val_loss: 0.4527 - val_acc: 0.8258\n",
      "Epoch 73/90\n",
      "4000/4000 [==============================] - 4s 971us/step - loss: 0.2784 - acc: 0.9210 - val_loss: 0.4493 - val_acc: 0.8308\n",
      "Epoch 74/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.2766 - acc: 0.9193 - val_loss: 0.4573 - val_acc: 0.8233\n",
      "Epoch 75/90\n",
      "4000/4000 [==============================] - 4s 951us/step - loss: 0.2733 - acc: 0.9230 - val_loss: 0.4388 - val_acc: 0.8375\n",
      "Epoch 76/90\n",
      "4000/4000 [==============================] - 4s 991us/step - loss: 0.2709 - acc: 0.9248 - val_loss: 0.4411 - val_acc: 0.8350\n",
      "Epoch 77/90\n",
      "4000/4000 [==============================] - 4s 941us/step - loss: 0.2691 - acc: 0.9233 - val_loss: 0.4444 - val_acc: 0.8408\n",
      "Epoch 78/90\n",
      "4000/4000 [==============================] - 4s 950us/step - loss: 0.2659 - acc: 0.9223 - val_loss: 0.4416 - val_acc: 0.8375\n",
      "Epoch 79/90\n",
      "4000/4000 [==============================] - 4s 879us/step - loss: 0.2627 - acc: 0.9245 - val_loss: 0.4397 - val_acc: 0.8383\n",
      "Epoch 80/90\n",
      "4000/4000 [==============================] - 4s 999us/step - loss: 0.2616 - acc: 0.9255 - val_loss: 0.4497 - val_acc: 0.8275\n",
      "Epoch 81/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.2582 - acc: 0.9280 - val_loss: 0.4436 - val_acc: 0.8292\n",
      "Epoch 82/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.2558 - acc: 0.9267 - val_loss: 0.4376 - val_acc: 0.8317\n",
      "Epoch 83/90\n",
      "4000/4000 [==============================] - 4s 1ms/step - loss: 0.2529 - acc: 0.9275 - val_loss: 0.4378 - val_acc: 0.8342\n",
      "Epoch 84/90\n",
      "4000/4000 [==============================] - 4s 978us/step - loss: 0.2509 - acc: 0.9325 - val_loss: 0.4556 - val_acc: 0.8392\n",
      "Epoch 85/90\n",
      "4000/4000 [==============================] - 4s 991us/step - loss: 0.2482 - acc: 0.9308 - val_loss: 0.4391 - val_acc: 0.8317\n",
      "Epoch 86/90\n",
      "4000/4000 [==============================] - 4s 956us/step - loss: 0.2451 - acc: 0.9317 - val_loss: 0.4482 - val_acc: 0.8325\n",
      "Epoch 87/90\n",
      "4000/4000 [==============================] - 4s 895us/step - loss: 0.2439 - acc: 0.9330 - val_loss: 0.4406 - val_acc: 0.8258\n",
      "Epoch 88/90\n",
      "4000/4000 [==============================] - 4s 891us/step - loss: 0.2410 - acc: 0.9368 - val_loss: 0.4373 - val_acc: 0.8308\n",
      "Epoch 89/90\n",
      "4000/4000 [==============================] - 4s 958us/step - loss: 0.2385 - acc: 0.9365 - val_loss: 0.4331 - val_acc: 0.8358\n",
      "Epoch 90/90\n",
      "4000/4000 [==============================] - 4s 917us/step - loss: 0.2362 - acc: 0.9377 - val_loss: 0.4309 - val_acc: 0.8383\n"
     ]
    }
   ],
   "source": [
    "#Train model\n",
    "\n",
    "history = model.fit(\n",
    "    train_data, \n",
    "    train_labels,\n",
    "    epochs=epochs,  \n",
    "    batch_size=batch_size, \n",
    "    validation_data=(validation_data, validation_labels))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Guardamos el modelo\n",
    "# h5py \n",
    "model.save_weights('InceptionV3/InceptionV3_FE_NS_model_1.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def print_graphics (history):\n",
    "    #Mostramos otro tipo de grafico\n",
    "    acc = history.history['acc']\n",
    "    val_acc = history.history['val_acc']\n",
    "    loss = history.history['loss']\n",
    "    val_loss = history.history['val_loss']\n",
    "\n",
    "    epochs = range(1, len(acc)+ 1)\n",
    "\n",
    "    plt.plot(epochs, acc, 'bo', label='Training acc')\n",
    "    plt.plot(epochs, val_acc, 'b', label='Validation acc')\n",
    "    #plt.tittle('Trainning and validation accuracy')\n",
    "    plt.legend()\n",
    "\n",
    "    plt.figure()\n",
    "    plt.plot(epochs, loss, 'bo', label='Training loss')\n",
    "    plt.plot(epochs, val_loss, 'b', label='Validation loss')\n",
    "    #plt.tittle('Trainning and validation loss')\n",
    "    plt.legend()\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def print_evaluate_model(model, history):\n",
    "    #Evaluate model\n",
    "    (loss, acc) = model.evaluate(\n",
    "        test_features, test_labels, \n",
    "        batch_size=batch_size, \n",
    "        verbose=0)\n",
    "\n",
    "    print(\"acc: {0:.2f}% - loss: {1:f}\".format(acc * 100, loss))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7effba3ab898>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7effba3d1b38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print_graphics (history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "acc: 79.42% - loss: 0.534959\n"
     ]
    }
   ],
   "source": [
    "print_evaluate_model(model, history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}