DenseNet_CheXnet.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import cv2\n",
    "import os\n",
    "import sys\n",
    "import random\n",
    "import glob\n",
    "import keras\n",
    "import math\n",
    "\n",
    "import numpy as np\n",
    "from numpy import argmax\n",
    "from keras.models import load_model\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "from sklearn.preprocessing import LabelBinarizer\n",
    "from keras.optimizers import Adam\n",
    "from keras.models import Sequential, load_model\n",
    "from keras.layers.normalization import BatchNormalization\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from keras.layers.convolutional import Conv2D, MaxPooling2D\n",
    "from keras.layers.core import Activation,Flatten,Dropout,Dense\n",
    "\n",
    "from keras.preprocessing.image import ImageDataGenerator, array_to_img, img_to_array, load_img\n",
    "\n",
    "from imutils import paths\n",
    "from keras.models import Model\n",
    "from keras.applications.inception_v3 import InceptionV3\n",
    "from keras.applications.densenet import DenseNet121\n",
    "\n",
    "from keras.layers import Conv2D, GlobalAveragePooling1D, AveragePooling2D, AveragePooling1D, AveragePooling3D, GlobalAveragePooling2D, GlobalAveragePooling3D\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[22. 28.]\n",
      " [49. 64.]]\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "with tf.device('/gpu:0'):\n",
    "    a = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[2, 3], name='a')\n",
    "    b = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[3, 2], name='b')\n",
    "    c = tf.matmul(a, b)\n",
    "with tf.Session() as sess:\n",
    "    print (sess.run(c))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1341"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import glob\n",
    "files = glob.glob(\"/home/cudacenter/BE/New_Chest_Xray/train/normal/*.*\")\n",
    "\n",
    "import cv2\n",
    "cv_img1 = []\n",
    "for file in files:\n",
    "    image= cv2.imread(file)\n",
    "    image=cv2.resize(image,(75,75))\n",
    "    cv_img1.append(image)\n",
    "len(cv_img1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5216"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import glob\n",
    "files = glob.glob(\"/home/cudacenter/BE/New_Chest_Xray/train/pneumonia/*.*\")\n",
    "\n",
    "import cv2\n",
    "\n",
    "for file in files:\n",
    "    image= cv2.imread(file)\n",
    "    image=cv2.resize(image,(75,75))\n",
    "    cv_img1.append(image)\n",
    "len(cv_img1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "ytrain=np.empty((0,2),int)\n",
    "a1=np.array([1,0.0])\n",
    "a1=np.expand_dims(a1,axis=0)\n",
    "a2=np.array([0,1.0])\n",
    "a2=np.expand_dims(a2,axis=0)\n",
    "\n",
    "for i in range(1341):\n",
    "    ytrain=np.append(ytrain,a1,axis=0)\n",
    "    \n",
    "for i in range(3875):\n",
    "    ytrain=np.append(ytrain,a2,axis=0) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5216\n"
     ]
    }
   ],
   "source": [
    "print(len(ytrain))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5216, 2)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ytrain.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "234"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import glob\n",
    "files = glob.glob(\"/home/cudacenter/BE/New_Chest_Xray/test/normal/*.*\")\n",
    "import cv2\n",
    "cv_img2 = []\n",
    "for file in files:\n",
    "    image= cv2.imread(file)\n",
    "    image=cv2.resize(image,(75,75))\n",
    "    cv_img2.append(image)\n",
    "len(cv_img2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "624"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "files = glob.glob(\"/home/cudacenter/BE/New_Chest_Xray/test/pneumonia/*.*\")\n",
    "\n",
    "import cv2\n",
    "\n",
    "for file in files:\n",
    "    image= cv2.imread(file)\n",
    "    image=cv2.resize(image,(75,75))\n",
    "    cv_img2.append(image)\n",
    "len(cv_img2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "ytest=np.empty((0,2),int)\n",
    "a1=np.array([1,0.0])\n",
    "a1=np.expand_dims(a1,axis=0)\n",
    "a2=np.array([0,1.0])\n",
    "a2=np.expand_dims(a2,axis=0)\n",
    "\n",
    "for i in range(234):\n",
    "    ytest=np.append(ytest,a1,axis=0)\n",
    "    \n",
    "for i in range(390):\n",
    "    ytest=np.append(ytest,a2,axis=0) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "624\n"
     ]
    }
   ],
   "source": [
    "print(len(ytest))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "cv_img1=np.array(cv_img1)\n",
    "cv_img2=np.array(cv_img2)\n",
    "\n",
    "cv_img1 = cv_img1.astype('float32')\n",
    "cv_img2 = cv_img2.astype('float32')\n",
    "#val_images=np.array(val_images)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "cv_img1=cv_img1/255\n",
    "cv_img2=cv_img2/255\n",
    "\n",
    "data_train,data_val,labels_train,labels_val=train_test_split(cv_img1,ytrain,test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\n    lrate = initial_lrate\\n    if (epoch == 10) :\\n        lrate = 0.00001\\n    elif (epoch == 20) :\\n        lrate = 0.000001\\n    elif (epoch == 30) :\\n        lrate = 0.0000001\\n'"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def step_decay(epoch):\n",
    "    initial_lrate=0.001\n",
    "    drop=0.1\n",
    "    epochs_drop=10\n",
    "    lrate=initial_lrate * math.pow(drop,math.floor((1+epoch)/epochs_drop))\n",
    "    return lrate\n",
    "'''\n",
    "    lrate = initial_lrate\n",
    "    if (epoch == 10) :\n",
    "        lrate = 0.00001\n",
    "    elif (epoch == 20) :\n",
    "        lrate = 0.000001\n",
    "    elif (epoch == 30) :\n",
    "        lrate = 0.0000001\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tensor(\"input_1:0\", shape=(?, 75, 75, 3), dtype=float32)\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            (None, 75, 75, 3)    0                                            \n",
      "__________________________________________________________________________________________________\n",
      "zero_padding2d_1 (ZeroPadding2D (None, 81, 81, 3)    0           input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv1/conv (Conv2D)             (None, 38, 38, 64)   9408        zero_padding2d_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv1/bn (BatchNormalization)   (None, 38, 38, 64)   256         conv1/conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv1/relu (Activation)         (None, 38, 38, 64)   0           conv1/bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "zero_padding2d_2 (ZeroPadding2D (None, 40, 40, 64)   0           conv1/relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool1 (MaxPooling2D)            (None, 19, 19, 64)   0           zero_padding2d_2[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_0_bn (BatchNormali (None, 19, 19, 64)   256         pool1[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_0_relu (Activation (None, 19, 19, 64)   0           conv2_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_1_conv (Conv2D)    (None, 19, 19, 128)  8192        conv2_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_1_bn (BatchNormali (None, 19, 19, 128)  512         conv2_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_1_relu (Activation (None, 19, 19, 128)  0           conv2_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_2_conv (Conv2D)    (None, 19, 19, 32)   36864       conv2_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_concat (Concatenat (None, 19, 19, 96)   0           pool1[0][0]                      \n",
      "                                                                 conv2_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_0_bn (BatchNormali (None, 19, 19, 96)   384         conv2_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_0_relu (Activation (None, 19, 19, 96)   0           conv2_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_1_conv (Conv2D)    (None, 19, 19, 128)  12288       conv2_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_1_bn (BatchNormali (None, 19, 19, 128)  512         conv2_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_1_relu (Activation (None, 19, 19, 128)  0           conv2_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_2_conv (Conv2D)    (None, 19, 19, 32)   36864       conv2_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_concat (Concatenat (None, 19, 19, 128)  0           conv2_block1_concat[0][0]        \n",
      "                                                                 conv2_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_0_bn (BatchNormali (None, 19, 19, 128)  512         conv2_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_0_relu (Activation (None, 19, 19, 128)  0           conv2_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_1_conv (Conv2D)    (None, 19, 19, 128)  16384       conv2_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_1_bn (BatchNormali (None, 19, 19, 128)  512         conv2_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_1_relu (Activation (None, 19, 19, 128)  0           conv2_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_2_conv (Conv2D)    (None, 19, 19, 32)   36864       conv2_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_concat (Concatenat (None, 19, 19, 160)  0           conv2_block2_concat[0][0]        \n",
      "                                                                 conv2_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_0_bn (BatchNormali (None, 19, 19, 160)  640         conv2_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_0_relu (Activation (None, 19, 19, 160)  0           conv2_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_1_conv (Conv2D)    (None, 19, 19, 128)  20480       conv2_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_1_bn (BatchNormali (None, 19, 19, 128)  512         conv2_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_1_relu (Activation (None, 19, 19, 128)  0           conv2_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_2_conv (Conv2D)    (None, 19, 19, 32)   36864       conv2_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_concat (Concatenat (None, 19, 19, 192)  0           conv2_block3_concat[0][0]        \n",
