final_code.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "final_code.ipynb",
      "version": "0.3.2",
      "views": {},
      "default_view": {},
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "metadata": {
        "id": "gTd93S58LPyd",
        "colab_type":  "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Following code insert drive into environment**\n",
        "\n"
      ]
    },
    {
      "metadata": {
        "id": "WD8dEhwo3PMH",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "In the output cell of follwing some link would be shown. Login and allow access and the copy the password into the block below the link "
      ]
    },
    {
      "metadata": {
        "id": "eOUsQ-0YKER_",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 105
        },
        "outputId": "72483420-9267-4da5-f9e6-e7c7fd856516",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1526276527316,
          "user_tz": -330,
          "elapsed": 26055,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "!apt-get install -y -qq software-properties-common python-software-properties module-init-tools\n",
        "!add-apt-repository -y ppa:alessandro-strada/ppa 2>&1 > /dev/null\n",
        "!apt-get update -qq 2>&1 > /dev/null\n",
        "!apt-get -y install -qq google-drive-ocamlfuse fuse\n",
        "from google.colab import auth\n",
        "auth.authenticate_user()\n",
        "from oauth2client.client import GoogleCredentials\n",
        "creds = GoogleCredentials.get_application_default()\n",
        "import getpass\n",
        "!google-drive-ocamlfuse -headless -id={creds.client_id} -secret={creds.client_secret} < /dev/null 2>&1 | grep URL\n",
        "vcode = getpass.getpass()\n",
        "!echo {vcode} | google-drive-ocamlfuse -headless -id={creds.client_id} -secret={creds.client_secret}"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Please, open the following URL in a web browser: https://accounts.google.com/o/oauth2/auth?client_id=32555940559.apps.googleusercontent.com&redirect_uri=urn%3Aietf%3Awg%3Aoauth%3A2.0%3Aoob&scope=https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fdrive&response_type=code&access_type=offline&approval_prompt=force\r\n",
            "··········\n",
            "Please, open the following URL in a web browser: https://accounts.google.com/o/oauth2/auth?client_id=32555940559.apps.googleusercontent.com&redirect_uri=urn%3Aietf%3Awg%3Aoauth%3A2.0%3Aoob&scope=https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fdrive&response_type=code&access_type=offline&approval_prompt=force\n",
            "Please enter the verification code: Access token retrieved correctly.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "-Kwu1rnVKIoX",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "!mkdir -p drive\n",
        "!google-drive-ocamlfuse drive"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "VUneXRozLqoM",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Check if shows drive folder**"
      ]
    },
    {
      "metadata": {
        "id": "dS6nb0btK9Py",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "22d4d2d7-a6fe-4af2-ccaf-4a1aedfa9249",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525867886189,
          "user_tz": -330,
          "elapsed": 2127,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "!ls"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "datalab  drive\r\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "_ZSfcgWcLwRo",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**switch into drive folder(place code and data files into same folder and move to that folder simillar to cd)**"
      ]
    },
    {
      "metadata": {
        "id": "3PLyj0IFL_ug",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "import os\n",
        "os.chdir('drive/SOP')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "sWRc8JKdMbn1",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 102
        },
        "outputId": "31e669b6-f566-4a1f-b983-4f1dd261ac03",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525867901643,
          "user_tz": -330,
          "elapsed": 2314,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "!ls"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "100epochs_3channel.h5  model.ipynb\t\t   tranfer_20ep_l2reg.h5\r\n",
            "data_1.h5\t       mycode.ipynb\t\t   tranfer_20ep_overfit.h5\r\n",
            "data.h5\t\t       Testdata_SigComp2011\t   trial.ipynb\r\n",
            "data_s_gen.h5\t       trainingSet\r\n",
            "data_siamese.h5        tranfer_20ep_l2+dropout.h5\r\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "FsNOpcArKVNk",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Data preprocessing(Optinal as all files are already created)**"
      ]
    },
    {
      "metadata": {
        "id": "h9e3QTSWKYR_",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "!wget http://www.iapr-tc11.org/dataset/ICDAR_SignatureVerification/SigComp2011/sigComp2011-trainingSet.zip"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "65wvZwwnKm3s",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "import zipfile\n",
        "zip_ref = zipfile.ZipFile('sigComp2011-test.zip', 'r')\n",
        "zip_ref.extractall('drive',pwd='I hereby accept the SigComp 2011 disclaimer.')\n",
        "zip_ref.close()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "zXJawpuzKnDF",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "from PIL import Image\n",
        "import pickle\n",
        "import os\n",
        "import pickle\n",
        "import cv2"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "hRQ72BvTKnG2",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "def Center(img, new_height, new_width):\n",
        "    cImg = 255*np.ones((new_height,new_width,3))\n",
        "    width =  np.size(img,1)\n",
        "    height = np.size(img,0)\n",
        "\n",
        "    left = int((new_width - width)/2.)\n",
        "    top = int(((new_height - height)/2.))\n",
        "    right = int(((width + new_width)/2.))\n",
        "    bottom = int(((height + new_height)/2.))\n",
        "    cImg[top:bottom, left:right,:] = img\n",
        "    return cImg"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "pbf_xBAhKnKR",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "x = []\n",
        "y = []\n",
        "\n",
        "j = k = 1\n",
        "arr = []\n",
        "path = os.listdir('trainingSet/OfflineSignatures/Chinese/TrainingSet/Offline Genuine/')\n",
        "path.sort(key=str.lower)\n",
        "\n",
        "for i in range(len(path)):\n",
        "    if(path[i][2] == str(j)):\n",
        "        img = cv2.imread('trainingSet/OfflineSignatures/Chinese/TrainingSet/Offline Genuine/'+path[i]+'')\n",
        "        #img  = cv2.cvtColor(img,cv2.COLOR_RGB2GRAY)\n",
        "        a = np.asarray(img)\n",
        "        a = Center(a,551,1117)\n",
        "        x.append(a)\n",
        "        y1 = np.zeros((10))\n",
        "        y1[j-1] = 1\n",
        "        y.append(y1)\n",
        "        k =k + 1\n",
        "    elif(j<10):\n",
        "        j = j + 1\n",
        "        print('j='+str(j))\n",
        "    else:\n",
        "        img = cv2.imread('trainingSet/OfflineSignatures/Chinese/TrainingSet/Offline Genuine/'+path[i]+'')\n",
        "        #img = cv2.cvtColor(img,cv2.COLOR_RGB2GRAY)\n",
        "        a = np.asarray(img)\n",
        "        a = Center(a,551,1117)\n",