      "                                                                 conv2_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_0_bn (BatchNormali (None, 19, 19, 192)  768         conv2_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_0_relu (Activation (None, 19, 19, 192)  0           conv2_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_1_conv (Conv2D)    (None, 19, 19, 128)  24576       conv2_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_1_bn (BatchNormali (None, 19, 19, 128)  512         conv2_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_1_relu (Activation (None, 19, 19, 128)  0           conv2_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_2_conv (Conv2D)    (None, 19, 19, 32)   36864       conv2_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_concat (Concatenat (None, 19, 19, 224)  0           conv2_block4_concat[0][0]        \n",
      "                                                                 conv2_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_0_bn (BatchNormali (None, 19, 19, 224)  896         conv2_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_0_relu (Activation (None, 19, 19, 224)  0           conv2_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_1_conv (Conv2D)    (None, 19, 19, 128)  28672       conv2_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_1_bn (BatchNormali (None, 19, 19, 128)  512         conv2_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_1_relu (Activation (None, 19, 19, 128)  0           conv2_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_2_conv (Conv2D)    (None, 19, 19, 32)   36864       conv2_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_concat (Concatenat (None, 19, 19, 256)  0           conv2_block5_concat[0][0]        \n",
      "                                                                 conv2_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "pool2_bn (BatchNormalization)   (None, 19, 19, 256)  1024        conv2_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "pool2_relu (Activation)         (None, 19, 19, 256)  0           pool2_bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "pool2_conv (Conv2D)             (None, 19, 19, 128)  32768       pool2_relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool2_pool (AveragePooling2D)   (None, 9, 9, 128)    0           pool2_conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_0_bn (BatchNormali (None, 9, 9, 128)    512         pool2_pool[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_0_relu (Activation (None, 9, 9, 128)    0           conv3_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_1_conv (Conv2D)    (None, 9, 9, 128)    16384       conv3_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_1_bn (BatchNormali (None, 9, 9, 128)    512         conv3_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_1_relu (Activation (None, 9, 9, 128)    0           conv3_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_2_conv (Conv2D)    (None, 9, 9, 32)     36864       conv3_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_concat (Concatenat (None, 9, 9, 160)    0           pool2_pool[0][0]                 \n",
      "                                                                 conv3_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_0_bn (BatchNormali (None, 9, 9, 160)    640         conv3_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_0_relu (Activation (None, 9, 9, 160)    0           conv3_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_1_conv (Conv2D)    (None, 9, 9, 128)    20480       conv3_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_1_bn (BatchNormali (None, 9, 9, 128)    512         conv3_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_1_relu (Activation (None, 9, 9, 128)    0           conv3_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_2_conv (Conv2D)    (None, 9, 9, 32)     36864       conv3_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_concat (Concatenat (None, 9, 9, 192)    0           conv3_block1_concat[0][0]        \n",
      "                                                                 conv3_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_0_bn (BatchNormali (None, 9, 9, 192)    768         conv3_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_0_relu (Activation (None, 9, 9, 192)    0           conv3_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_1_conv (Conv2D)    (None, 9, 9, 128)    24576       conv3_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_1_bn (BatchNormali (None, 9, 9, 128)    512         conv3_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_1_relu (Activation (None, 9, 9, 128)    0           conv3_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_2_conv (Conv2D)    (None, 9, 9, 32)     36864       conv3_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_concat (Concatenat (None, 9, 9, 224)    0           conv3_block2_concat[0][0]        \n",
      "                                                                 conv3_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_0_bn (BatchNormali (None, 9, 9, 224)    896         conv3_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_0_relu (Activation (None, 9, 9, 224)    0           conv3_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_1_conv (Conv2D)    (None, 9, 9, 128)    28672       conv3_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_1_bn (BatchNormali (None, 9, 9, 128)    512         conv3_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_1_relu (Activation (None, 9, 9, 128)    0           conv3_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_2_conv (Conv2D)    (None, 9, 9, 32)     36864       conv3_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_concat (Concatenat (None, 9, 9, 256)    0           conv3_block3_concat[0][0]        \n",
      "                                                                 conv3_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_0_bn (BatchNormali (None, 9, 9, 256)    1024        conv3_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_0_relu (Activation (None, 9, 9, 256)    0           conv3_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_1_conv (Conv2D)    (None, 9, 9, 128)    32768       conv3_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_1_bn (BatchNormali (None, 9, 9, 128)    512         conv3_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_1_relu (Activation (None, 9, 9, 128)    0           conv3_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_2_conv (Conv2D)    (None, 9, 9, 32)     36864       conv3_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_concat (Concatenat (None, 9, 9, 288)    0           conv3_block4_concat[0][0]        \n",
      "                                                                 conv3_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_0_bn (BatchNormali (None, 9, 9, 288)    1152        conv3_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_0_relu (Activation (None, 9, 9, 288)    0           conv3_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_1_conv (Conv2D)    (None, 9, 9, 128)    36864       conv3_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_1_bn (BatchNormali (None, 9, 9, 128)    512         conv3_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_1_relu (Activation (None, 9, 9, 128)    0           conv3_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_2_conv (Conv2D)    (None, 9, 9, 32)     36864       conv3_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_concat (Concatenat (None, 9, 9, 320)    0           conv3_block5_concat[0][0]        \n",
      "                                                                 conv3_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_0_bn (BatchNormali (None, 9, 9, 320)    1280        conv3_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_0_relu (Activation (None, 9, 9, 320)    0           conv3_block7_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_1_conv (Conv2D)    (None, 9, 9, 128)    40960       conv3_block7_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_1_bn (BatchNormali (None, 9, 9, 128)    512         conv3_block7_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_1_relu (Activation (None, 9, 9, 128)    0           conv3_block7_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_2_conv (Conv2D)    (None, 9, 9, 32)     36864       conv3_block7_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_concat (Concatenat (None, 9, 9, 352)    0           conv3_block6_concat[0][0]        \n",
      "                                                                 conv3_block7_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_0_bn (BatchNormali (None, 9, 9, 352)    1408        conv3_block7_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_0_relu (Activation (None, 9, 9, 352)    0           conv3_block8_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_1_conv (Conv2D)    (None, 9, 9, 128)    45056       conv3_block8_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_1_bn (BatchNormali (None, 9, 9, 128)    512         conv3_block8_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_1_relu (Activation (None, 9, 9, 128)    0           conv3_block8_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_2_conv (Conv2D)    (None, 9, 9, 32)     36864       conv3_block8_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_concat (Concatenat (None, 9, 9, 384)    0           conv3_block7_concat[0][0]        \n",
      "                                                                 conv3_block8_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_0_bn (BatchNormali (None, 9, 9, 384)    1536        conv3_block8_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_0_relu (Activation (None, 9, 9, 384)    0           conv3_block9_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_1_conv (Conv2D)    (None, 9, 9, 128)    49152       conv3_block9_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_1_bn (BatchNormali (None, 9, 9, 128)    512         conv3_block9_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_1_relu (Activation (None, 9, 9, 128)    0           conv3_block9_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_2_conv (Conv2D)    (None, 9, 9, 32)     36864       conv3_block9_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_concat (Concatenat (None, 9, 9, 416)    0           conv3_block8_concat[0][0]        \n",
      "                                                                 conv3_block9_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_0_bn (BatchNormal (None, 9, 9, 416)    1664        conv3_block9_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_0_relu (Activatio (None, 9, 9, 416)    0           conv3_block10_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_1_conv (Conv2D)   (None, 9, 9, 128)    53248       conv3_block10_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_1_bn (BatchNormal (None, 9, 9, 128)    512         conv3_block10_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_1_relu (Activatio (None, 9, 9, 128)    0           conv3_block10_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_2_conv (Conv2D)   (None, 9, 9, 32)     36864       conv3_block10_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_concat (Concatena (None, 9, 9, 448)    0           conv3_block9_concat[0][0]        \n",
      "                                                                 conv3_block10_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_0_bn (BatchNormal (None, 9, 9, 448)    1792        conv3_block10_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_0_relu (Activatio (None, 9, 9, 448)    0           conv3_block11_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_1_conv (Conv2D)   (None, 9, 9, 128)    57344       conv3_block11_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_1_bn (BatchNormal (None, 9, 9, 128)    512         conv3_block11_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_1_relu (Activatio (None, 9, 9, 128)    0           conv3_block11_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_2_conv (Conv2D)   (None, 9, 9, 32)     36864       conv3_block11_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_concat (Concatena (None, 9, 9, 480)    0           conv3_block10_concat[0][0]       \n",
      "                                                                 conv3_block11_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_0_bn (BatchNormal (None, 9, 9, 480)    1920        conv3_block11_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_0_relu (Activatio (None, 9, 9, 480)    0           conv3_block12_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_1_conv (Conv2D)   (None, 9, 9, 128)    61440       conv3_block12_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_1_bn (BatchNormal (None, 9, 9, 128)    512         conv3_block12_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_1_relu (Activatio (None, 9, 9, 128)    0           conv3_block12_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_2_conv (Conv2D)   (None, 9, 9, 32)     36864       conv3_block12_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_concat (Concatena (None, 9, 9, 512)    0           conv3_block11_concat[0][0]       \n",
      "                                                                 conv3_block12_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool3_bn (BatchNormalization)   (None, 9, 9, 512)    2048        conv3_block12_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool3_relu (Activation)         (None, 9, 9, 512)    0           pool3_bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "pool3_conv (Conv2D)             (None, 9, 9, 256)    131072      pool3_relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool3_pool (AveragePooling2D)   (None, 4, 4, 256)    0           pool3_conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_0_bn (BatchNormali (None, 4, 4, 256)    1024        pool3_pool[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_0_relu (Activation (None, 4, 4, 256)    0           conv4_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_1_conv (Conv2D)    (None, 4, 4, 128)    32768       conv4_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_1_bn (BatchNormali (None, 4, 4, 128)    512         conv4_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_1_relu (Activation (None, 4, 4, 128)    0           conv4_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_2_conv (Conv2D)    (None, 4, 4, 32)     36864       conv4_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_concat (Concatenat (None, 4, 4, 288)    0           pool3_pool[0][0]                 \n",
      "                                                                 conv4_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_0_bn (BatchNormali (None, 4, 4, 288)    1152        conv4_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_0_relu (Activation (None, 4, 4, 288)    0           conv4_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_1_conv (Conv2D)    (None, 4, 4, 128)    36864       conv4_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_1_bn (BatchNormali (None, 4, 4, 128)    512         conv4_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_1_relu (Activation (None, 4, 4, 128)    0           conv4_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_2_conv (Conv2D)    (None, 4, 4, 32)     36864       conv4_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_concat (Concatenat (None, 4, 4, 320)    0           conv4_block1_concat[0][0]        \n",
      "                                                                 conv4_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_0_bn (BatchNormali (None, 4, 4, 320)    1280        conv4_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_0_relu (Activation (None, 4, 4, 320)    0           conv4_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_1_conv (Conv2D)    (None, 4, 4, 128)    40960       conv4_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_1_bn (BatchNormali (None, 4, 4, 128)    512         conv4_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_1_relu (Activation (None, 4, 4, 128)    0           conv4_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_2_conv (Conv2D)    (None, 4, 4, 32)     36864       conv4_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_concat (Concatenat (None, 4, 4, 352)    0           conv4_block2_concat[0][0]        \n",
      "                                                                 conv4_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_0_bn (BatchNormali (None, 4, 4, 352)    1408        conv4_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_0_relu (Activation (None, 4, 4, 352)    0           conv4_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_1_conv (Conv2D)    (None, 4, 4, 128)    45056       conv4_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_1_bn (BatchNormali (None, 4, 4, 128)    512         conv4_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_1_relu (Activation (None, 4, 4, 128)    0           conv4_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_2_conv (Conv2D)    (None, 4, 4, 32)     36864       conv4_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_concat (Concatenat (None, 4, 4, 384)    0           conv4_block3_concat[0][0]        \n",
      "                                                                 conv4_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_0_bn (BatchNormali (None, 4, 4, 384)    1536        conv4_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_0_relu (Activation (None, 4, 4, 384)    0           conv4_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_1_conv (Conv2D)    (None, 4, 4, 128)    49152       conv4_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_1_bn (BatchNormali (None, 4, 4, 128)    512         conv4_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_1_relu (Activation (None, 4, 4, 128)    0           conv4_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_2_conv (Conv2D)    (None, 4, 4, 32)     36864       conv4_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_concat (Concatenat (None, 4, 4, 416)    0           conv4_block4_concat[0][0]        \n",
      "                                                                 conv4_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_0_bn (BatchNormali (None, 4, 4, 416)    1664        conv4_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_0_relu (Activation (None, 4, 4, 416)    0           conv4_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_1_conv (Conv2D)    (None, 4, 4, 128)    53248       conv4_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_1_bn (BatchNormali (None, 4, 4, 128)    512         conv4_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_1_relu (Activation (None, 4, 4, 128)    0           conv4_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_2_conv (Conv2D)    (None, 4, 4, 32)     36864       conv4_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_concat (Concatenat (None, 4, 4, 448)    0           conv4_block5_concat[0][0]        \n",
      "                                                                 conv4_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_0_bn (BatchNormali (None, 4, 4, 448)    1792        conv4_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_0_relu (Activation (None, 4, 4, 448)    0           conv4_block7_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_1_conv (Conv2D)    (None, 4, 4, 128)    57344       conv4_block7_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_1_bn (BatchNormali (None, 4, 4, 128)    512         conv4_block7_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_1_relu (Activation (None, 4, 4, 128)    0           conv4_block7_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_2_conv (Conv2D)    (None, 4, 4, 32)     36864       conv4_block7_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_concat (Concatenat (None, 4, 4, 480)    0           conv4_block6_concat[0][0]        \n",
      "                                                                 conv4_block7_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_0_bn (BatchNormali (None, 4, 4, 480)    1920        conv4_block7_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_0_relu (Activation (None, 4, 4, 480)    0           conv4_block8_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_1_conv (Conv2D)    (None, 4, 4, 128)    61440       conv4_block8_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_1_bn (BatchNormali (None, 4, 4, 128)    512         conv4_block8_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_1_relu (Activation (None, 4, 4, 128)    0           conv4_block8_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_2_conv (Conv2D)    (None, 4, 4, 32)     36864       conv4_block8_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_concat (Concatenat (None, 4, 4, 512)    0           conv4_block7_concat[0][0]        \n",
      "                                                                 conv4_block8_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_0_bn (BatchNormali (None, 4, 4, 512)    2048        conv4_block8_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_0_relu (Activation (None, 4, 4, 512)    0           conv4_block9_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_1_conv (Conv2D)    (None, 4, 4, 128)    65536       conv4_block9_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_1_bn (BatchNormali (None, 4, 4, 128)    512         conv4_block9_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_1_relu (Activation (None, 4, 4, 128)    0           conv4_block9_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_2_conv (Conv2D)    (None, 4, 4, 32)     36864       conv4_block9_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_concat (Concatenat (None, 4, 4, 544)    0           conv4_block8_concat[0][0]        \n",
      "                                                                 conv4_block9_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_0_bn (BatchNormal (None, 4, 4, 544)    2176        conv4_block9_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_0_relu (Activatio (None, 4, 4, 544)    0           conv4_block10_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_1_conv (Conv2D)   (None, 4, 4, 128)    69632       conv4_block10_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block10_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block10_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block10_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_concat (Concatena (None, 4, 4, 576)    0           conv4_block9_concat[0][0]        \n",
      "                                                                 conv4_block10_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_0_bn (BatchNormal (None, 4, 4, 576)    2304        conv4_block10_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_0_relu (Activatio (None, 4, 4, 576)    0           conv4_block11_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_1_conv (Conv2D)   (None, 4, 4, 128)    73728       conv4_block11_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block11_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block11_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block11_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_concat (Concatena (None, 4, 4, 608)    0           conv4_block10_concat[0][0]       \n",
      "                                                                 conv4_block11_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_0_bn (BatchNormal (None, 4, 4, 608)    2432        conv4_block11_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_0_relu (Activatio (None, 4, 4, 608)    0           conv4_block12_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_1_conv (Conv2D)   (None, 4, 4, 128)    77824       conv4_block12_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block12_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block12_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block12_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_concat (Concatena (None, 4, 4, 640)    0           conv4_block11_concat[0][0]       \n",
      "                                                                 conv4_block12_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_0_bn (BatchNormal (None, 4, 4, 640)    2560        conv4_block12_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_0_relu (Activatio (None, 4, 4, 640)    0           conv4_block13_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_1_conv (Conv2D)   (None, 4, 4, 128)    81920       conv4_block13_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block13_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block13_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block13_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_concat (Concatena (None, 4, 4, 672)    0           conv4_block12_concat[0][0]       \n",
      "                                                                 conv4_block13_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_0_bn (BatchNormal (None, 4, 4, 672)    2688        conv4_block13_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_0_relu (Activatio (None, 4, 4, 672)    0           conv4_block14_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_1_conv (Conv2D)   (None, 4, 4, 128)    86016       conv4_block14_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block14_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block14_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block14_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_concat (Concatena (None, 4, 4, 704)    0           conv4_block13_concat[0][0]       \n",
      "                                                                 conv4_block14_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_0_bn (BatchNormal (None, 4, 4, 704)    2816        conv4_block14_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_0_relu (Activatio (None, 4, 4, 704)    0           conv4_block15_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_1_conv (Conv2D)   (None, 4, 4, 128)    90112       conv4_block15_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block15_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block15_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block15_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_concat (Concatena (None, 4, 4, 736)    0           conv4_block14_concat[0][0]       \n",
      "                                                                 conv4_block15_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_0_bn (BatchNormal (None, 4, 4, 736)    2944        conv4_block15_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_0_relu (Activatio (None, 4, 4, 736)    0           conv4_block16_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_1_conv (Conv2D)   (None, 4, 4, 128)    94208       conv4_block16_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block16_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block16_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block16_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_concat (Concatena (None, 4, 4, 768)    0           conv4_block15_concat[0][0]       \n",
      "                                                                 conv4_block16_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_0_bn (BatchNormal (None, 4, 4, 768)    3072        conv4_block16_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_0_relu (Activatio (None, 4, 4, 768)    0           conv4_block17_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_1_conv (Conv2D)   (None, 4, 4, 128)    98304       conv4_block17_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block17_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block17_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block17_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_concat (Concatena (None, 4, 4, 800)    0           conv4_block16_concat[0][0]       \n",
      "                                                                 conv4_block17_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_0_bn (BatchNormal (None, 4, 4, 800)    3200        conv4_block17_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_0_relu (Activatio (None, 4, 4, 800)    0           conv4_block18_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_1_conv (Conv2D)   (None, 4, 4, 128)    102400      conv4_block18_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block18_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block18_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block18_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_concat (Concatena (None, 4, 4, 832)    0           conv4_block17_concat[0][0]       \n",
      "                                                                 conv4_block18_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_0_bn (BatchNormal (None, 4, 4, 832)    3328        conv4_block18_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_0_relu (Activatio (None, 4, 4, 832)    0           conv4_block19_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_1_conv (Conv2D)   (None, 4, 4, 128)    106496      conv4_block19_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block19_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block19_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block19_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_concat (Concatena (None, 4, 4, 864)    0           conv4_block18_concat[0][0]       \n",
      "                                                                 conv4_block19_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_0_bn (BatchNormal (None, 4, 4, 864)    3456        conv4_block19_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_0_relu (Activatio (None, 4, 4, 864)    0           conv4_block20_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_1_conv (Conv2D)   (None, 4, 4, 128)    110592      conv4_block20_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block20_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block20_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block20_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_concat (Concatena (None, 4, 4, 896)    0           conv4_block19_concat[0][0]       \n",
      "                                                                 conv4_block20_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_0_bn (BatchNormal (None, 4, 4, 896)    3584        conv4_block20_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_0_relu (Activatio (None, 4, 4, 896)    0           conv4_block21_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_1_conv (Conv2D)   (None, 4, 4, 128)    114688      conv4_block21_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block21_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block21_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block21_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_concat (Concatena (None, 4, 4, 928)    0           conv4_block20_concat[0][0]       \n",
      "                                                                 conv4_block21_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_0_bn (BatchNormal (None, 4, 4, 928)    3712        conv4_block21_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_0_relu (Activatio (None, 4, 4, 928)    0           conv4_block22_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_1_conv (Conv2D)   (None, 4, 4, 128)    118784      conv4_block22_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block22_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block22_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block22_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_concat (Concatena (None, 4, 4, 960)    0           conv4_block21_concat[0][0]       \n",
      "                                                                 conv4_block22_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_0_bn (BatchNormal (None, 4, 4, 960)    3840        conv4_block22_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_0_relu (Activatio (None, 4, 4, 960)    0           conv4_block23_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_1_conv (Conv2D)   (None, 4, 4, 128)    122880      conv4_block23_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block23_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block23_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block23_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_concat (Concatena (None, 4, 4, 992)    0           conv4_block22_concat[0][0]       \n",
      "                                                                 conv4_block23_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_0_bn (BatchNormal (None, 4, 4, 992)    3968        conv4_block23_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_0_relu (Activatio (None, 4, 4, 992)    0           conv4_block24_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_1_conv (Conv2D)   (None, 4, 4, 128)    126976      conv4_block24_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_1_bn (BatchNormal (None, 4, 4, 128)    512         conv4_block24_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_1_relu (Activatio (None, 4, 4, 128)    0           conv4_block24_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_2_conv (Conv2D)   (None, 4, 4, 32)     36864       conv4_block24_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_concat (Concatena (None, 4, 4, 1024)   0           conv4_block23_concat[0][0]       \n",
      "                                                                 conv4_block24_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool4_bn (BatchNormalization)   (None, 4, 4, 1024)   4096        conv4_block24_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool4_relu (Activation)         (None, 4, 4, 1024)   0           pool4_bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "pool4_conv (Conv2D)             (None, 4, 4, 512)    524288      pool4_relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool4_pool (AveragePooling2D)   (None, 2, 2, 512)    0           pool4_conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_0_bn (BatchNormali (None, 2, 2, 512)    2048        pool4_pool[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_0_relu (Activation (None, 2, 2, 512)    0           conv5_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_1_conv (Conv2D)    (None, 2, 2, 128)    65536       conv5_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_1_bn (BatchNormali (None, 2, 2, 128)    512         conv5_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_1_relu (Activation (None, 2, 2, 128)    0           conv5_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_2_conv (Conv2D)    (None, 2, 2, 32)     36864       conv5_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_concat (Concatenat (None, 2, 2, 544)    0           pool4_pool[0][0]                 \n",
      "                                                                 conv5_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_0_bn (BatchNormali (None, 2, 2, 544)    2176        conv5_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_0_relu (Activation (None, 2, 2, 544)    0           conv5_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_1_conv (Conv2D)    (None, 2, 2, 128)    69632       conv5_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_1_bn (BatchNormali (None, 2, 2, 128)    512         conv5_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_1_relu (Activation (None, 2, 2, 128)    0           conv5_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_2_conv (Conv2D)    (None, 2, 2, 32)     36864       conv5_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_concat (Concatenat (None, 2, 2, 576)    0           conv5_block1_concat[0][0]        \n",
      "                                                                 conv5_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_0_bn (BatchNormali (None, 2, 2, 576)    2304        conv5_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_0_relu (Activation (None, 2, 2, 576)    0           conv5_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_1_conv (Conv2D)    (None, 2, 2, 128)    73728       conv5_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_1_bn (BatchNormali (None, 2, 2, 128)    512         conv5_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_1_relu (Activation (None, 2, 2, 128)    0           conv5_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_2_conv (Conv2D)    (None, 2, 2, 32)     36864       conv5_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_concat (Concatenat (None, 2, 2, 608)    0           conv5_block2_concat[0][0]        \n",
      "                                                                 conv5_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_0_bn (BatchNormali (None, 2, 2, 608)    2432        conv5_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_0_relu (Activation (None, 2, 2, 608)    0           conv5_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_1_conv (Conv2D)    (None, 2, 2, 128)    77824       conv5_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_1_bn (BatchNormali (None, 2, 2, 128)    512         conv5_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_1_relu (Activation (None, 2, 2, 128)    0           conv5_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_2_conv (Conv2D)    (None, 2, 2, 32)     36864       conv5_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_concat (Concatenat (None, 2, 2, 640)    0           conv5_block3_concat[0][0]        \n",
      "                                                                 conv5_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_0_bn (BatchNormali (None, 2, 2, 640)    2560        conv5_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_0_relu (Activation (None, 2, 2, 640)    0           conv5_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_1_conv (Conv2D)    (None, 2, 2, 128)    81920       conv5_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_1_bn (BatchNormali (None, 2, 2, 128)    512         conv5_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_1_relu (Activation (None, 2, 2, 128)    0           conv5_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_2_conv (Conv2D)    (None, 2, 2, 32)     36864       conv5_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_concat (Concatenat (None, 2, 2, 672)    0           conv5_block4_concat[0][0]        \n",
      "                                                                 conv5_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_0_bn (BatchNormali (None, 2, 2, 672)    2688        conv5_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_0_relu (Activation (None, 2, 2, 672)    0           conv5_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_1_conv (Conv2D)    (None, 2, 2, 128)    86016       conv5_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_1_bn (BatchNormali (None, 2, 2, 128)    512         conv5_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_1_relu (Activation (None, 2, 2, 128)    0           conv5_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_2_conv (Conv2D)    (None, 2, 2, 32)     36864       conv5_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_concat (Concatenat (None, 2, 2, 704)    0           conv5_block5_concat[0][0]        \n",
      "                                                                 conv5_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_0_bn (BatchNormali (None, 2, 2, 704)    2816        conv5_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_0_relu (Activation (None, 2, 2, 704)    0           conv5_block7_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_1_conv (Conv2D)    (None, 2, 2, 128)    90112       conv5_block7_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_1_bn (BatchNormali (None, 2, 2, 128)    512         conv5_block7_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_1_relu (Activation (None, 2, 2, 128)    0           conv5_block7_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_2_conv (Conv2D)    (None, 2, 2, 32)     36864       conv5_block7_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_concat (Concatenat (None, 2, 2, 736)    0           conv5_block6_concat[0][0]        \n",
      "                                                                 conv5_block7_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_0_bn (BatchNormali (None, 2, 2, 736)    2944        conv5_block7_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_0_relu (Activation (None, 2, 2, 736)    0           conv5_block8_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_1_conv (Conv2D)    (None, 2, 2, 128)    94208       conv5_block8_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_1_bn (BatchNormali (None, 2, 2, 128)    512         conv5_block8_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_1_relu (Activation (None, 2, 2, 128)    0           conv5_block8_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_2_conv (Conv2D)    (None, 2, 2, 32)     36864       conv5_block8_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_concat (Concatenat (None, 2, 2, 768)    0           conv5_block7_concat[0][0]        \n",
      "                                                                 conv5_block8_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_0_bn (BatchNormali (None, 2, 2, 768)    3072        conv5_block8_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_0_relu (Activation (None, 2, 2, 768)    0           conv5_block9_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_1_conv (Conv2D)    (None, 2, 2, 128)    98304       conv5_block9_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_1_bn (BatchNormali (None, 2, 2, 128)    512         conv5_block9_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_1_relu (Activation (None, 2, 2, 128)    0           conv5_block9_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_2_conv (Conv2D)    (None, 2, 2, 32)     36864       conv5_block9_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_concat (Concatenat (None, 2, 2, 800)    0           conv5_block8_concat[0][0]        \n",
      "                                                                 conv5_block9_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_0_bn (BatchNormal (None, 2, 2, 800)    3200        conv5_block9_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_0_relu (Activatio (None, 2, 2, 800)    0           conv5_block10_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_1_conv (Conv2D)   (None, 2, 2, 128)    102400      conv5_block10_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_1_bn (BatchNormal (None, 2, 2, 128)    512         conv5_block10_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_1_relu (Activatio (None, 2, 2, 128)    0           conv5_block10_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_2_conv (Conv2D)   (None, 2, 2, 32)     36864       conv5_block10_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_concat (Concatena (None, 2, 2, 832)    0           conv5_block9_concat[0][0]        \n",
      "                                                                 conv5_block10_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_0_bn (BatchNormal (None, 2, 2, 832)    3328        conv5_block10_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_0_relu (Activatio (None, 2, 2, 832)    0           conv5_block11_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_1_conv (Conv2D)   (None, 2, 2, 128)    106496      conv5_block11_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_1_bn (BatchNormal (None, 2, 2, 128)    512         conv5_block11_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_1_relu (Activatio (None, 2, 2, 128)    0           conv5_block11_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_2_conv (Conv2D)   (None, 2, 2, 32)     36864       conv5_block11_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_concat (Concatena (None, 2, 2, 864)    0           conv5_block10_concat[0][0]       \n",
      "                                                                 conv5_block11_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_0_bn (BatchNormal (None, 2, 2, 864)    3456        conv5_block11_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_0_relu (Activatio (None, 2, 2, 864)    0           conv5_block12_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_1_conv (Conv2D)   (None, 2, 2, 128)    110592      conv5_block12_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_1_bn (BatchNormal (None, 2, 2, 128)    512         conv5_block12_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_1_relu (Activatio (None, 2, 2, 128)    0           conv5_block12_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_2_conv (Conv2D)   (None, 2, 2, 32)     36864       conv5_block12_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_concat (Concatena (None, 2, 2, 896)    0           conv5_block11_concat[0][0]       \n",
      "                                                                 conv5_block12_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_0_bn (BatchNormal (None, 2, 2, 896)    3584        conv5_block12_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_0_relu (Activatio (None, 2, 2, 896)    0           conv5_block13_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_1_conv (Conv2D)   (None, 2, 2, 128)    114688      conv5_block13_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_1_bn (BatchNormal (None, 2, 2, 128)    512         conv5_block13_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_1_relu (Activatio (None, 2, 2, 128)    0           conv5_block13_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_2_conv (Conv2D)   (None, 2, 2, 32)     36864       conv5_block13_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_concat (Concatena (None, 2, 2, 928)    0           conv5_block12_concat[0][0]       \n",
      "                                                                 conv5_block13_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_0_bn (BatchNormal (None, 2, 2, 928)    3712        conv5_block13_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_0_relu (Activatio (None, 2, 2, 928)    0           conv5_block14_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_1_conv (Conv2D)   (None, 2, 2, 128)    118784      conv5_block14_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_1_bn (BatchNormal (None, 2, 2, 128)    512         conv5_block14_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_1_relu (Activatio (None, 2, 2, 128)    0           conv5_block14_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_2_conv (Conv2D)   (None, 2, 2, 32)     36864       conv5_block14_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_concat (Concatena (None, 2, 2, 960)    0           conv5_block13_concat[0][0]       \n",
      "                                                                 conv5_block14_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_0_bn (BatchNormal (None, 2, 2, 960)    3840        conv5_block14_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_0_relu (Activatio (None, 2, 2, 960)    0           conv5_block15_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_1_conv (Conv2D)   (None, 2, 2, 128)    122880      conv5_block15_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_1_bn (BatchNormal (None, 2, 2, 128)    512         conv5_block15_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_1_relu (Activatio (None, 2, 2, 128)    0           conv5_block15_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_2_conv (Conv2D)   (None, 2, 2, 32)     36864       conv5_block15_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_concat (Concatena (None, 2, 2, 992)    0           conv5_block14_concat[0][0]       \n",
      "                                                                 conv5_block15_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_0_bn (BatchNormal (None, 2, 2, 992)    3968        conv5_block15_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_0_relu (Activatio (None, 2, 2, 992)    0           conv5_block16_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_1_conv (Conv2D)   (None, 2, 2, 128)    126976      conv5_block16_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_1_bn (BatchNormal (None, 2, 2, 128)    512         conv5_block16_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_1_relu (Activatio (None, 2, 2, 128)    0           conv5_block16_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_2_conv (Conv2D)   (None, 2, 2, 32)     36864       conv5_block16_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_concat (Concatena (None, 2, 2, 1024)   0           conv5_block15_concat[0][0]       \n",
      "                                                                 conv5_block16_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "bn (BatchNormalization)         (None, 2, 2, 1024)   4096        conv5_block16_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "relu (Activation)               (None, 2, 2, 1024)   0           bn[0][0]                         \n",
      "__________________________________________________________________________________________________\n",
      "avg_pool (GlobalAveragePooling2 (None, 1024)         0           relu[0][0]                       \n",
      "__________________________________________________________________________________________________\n",
      "fc1000 (Dense)                  (None, 14)           14350       avg_pool[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "dense_1 (Dense)                 (None, 32)           480         fc1000[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "dropout_1 (Dropout)             (None, 32)           0           dense_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "dense_2 (Dense)                 (None, 2)            66          dropout_1[0][0]                  \n",
      "==================================================================================================\n",
      "Total params: 7,052,400\n",
      "Trainable params: 6,968,752\n",
      "Non-trainable params: 83,648\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "base_model = DenseNet121(include_top=True, weights=None,input_shape=(75,75,3),classes=14)\n",
    "base_model.load_weights('/home/cudacenter/BE/New_Chest_Xray/Chexnet weights/brucechou1983_CheXNet_Keras_0.3.0_weights.h5') \n",
    "\n",
    "\n",
    "x = base_model.output\n",
    "# x = GlobalAveragePooling1D()(x)\n",
    "\n",
    "# let's add a fully-connected layer\n",
    "x = Dense(32, activation='relu')(x)\n",
    "y = Dropout(0.3)(x)\n",
    "\n",
    "# and a logistic layer -- let's say we have 2 classes\n",
    "predictions = Dense(2, activation='softmax')(y)\n",
    "\n",
    "model = Model(inputs=base_model.input, outputs=predictions)\n",
    "\n",
    "print(base_model.input)\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss='categorical_crossentropy',optimizer=Adam(lr=1e-04, decay = 1e-02),metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "\n",
      "Epoch 00001: LearningRateScheduler setting learning rate to 0.001.\n",
      "25/25 [==============================] - 28s 1s/step - loss: 0.6427 - acc: 0.6800 - val_loss: 0.5880 - val_acc: 0.7241\n",
      "Epoch 2/20\n",
      "\n",
      "Epoch 00002: LearningRateScheduler setting learning rate to 0.001.\n",
      "25/25 [==============================] - 6s 248ms/step - loss: 0.5597 - acc: 0.7750 - val_loss: 0.5437 - val_acc: 0.7251\n",
      "Epoch 3/20\n",
      "\n",
      "Epoch 00003: LearningRateScheduler setting learning rate to 0.001.\n",
      "25/25 [==============================] - 6s 237ms/step - loss: 0.5325 - acc: 0.7450 - val_loss: 0.5565 - val_acc: 0.7241\n",
      "Epoch 4/20\n",
      "\n",
      "Epoch 00004: LearningRateScheduler setting learning rate to 0.001.\n",
      "25/25 [==============================] - 6s 232ms/step - loss: 0.5193 - acc: 0.7350 - val_loss: 0.5290 - val_acc: 0.7241\n",
      "Epoch 5/20\n",
      "\n",
      "Epoch 00005: LearningRateScheduler setting learning rate to 0.001.\n",
      "25/25 [==============================] - 6s 236ms/step - loss: 0.4952 - acc: 0.7450 - val_loss: 0.5296 - val_acc: 0.7251\n",
      "Epoch 6/20\n",
      "\n",
      "Epoch 00006: LearningRateScheduler setting learning rate to 0.001.\n",
      "25/25 [==============================] - 6s 233ms/step - loss: 0.4593 - acc: 0.7500 - val_loss: 0.4585 - val_acc: 0.7251\n",
      "Epoch 7/20\n",
      "\n",
      "Epoch 00007: LearningRateScheduler setting learning rate to 0.001.\n",
      "25/25 [==============================] - 6s 238ms/step - loss: 0.4234 - acc: 0.8000 - val_loss: 0.4552 - val_acc: 0.7241\n",
      "Epoch 8/20\n",
      "\n",
      "Epoch 00008: LearningRateScheduler setting learning rate to 0.001.\n",
      "25/25 [==============================] - 6s 235ms/step - loss: 0.4341 - acc: 0.7900 - val_loss: 0.4486 - val_acc: 0.8487\n",
      "Epoch 9/20\n",
      "\n",
      "Epoch 00009: LearningRateScheduler setting learning rate to 0.001.\n",
      "25/25 [==============================] - 6s 236ms/step - loss: 0.3799 - acc: 0.8550 - val_loss: 0.4670 - val_acc: 0.7921\n",
      "Epoch 10/20\n",
      "\n",
      "Epoch 00010: LearningRateScheduler setting learning rate to 0.0001.\n",
      "25/25 [==============================] - 6s 236ms/step - loss: 0.3780 - acc: 0.8850 - val_loss: 0.4532 - val_acc: 0.8190\n",
      "Epoch 11/20\n",
      "\n",
      "Epoch 00011: LearningRateScheduler setting learning rate to 0.0001.\n",
      "25/25 [==============================] - 6s 238ms/step - loss: 0.3919 - acc: 0.8650 - val_loss: 0.4328 - val_acc: 0.8458\n",
      "Epoch 12/20\n",
      "\n",
      "Epoch 00012: LearningRateScheduler setting learning rate to 0.0001.\n",
      "25/25 [==============================] - 6s 235ms/step - loss: 0.3808 - acc: 0.8450 - val_loss: 0.4132 - val_acc: 0.8688\n",
      "Epoch 13/20\n",
      "\n",
      "Epoch 00013: LearningRateScheduler setting learning rate to 0.0001.\n",
      "25/25 [==============================] - 6s 235ms/step - loss: 0.3723 - acc: 0.8550 - val_loss: 0.4009 - val_acc: 0.8908\n",
      "Epoch 14/20\n",
      "\n",
      "Epoch 00014: LearningRateScheduler setting learning rate to 0.0001.\n",
      "25/25 [==============================] - 6s 240ms/step - loss: 0.3581 - acc: 0.9200 - val_loss: 0.3897 - val_acc: 0.9080\n",
      "Epoch 15/20\n",
      "\n",
      "Epoch 00015: LearningRateScheduler setting learning rate to 0.0001.\n",
      "25/25 [==============================] - 6s 236ms/step - loss: 0.3851 - acc: 0.8450 - val_loss: 0.3882 - val_acc: 0.9071\n",
      "Epoch 16/20\n",
      "\n",
      "Epoch 00016: LearningRateScheduler setting learning rate to 0.0001.\n",
      "25/25 [==============================] - 6s 236ms/step - loss: 0.3604 - acc: 0.9050 - val_loss: 0.3807 - val_acc: 0.9128\n",
      "Epoch 17/20\n",
      "\n",
      "Epoch 00017: LearningRateScheduler setting learning rate to 0.0001.\n",
      "25/25 [==============================] - 6s 233ms/step - loss: 0.3653 - acc: 0.8750 - val_loss: 0.3799 - val_acc: 0.9119\n",
      "Epoch 18/20\n",
      "\n",
      "Epoch 00018: LearningRateScheduler setting learning rate to 0.0001.\n",
      "25/25 [==============================] - 6s 236ms/step - loss: 0.3470 - acc: 0.8800 - val_loss: 0.3769 - val_acc: 0.9138\n",
      "Epoch 19/20\n",
      "\n",
      "Epoch 00019: LearningRateScheduler setting learning rate to 0.0001.\n",
      "25/25 [==============================] - 6s 233ms/step - loss: 0.3427 - acc: 0.8750 - val_loss: 0.3778 - val_acc: 0.9119\n",
      "Epoch 20/20\n",
      "\n",
      "Epoch 00020: LearningRateScheduler setting learning rate to 1.0000000000000003e-05.\n",
      "25/25 [==============================] - 6s 236ms/step - loss: 0.3726 - acc: 0.8650 - val_loss: 0.3784 - val_acc: 0.9100\n"
     ]
    }
   ],
   "source": [
    "lr=keras.callbacks.LearningRateScheduler(step_decay,verbose=1)\n",
    "    \n",
    "# This will do preprocessing and realtime data augmentation:\n",
    "tbCallback=keras.callbacks.TensorBoard(log_dir='./Logs/Tensorboard_Logs', histogram_freq=0, \n",
    "                                           batch_size=16, write_graph=False, write_grads=False, \n",
    "                                           write_images=False, embeddings_freq=0, embeddings_layer_names=None, \n",
    "                                           embeddings_metadata=None)\n",
    "checkpoint=keras.callbacks.ModelCheckpoint('./Logs/Checkpoint', monitor='val_loss', \n",
    "                                               verbose=0, save_best_only=True, \n",
    "                                               save_weights_only=False, \n",
    "                                               mode='auto',\n",
    "                                               period=1)\n",
    "\n",
    "datagen = ImageDataGenerator(\n",
    "        featurewise_center=False,  # set input mean to 0 over the dataset\n",
    "        samplewise_center=False,  # set each sample mean to 0\n",
    "        featurewise_std_normalization=False,  # divide inputs by std of the dataset\n",
    "        samplewise_std_normalization=False,  # divide each input by its std\n",
    "        zca_whitening=False,  # apply ZCA whitening\n",
    "        rotation_range=0,  # randomly rotate images in the range (degrees, 0 to 180)\n",
    "        width_shift_range=0.2,  # randomly shift images horizontally (fraction of total width)\n",
    "        height_shift_range=0.2,  # randomly shift images vertically (fraction of total height)\n",
    "        horizontal_flip=False,  # randomly flip images\n",
    "        vertical_flip=False)  # randomly flip images\n",
    "\n",
    "datagen.fit(data_train)\n",
    "    # Fit the model on the batches generated by datagen.flow().\n",
    "history = model.fit_generator(datagen.flow(data_train,labels_train,batch_size=8),steps_per_epoch=25,epochs=20, validation_data=(data_val,labels_val),workers=4,callbacks=[tbCallback, checkpoint,lr])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.title('Model Accuracy')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.plot(history.history['acc'])\n",
    "plt.plot(history.history['val_acc'])\n",
    "plt.legend(['train','test'],loc='best')\n",
    "plt.show\n",
    "plt.savefig('Comparision.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "624/624 [==============================] - 1s 2ms/step\n",
      "0.43035589769864696\n",
      "0.8910256410256411\n"
     ]
    }
   ],
   "source": [
    "import itertools\n",
    "score=model.evaluate(cv_img2,ytest,verbose=1)\n",
    "print(score[0])\n",
    "print(score[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.save('model_Dense_89.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion matrix, without normalization')\n",
    "\n",
    "    print(cm)\n",
    "\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    fmt = '.2f' if normalize else 'd'\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, format(cm[i, j], fmt),\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.43035589769864696\n",
      "0.8910256410256411\n"
     ]
    }
   ],
   "source": [
    "import itertools\n",
    "score=model.evaluate(cv_img2,ytest,verbose=2)\n",
    "print(score[0])\n",
    "print(score[1])\n",
    "cm=confusion_matrix(np.argmax(ytest,axis=1), np.argmax(model.predict(cv_img2),axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix, without normalization\n",
      "[[189  45]\n",
      " [ 23 367]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVcAAAEmCAYAAADWT9N8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4xLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvDW2N/gAAIABJREFUeJzt3XmcnePdx/HPdyYr2RCJiCUeYklpIom11FaKaqna0rRFtXl4bC1dtKWW0qpWtWprVAlqrVprD1o7QRIJSogtQhJLJERk+T1/3NfEyWRmzsnknDlzz3zfXvdrzrnu7TcZ85vrXPe1KCIwM7Pyqql2AGZmbZGTq5lZBTi5mplVgJOrmVkFOLmamVWAk6uZWQU4uVqLktRV0q2SZku6fgWuM1LS3eWMrRok3SHp4GrHYeXn5GoNkvRNSeMkzZU0PSWB7cpw6f2AvsBqEbF/cy8SEX+PiN3KEM9SJO0oKSTdWK98cCp/oMTrnCLpymLHRcQeETGmmeFaK+bkasuQdBzwR+DXZIlwHeACYO8yXH5d4MWIWFiGa1XKTGAbSasVlB0MvFiuGyjj37+2LCK8eVuyAT2BucD+TRzTmSz5vpW2PwKd074dgTeB44EZwHTg0LTvVOBTYEG6x2HAKcCVBdceAATQIb0/BHgFmANMBUYWlD9UcN62wJPA7PR124J9DwC/Ah5O17kb6N3I91YX/0XAkamsFpgG/BJ4oODYPwFvAB8CTwHbp/Ld632fEwriOCPFMQ/YIJV9L+2/ELih4Pq/BcYCqvb/F96Wf/NfTqtvG6ALcGMTx/wC2BoYAgwGtgROLNi/BlmS7k+WQM+XtEpEnExWG742IrpFxCVNBSJpZeBcYI+I6E6WQMc3cNyqwL/SsasBfwD+Va/m+U3gUKAP0An4UVP3Bi4HvpNefxmYRPaHpNCTZP8GqwJXAddL6hIRd9b7PgcXnPNtYBTQHXit3vWOBzaTdIik7cn+7Q6OlGktX5xcrb7VgFnR9Mf2kcBpETEjImaS1Ui/XbB/Qdq/ICJuJ6u9bdTMeBYDm0rqGhHTI2JyA8d8BXgpIq6IiIURcTXwAvDVgmMujYgXI2IecB1ZUmxURDwCrCppI7Ike3kDx1wZEe+me55NVqMv9n1eFhGT0zkL6l3vY7J/xz8AVwJHR8SbRa5nrZSTq9X3LtBbUocmjlmTpWtdr6WyJdeol5w/BrotbyAR8RFwIHA4MF3SvyRtXEI8dTH1L3j/djPiuQI4CtiJBmrykn4k6fnU8+EDstp67yLXfKOpnRHxOFkziMj+CFhOOblafY8C84F9mjjmLbIHU3XWYdmPzKX6CFip4P0ahTsj4q6I2BXoR1YbvbiEeOpimtbMmOpcAfwfcHuqVS6RPrb/BDgAWCUiepG196ou9Eau2eRHfElHktWA30rXt5xycrWlRMRssgc350vaR9JKkjpK2kPSWemwq4ETJa0uqXc6vmi3o0aMB74oaR1JPYGf1e2Q1FfS3qntdT5Z88LiBq5xO7Bh6j7WQdKBwCDgtmbGBEBETAV2IGtjrq87sJCsZ0EHSb8EehTsfwcYsDw9AiRtCJwOfIuseeAnkppsvrDWy8nVlpHaD48je0g1k+yj7FHATemQ04FxwETgWeDpVNace90DXJuu9RRLJ8SaFMdbwHtkie6IBq7xLrAX2QOhd8lqfHtFxKzmxFTv2g9FREO18ruAO8m6Z70GfMLSH/nrBki8K+npYvdJzTBXAr+NiAkR8RLwc+AKSZ1X5Huw6pAfRJqZlZ9rrmZmFeDkamZWAU6uZmYV4ORqZlYBTXUUt2bo2mOV6Nmnf/EDraLW6OEH7K3FhGeenhURq5frerU91o1YOK/ocTFv5l0RsXu57ru8nFzLrGef/oz8wz+qHUa7d8JO61c7BEv69OhUf/TcComF8+i80QFFj/tk/PnFRstVlJOrmeWLBDW11Y6iKCdXM8ufHEyF6+RqZvkjFT+mypxczSxn5JqrmVnZiVy0ubb+9G9mthRlzQLFtqauIHWR9ISkCZImSzo1lV8maaqk8Wkbksol6VxJUyRNlDS0WJSuuZpZ/qx4s8B8YOeImCupI/CQpDvSvh9HRP3+lHsAA9O2Fdl6Z1s1dQMnVzPLmRXvipXWJZub3nZMW1NTBO4NXJ7Oe0xSL0n9ImJ6Yye4WcDM8kWU2izQW9K4gm3UUpeRaiWNJ1ul+J60xA7AGemj/zkFc+n2Z+n5et9k6WWEluGaq5nlT2nNArMiYnhjOyNiETBEUi/gRkmbkq2E8TbZCsGjgZ8CpzUnRNdczSxnBLW1xbcSRcQHwP3A7mmF4YiI+cClZMvGQ7Ye29oFp61FkTXanFzNLF9EVnMttjV1iWz9t17pdVdgV+AFSf1SmcgW6ZyUTrkF+E7qNbA1MLup9lZws4CZ5dGKj9DqB4yRVEtWybwuIm6TdJ+k1clS+HiyZd0hWwRzT2AK2dLshxa7gZOrmeXMio/QioiJwOYNlO/cyPEBHLk893ByNbP8ycEILSdXM8uXEkZgtQZOrmaWP564xcys3DxZtplZZbhZwMyszOr6ubZyTq5mljNuFjAzqwzXXM3MKsBtrmZmZSavoWVmVhGqcXI1MyurbK5sNwuYmZWX0tbKObmaWc6IGjcLmJmVn5sFzMwqwMnVzKzMJKEaJ1czs7JzzdXMrAKcXM3Myk3kolmg9fdnMDOrR1LRrcj5XSQ9IWmCpMmSTk3l60l6XNIUSddK6pTKO6f3U9L+AcVidHI1s1wRxRNrCc0G84GdI2IwMATYXdLWwG+BcyJiA+B94LB0/GHA+6n8nHRck5xczSx3VjS5RmZuetsxbQHsDPwjlY8B9kmv907vSft3UZGbOLmaWb6kNtdiG9Bb0riCbdRSl5FqJY0HZgD3AC8DH0TEwnTIm0D/9Lo/8AZA2j8bWK2pMP1Ay8xyp8TeArMiYnhjOyNiETBEUi/gRmDjMoUHuOZqZjlUhjbXJSLiA+B+YBugl6S6SudawLT0ehqwdrp3B6An8G5T13XNtR06cMgabNK3G3PnL+L3D0wFYM0endnv82vQoVYsjuCGie/wxgef0LVjDQcO6cdqK3dk4aLg2vHTeXvOp1X+DtquRYsWsesOW9OvX3/+fv1NHH34YTz68IN079EDgHMv/CubfX5IlaOsLrHiI7QkrQ4siIgPJHUFdiV7SHU/sB9wDXAwcHM65Zb0/tG0/76IiKbu4eTaDj35+mwemvo+IzZfc0nZXoP6cPeLs3hhxkds3Gdl9hrUhwsfeZ1dBq7GW7M/4bInp9GnWyf23awvFz36RhWjb9tGX/hnNtxwY+bMmbOk7ORf/Yav7vONKkbVyqgsgwj6AWMk1ZJ9gr8uIm6T9BxwjaTTgWeAS9LxlwBXSJoCvAccVOwGTq7t0CvvzWOVrh3rlQZdOmStRF071vDhJwsA6Nu9M/e9lH36mTH3U1ZZqSPdOtcyd/6ilgy5XXhr2pvce9cd/OBHJ3DReX+qdjit2oom14iYCGzeQPkrwJYNlH8C7L8893CbqwFw06QZ7DWoDyftuj5fHdSH25+fCcBbsz9hs37dAVi7VxdW6dqRXl38N7kSTjzheH552m+Wmav016f9kh22GcpJJ/yI+fPnVym61qXE3gJVlZvkKikknV3w/keSTmnhGC6TtF9L3rOlbDugFzdPnsGv7nmZmyfP4IAh/QC4b8p7dO1Yw3E7DGC79VZh2uxPWNxkS5M1x913/IvevfswePOhS5WfeMrpPPLUJO5+4FHef/89/nzO76oUYetSzgdalZKb5Eo2omJfSb2bc3LBE0BrwPC1e/Ls9Kydb8Jbc1inVxcA5i9czLXj3+YP/36Vq5+ZTrfOHXj34wXVDLVNeuLxR7jrjtsYtulARh36LR76z/0c8b2D6btGPyTRuXNnRnzrYJ55aly1Q626UhKrk+vyWQiMBn5Yf4ekAZLukzRR0lhJ66TyyyRdJOlx4CxJp0gaI+lBSa9J2lfSWZKelXSnpI7pvF9KelLSJEmji43EaAs+/GQh66+2EgADe6/EzI+yBNqlQw216bvfap2evPLux8xfuLhaYbZZJ55yBhNemMpTk15i9KVXst0Xd+LCv47hnbenAxAR3HHbLWw8aFCVI20d8pBc81abOx+YKOmseuV/BsZExBhJ3wXO5bNha2sB20bEotSMsD6wEzCIrFvFNyLiJ5JuBL4C3AScFxGnAUi6AtgLuLWy31rL+dbQNVm/90qs3KmWk3Zdn7v+O4vrJ7zN3pv2pVawYHHwjwnZL3Xf7p0YsfmaBMHbcz7luvHTqxx9+3LE9w7m3VkziQg+t9lgfvfH86sdUqvQGtpUi8lVco2IDyVdDhwDzCvYtQ2wb3p9BVCYfK9PIzHq3BERCyQ9C9QCd6byZ4EB6fVOkn4CrASsCkymieSahtWNAui++pqNHdZqXPn0Ww2W//E/ry5T9tr7n3Dmfa9UOCIr9IXtd+AL2+8AwD9vu7vK0bROraFmWkyemgXq/JFshpqVSzz+o3rv5wNExGKyTsR1j2cWAx0kdQEuAPaLiM2Ai4EuTd0gIkZHxPCIGL5Sj1VKDMvMmkX5aBbIXXKNiPeA6/hsKjCAR/isU+9I4MEVuEVdIp0lqRvZaAwzayWEqKkpvlVb7pJrcjZQ2GvgaOBQSROBbwPHNvfCaZzxxcAk4C7gyRWI08wqQCq+VVtu2lwjolvB63fI2kPr3r9GNg9j/XMOqff+lCaueUrB6xOBE4tdz8yqozV87C8mN8nVzAyyWmltrZOrmVnZ5aDi6uRqZvnjZgEzs3JrJQ+sinFyNbNcybpitf6OTk6uZpY7rrmamVWA21zNzMpMolWMwCrGydXMcicHFdfcDn81s3ZsRSdukbS2pPslPSdpsqRjU/kpkqZJGp+2PQvO+ZmkKZL+K+nLxWJ0zdXM8qU8zQILgeMj4mlJ3YGnJN2T9p0TEb9f6pbSILLJoT4HrAncK2nDetOZLsU1VzPLFbHiE7dExPSIeDq9ngM8D/Rv4pS9gWsiYn5ETAWm0MAqsYWcXM0sZ0peQ6u3pHEF26gGryYNIFtm+/FUdFRaMupvkuomaO4PvFFw2ps0nYydXM0sf0qsuc6qm8Q+baOXvY66ATcAP4iID4ELyZaCGgJMJ5vetFnc5mpm+VKmrlhpQdIbgL9HxD9hyXSmdfsvBm5Lb6cBaxecvlYqa5RrrmaWK1mb6wr3FhBwCfB8RPyhoLxfwWFfJ5s0H+AW4CBJnSWtBwwEnmjqHq65mlnulGGE1hfIVi15VtL4VPZzYISkIUAArwL/CxARkyVdBzxH1tPgyKZ6CoCTq5nl0Io2C0TEQ2SV4Ppub+KcM4AzSr2Hk6uZ5YunHDQzKz/ROpbOLsbJ1cxypzbPE7dI6tHUialPmJlZi8tBxbXJmutksidmhd9G3fsA1qlgXGZmDcoGCbT+7Npoco2ItRvbZ2ZWTTloFShtEIGkgyT9PL1eS9KwyoZlZta4mhoV3aqtaHKVdB6wE1mHW4CPgYsqGZSZWWNE6jFQ5L9qK6W3wLYRMVTSMwAR8Z6kThWOy8ysUa2gYlpUKcl1gaQasodYSFoNWFzRqMzMGqPW8bG/mFLaXM8nmzlmdUmnAg8Bv61oVGZmjRBQIxXdqq1ozTUiLpf0FPClVLR/RExq6hwzs0pqBbmzqFJHaNUCC8iaBjxNoZlVTV6W1i6lt8AvgKvJFuVaC7hK0s8qHZiZWWPaRLMA8B1g84j4GEDSGcAzwG8qGZiZWWOqnzqLKyW5Tq93XIdUZmZWFbke/irpHLI21veAyZLuSu93A55smfDMzJYmKd+zYvHZ2jGTgX8VlD9WuXDMzIrLQcW1yYlbLmnJQMzMSpXrZoE6ktYnWzdmENClrjwiNqxgXGZmDRL5mCy7lD6rlwGXkn1PewDXAddWMCYzsyaphK3J86W1Jd0v6TlJkyUdm8pXlXSPpJfS11VSuSSdK2mKpImShhaLsZTkulJE3AUQES9HxIlkSdbMrMVJZennuhA4PiIGAVsDR0oaBJwAjI2IgcDY9B6ynDcwbaOAC4vdoJSuWPPTxC0vSzocmAZ0L+E8M7OKKMPS2tNJXUojYo6k54H+wN7AjumwMcADwE9T+eUREcBjknpJ6peu06BSkusPgZWBY8jaXnsC323ON2RmVg4lPs/qLWlcwfvRETF62WtpALA58DjQtyBhvg30Ta/7A28UnPZmKmt+co2Ix9PLOXw2YbaZWVWIkoe3zoqI4U1eS+pGNuvfDyLiw8JeCBERkqK5cTY1iOBG0hyuDYmIfZt707asf88unLHnxtUOo91bZYujqh2CVYrK089VUkeyxPr3iPhnKn6n7uO+pH7AjFQ+DShcV3CtVNaopmqu5zUzZjOziqpdweyqrIp6CfB8RPyhYNctwMHAmenrzQXlR0m6BtgKmN1Ueys0PYhg7ArEbmZWEaIsgwi+QNbM+ayk8ans52RJ9TpJhwGvAQekfbcDewJTyNYRPLTYDUqdz9XMrNVY0TEEEfEQjXeH3aWB4wM4cnnu4eRqZrki5WOEVsnJVVLniJhfyWDMzEqRg9xa0koEW0p6FngpvR8s6c8Vj8zMrBFS8a3aShn+ei6wF/AuQERMAHaqZFBmZo0R0EEqulVbKc0CNRHxWr2nc4sqFI+ZWVGtIHcWVUpyfUPSlkBIqgWOBl6sbFhmZg1TK1mAsJhSkusRZE0D6wDvAPemMjOzqshBbi1pboEZwEEtEIuZWVECOuSgu0ApKxFcTANzDETEqIpEZGZWRJuouZI1A9TpAnydpafeMjNrOcpHP9dSmgWWWtJF0hXAQxWLyMysCWLFJ25pCc0Z/roen00ga2bW4tpEzVXS+3zW5loDvMdn68qYmbW43C+tneY8HMxnk8IuTrPDmJlVRTZxS7WjKK7JEFMivT0iFqXNidXMqq4Mq79WPsYSjhkvafOKR2JmVgKRtbkW26qtqTW0OkTEQrJVEZ+U9DLwEdn3FhExtIViNDNbSiuomBbVVJvrE8BQ4GstFIuZWVFCue+KJYCIeLmFYjEzK66VfOwvpqnkurqk4xrbWW/FRDOzFlOOB1aS/kY2V/WMiNg0lZ0CfB+YmQ77eUTcnvb9DDiMbMrVYyLirqau31RyrQW60fgiXmZmLU6UbQ2ty4DzgMvrlZ8TEb9f6p7SILIJrD4HrAncK2nDiGh0buumkuv0iDitWSGbmVVQOZpcI+I/kgaUePjewDVpHcGpkqYAWwKPNnZCU12xXGM1s1ZHZImr2LYCjpI0UdLfJK2Syvqz9IRVb6ayRjUVwzJrd5uZVZ1KHkTQW9K4gq2UaVIvBNYHhgDTgbObG2ajzQIR8V5zL2pmVinZIIKSPljPiojhy3PtiHhnyX2yuaxvS2+nAWsXHLoWn00L0KAcjNA1M1uaStiadV2pX8HbrwOT0utbgIMkdZa0HjCQbCxAo5oz5aCZWVWV44GWpKuBHcmaD94ETgZ2lDSEbCbAV4H/BYiIyZKuA54DFgJHNtVTAJxczSxnyjVCKyJGNFB8SRPHnwGcUer1nVzNLHdyP5+rmVlr1PpTq5OrmeWM1HbX0DIzqyo3C5iZVUDrT61OrmaWM215aW0zs6rKQW51cjWzvBHKQcOAk6uZ5Y5rrmZmZeauWGZmFZKD3Ork2t698cYbfO/Q7zBjxjtI4ruHjeKoY47l1JNP4rZbbqampobV+/Rh9CWXseaaa1Y73Dalc6cO3HvJD+jUqQMdamu58d5nOP2i2wE45civsu+um7No0WIu/seDXHD1v/nhd3bhwD23AKBDbQ0br7cGa+98Au9/+HE1v42qcJurtXodOnTgzLPOZvOhQ5kzZw7bbjWMXb60Kz88/secfOqvADj/z+fym9NP488XXFTlaNuW+Z8uZPdR5/LRvE/p0KGG+/52HHc//BwbrbcGa63Ri8Ff/xURweqrdAPgnMvHcs7lYwHY84ubcvTIndppYs1Hs4Dnc23n+vXrx+ZDhwLQvXt3Nt54E956axo9evRYcszHH3+UixExefTRvE8B6Nihlg4daokIRu2/Hb8efQcRAcDM9+cuc94Buw/nujufatFYWxOp+FZtrrnaEq+9+irjxz/DFltuBcDJJ/2Cv195OT179uTOe+6vcnRtU02NeOSqn7L+2qvzl2v/w5OTXmO9tVZnv92G8bWdBzPr/Tkcf9Y/ePn1mUvO6dqlI7tuuwk/PPO6KkZeXXloFqhYzVXSIknjJU2SdL2klSp1r3KStKakf1Q7jpY2d+5cRhzwDX539h+X1FpP/dUZTJn6BgeNGMlFF5xX5QjbpsWLg60POpMNvnwiwzddl0Hr96Nzpw7M/3QB2408i0v/+Qh/OXnkUud85Yub8ej4V9plkwB8Np9rsa3aKtksMC8ihkTEpsCnwOEVvFfZRMRbEbFfteNoSQsWLGDEAd/gwBEj2efr+y6z/8ARI7npxhuqEFn7MXvuPP497kV223YQ0955n5vGTgDg5vsmsOnApRcZ3f/Lw7i+HTcJUEKTQCvIrS3W5vogsIGkAZKel3SxpMmS7pbUFUDS+pLulPSUpAclbZzKL5O0JNlJmpu+7ijp35JulvSKpDMljZT0hKRnJa2fjhsg6b60VO5YSesUXPdcSY+k8/crOH5SwesHJT2dtm1b6N+rxUQEh3//MDbaeBOO/eFxS8qnvPTSkte33XIzG260cTXCa9N6r9KNnt26AtClc0d22Wpj/vvqO9z6wER22GIgANsPG8iU12csOadHty5sN2wDbn1gYlVibi0qtYZWOVW8zVVSB2AP4M5UNBAYERHfT2vSfAO4EhgNHB4RL0naCrgA2LnI5QcDmwDvAa8Af42ILSUdCxwN/AD4MzAmIsZI+i5wLrBPOr8fsB2wMdkCZPWbA2YAu0bEJ5IGAlcDy6wmmZbsHQWw9jrrlPCv0no88vDDXPX3K9h0083YatgQAE49/ddcduklvPTif6lRDeusuy7nnu+eAuW2Ru8eXHzat6mtqaGmRtxwz9Pc8eAkHnnmZS799cEcPXJnPpo3nyNOu2rJOV/baTBjH3uBjz/5tIqRV9dyrP5aVZVMrl0ljU+vHyRbm2ZNYGpE1JU/BQyQ1A3YFri+4Kl05xLu8WRETAeQ9DJwdyp/Ftgpvd4GqPusewVwVsH5N0XEYuA5SX0buH5H4Ly0YNkiYMOGgoiI0WR/HBg2bHiUEHer8YXttmPegmVD3n2PPasQTfsy6aW32GbEb5cpnz13Hvse0/AfsytvfZwrb3280qG1ejnIrRVNrvMiYkhhQUqc8wuKFgFdyZonPqh/fLIw7UdSDdCpYF/htRYXvF9Mad9b4fkN/bh+CLxDVkOuAT4p4ZpmVmHl6C0g6W/AXsCM9GwISasC1wIDyFZ/PSAi3leWvP4E7Al8DBwSEU83df1W0c81Ij4EpkraH0CZwWn3q8Cw9PprZLXJ5fEIcFB6PZKsFl2qnsD0VLv9NlC7nPc2swoo0wOty4Dd65WdAIyNiIHA2PQesqbNgWkbBVxY7OKtIrkmI4HDJE0AJgN7p/KLgR1S+TbAR8t53aOBQyVNJEuQxy7HuRcAB6d7b9yMe5tZBZQjuUbEf8ie1xTaGxiTXo/hs+czewOXR+YxoJekfk1dv2LNAhHRrYGyV4FNC97/vuD1VJb9K0JEvANsXVD001T+APBAwXE7Frxesi8iXqOBB2MRcUhD8RbGGBEvAZ+vf28zq56sN0BJVdPeksYVvB+dno80pW/dcxzgbaDuWUx/4I2C495MZdNphEdomVm+lP6xf1ZELNO7p1QREZKa/YC6NTULmJmVpIKDCN6p+7ifvtZ1Mp4GrF1w3FqprFFOrmaWMyrpv2a6BTg4vT4YuLmg/DvpYfvWwOyC5oMGuVnAzHKnHP1cJV0N7EjWNvsmcDJwJnCdpMOA14AD0uG3k3XDmkLWFevQYtd3cjWzXCnX8NaIGNHIrl0aODaAI5fn+k6uZpY7eZhf2MnVzHInB7nVydXM8icHudXJ1cxyRm4WMDMrO+FmATOzishBbnVyNbP8cbOAmVkF5CC3OrmaWf7kILc6uZpZDuUguzq5mlmuSF6g0MysIlp/anVyNbM8ykF2dXI1s5yRmwXMzMqtXFMOVpqTq5nlTw6yq5OrmeWOmwXMzCqg9adWJ1czy5sVW921xTi5mlmuZFMOtv7s6uRqZrlTjtQq6VVgDrAIWBgRwyWtClwLDABeBQ6IiPebc/2aMsRoZtaipOJbiXaKiCERMTy9PwEYGxEDgbHpfbM4uZpZ7qiE/5ppb2BMej0G2Ke5F3JyNbPcKbHm2lvSuIJtVL3LBHC3pKcK9vWNiOnp9dtA3+bG6DZXM8uV5fjYP6vg435DtouIaZL6APdIeqFwZ0SEpGhunK65mlnulKNZICKmpa8zgBuBLYF3JPUDSF9nNDdGJ1czy50VfaAlaWVJ3eteA7sBk4BbgIPTYQcDNzc3RjcLmFnulKGba1/gxtRftgNwVUTcKelJ4DpJhwGvAQc09wZOrmaWMyvUGwCAiHgFGNxA+bvALit08cTJ1cxyJRuhVe0oinNyNbPccXI1M6uAFW0WaAlOrmaWL54Vy8ys/NzmamZWIW4WMDOrANdczcwqwMnVzKwC8tAsoIhmT/piDZA0k2zYXJ71BmZVOwgD2sbPYt2IWL1cF5N0J9m/SzGzImL3ct13eTm52jIkjSsyVZu1EP8s8suzYpmZVYCTq5lZBTi5WkNGVzsAW8I/i5xym6uZWQW45mpmVgFOrmZmFeDkamZWAU6uttwkbSJpZ0kdqx2LWWvl4a/WHAcBawOLJD0SEQuqHZB9RpKi3pPqhsqsstxbwJabpBrgRGAN4HrgISfY1qEwiUraEvgUeC4iPpVUGxGLqhth++FmASuJ9Nk8RBGxGDgDmA4cCGznJoLqqvv5FCTWo4FzgUOA+yV1iYhFkmqrF2X74uRqRdWrDe0maUegF3A68DpZgt3WCbaq+te9kLQ3WdPNzsBMYBNgsqSuKcH6974F+B/ZiipIrMcBJwEjyGpFO0bEr4GpwChgq6oF2Y5JWhX4naTuqWg8sD/wTWA00LuLAAAKaElEQVT7iFgVeAN4QVLn9MnDKszJ1Uoi6UvAThGxPfAeMAgYIWmniPgtMA6YUs0Y26uIeA84FNhC0v9FxGsR8RawGXBFOuwm4E1gzSqF2e64t4A1qIGny28AR0s6BNgC2BM4BzhFUseIOKcKYbZrhT+jiPhEUifgGEmfRMTfyJoEtpI0hOxntm9EvFPFkNsVJ1dbRr021q2AF4Gp6YnzQODCiJguaSIwg+xjqLWgej+j1YAFEXGnpKOAsyXNAc4ja8IZDhzjxNqynFxtGQW/tIcDPwYmA3dLugaYBIyRNBTYF9grImZULdh2qF5i/THZp4iukn4SEfemsjOBlSPiwmrG2p45udoS9X5p+wCfB7Ykq/nsChxGVhuaTfbwat+IeLlK4bZbBT+j7YEvAd8GdgNukfS1iLg79dw4QdJNwGwPIGh5HkRgwDKJ9SiyAQLbR8QOqWwPsl/kucCf0kMUqxJJ2wI/A16NiKNT2SHA74EDI2KspJUi4uMqhtmuubeAAUvVhvYFDgaeANaSdG3afwfwH6Aj5GDpzTamcBBHMgF4Cugvads0+uoyspFzf5PUFZjXwmFaAddc27l6NdZhwM+BOyPi4vQL/TTwQkSMSMesHBEfVS/i9qfez+gAoAvwcUT8Q9LJZCuhXgM8HhELJfWMiNlVDNlwzbVdq/dLux/Zg5H3gR0lDU77hgJbS7oMwIm15RX8jI4AfkLWZHOcpDERcSpZl6vvAsPSKR9WJVBbimuuhqTdyX5pv0w2VPJbwBzgpoh4Nh2zXkRMrV6U7ZuklYBrgVMjYlwqexi4m6xnwC+A893dqvVwzbWdS/MEHAE8GRELImIicDOwMvBNSZ8DcGJtWQ2M/18AfAR0Lyg7Flg9IuYDJzuxti5Oru1MAw9GppLNbvU/kgYDRMTDwJ1kv9D+ha2CuvH/knZJf+A6AbcBV0paPx22ObCepC5VCtOa4GaBdqReG+tXgYXAB2TzAvyJbM6AawuaArpGhJ84t6B6P6NDgV8DtwLdgCPJJmQ5kuxnNhT4VkRMrlK41gQn13ak7hdX0v8B3wPuIBtldSlZcj0bmA9cEhHPVS/S9qleYl2TLJHeQNa3eBSwTfpaC6wGfBgRb1QpXCvCI7TaAUnrAO9GxEdp5NUBwMiIeF7S78n6S75FNgH2T8nmC7AWVC+xHg3sCKwPPAo8SfYHELKHWke7ttr6uc21jZPUFzgeOEJStzQPwCyy5T+IiPeBHwCbRcR04McRMatqAbdT9QZx7AH8hqwt/CBgjYiYCYwhmzrQfVhzwMm17ZtJVvNZEzg0PdCaAlwjqe6Ty7pko7FqydphrYUU9gqQtAlwMvBI6m71TWAt4KeS+qfeAOdFxJvVidaWh9tc26g0NWBNRPw3JdS9yGpE4yNitKQLgcHARLJJWEa6nbV6JPWIiA8lnUS2bM5REfFAWl3gOrKf08/DCwzmhpNrG5Tm95xJ9vH/VGARMJqsJrQBMD0i/pLmau0CvO5+rNWRaq6bAw8Dm0bElDRxzu7A71OC7Qb0jIhp1YzVlo8faLVBEfFuWpblXrKmn8FkD0LmkrW1bpZqs5emDuhWJak/61OSfgM8IGmHiDhP0mLgV5J+FhEPkf3sLEecXNuoiLhP0pfJFhIcDPQlWw30ILI5WjcCribremVVIGlrgIh4LCJOlTQfeFLS8Ii4QNIC4LXqRmnN5WaBNk7SV8jWuto6It6TtArZtIErRcSrVQ2unSnoZ1xD1hxzHtkgjqsK5gu4guwP4P+4D2u+uebaxkXEv9JHzMckbRMR71Y7pvaosB8r0DsiZqS21dOA/STVRMQTZPM6rAp0rlasVh6uubYTkvYGTgGGhdetrxpJx5DNPvYh2cKO55ENce1ENinLRsDekS2NbTnm5NqOpEEEfjBSJZIOAr4P7Ec2TeA6EbGHpFXJ5mIdDtzsLnFtg5OrWQtJnx6mAV8gm5h8r4hYIGmQE2rb4xFaZhXQwNSOkH3svxfYNSK+nBLr94Aj02TY1ob4gZZZmTUwbeC6wNiIuFLSRsBeaajrbmTLlY/wKq1tj5sFzMqkXo+AumaAXwCPk63sMJ5spNyPgAGp7LSIeL7lo7VKc3I1KxNJHSJiYXq9F3ASsE9ETE+zXX2RbNKc0RHxaeHx1va4zdWsDCTtClwh6YQ0cONtssUeDwOIiH8CDwBDgFFpIIEnYWnDnFzNVlBaPfcM4BGyj/r7kc3h8DVgf0mHA0TETWTzsV4fEYvDHxvbNDcLmK2A1Ed1FlnH/1vTqg+/A66JiBsl7ZDeXxURf6xmrNayXHM1WwER8R7wVeDMNCfr62QTjvdJ+/8N/Bz4uqRejXTRsjbINVezMpC0B9kMZHeRrfowsnDlXK+k2/44uZqVSZpD926yNa9mOKG2b24WMCuTiLgX+Apwv6Q+Tqztm0domZVRRNwhqRNwp6ThWZE/HrZHbhYwqwDPQGZOrmZmFeA2VzOzCnByNTOrACdXM7MKcHI1M6sAJ1crC0mLJI2XNEnS9Ssys76kHSXdll5/TdIJTRzbS9L/NeMep0j6Uanl9Y65TNJ+y3GvAZImLW+Mlm9OrlYu8yJiSERsSjYj1OGFO5VZ7v/fIuKWiDiziUN6AcudXM0qzcnVKuFBYINUY/uvpMuBScDaknaT9Kikp1MNtxtk0/ZJekHS08C+dReSdIik89LrvpJulDQhbduSraK6fqo1/y4d92NJT0qaKOnUgmv9QtKLkh4iW8K6SZK+n64zQdIN9WrjX5I0Ll1vr3R8raTfFdz7f1f0H9Lyy8nVykpSB2AP4NlUNBC4ICI+B3wEnAh8KSKGAuOA4yR1AS4mm11qGLBGI5c/F/h3RAwGhgKTgROAl1Ot+ceSdkv33JJsYuphkr4oaRhwUCrbE9iihG/nnxGxRbrf86SJr5MB6R5fAS5K38NhwOyI2CJd//uS1ivhPtYGefirlUtXSePT6weBS8hmh3otIh5L5VsDg4CH08x7nYBHgY2BqRHxEoCkK4FRDdxjZ+A7ABGxCJgtaZV6x+yWtmfS+25kybY7cGPdQoCSbinhe9pU0ulkTQ/dyGa8qnNdRCwGXpL0SvoedgM+X9Ae2zPd+8US7mVtjJOrlcu8iBhSWJAS6EeFRcA9ETGi3nFLnbeCBPwmIv5S7x4/aMa1LiNbA2uCpEOAHQv21R/aGOneR0dEYRJG0oBm3Ntyzs0C1pIeA74gaQMASStL2hB4ARggaf103IhGzh8LHJHOrZXUE5hDViutcxfw3YK23P6S+gD/AfaR1FVSd7ImiGK6A9MldQRG1tu3v6SaFPP/AP9N9z4iHY+kDSWtXMJ9rA1yzdVaTETMTDXAqyV1TsUnRsSLkkYB/5L0MVmzQvcGLnEsMFrSYWSL+x0REY9Kejh1dbojtbtuAjyaas5zgW9FxNOSrgUmADOAJ0sI+SSyZbFnpq+FMb0OPAH0AA6PiE8k/ZWsLfbptOLATGCf0v51rK3xxC1mZhXgZgEzswpwcjUzqwAnVzOzCnByNTOrACdXM7MKcHI1M6sAJ1czswr4f2YUa41pi4CNAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm_plot_labels=[\"Normal\",\"Pneumonia\"]\n",
    "plot_confusion_matrix(cm,cm_plot_labels,title=\"Confusion Matrix\")\n",
    "plt.savefig('Confusion.eps')"
   ]
  }
 ],
 "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": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}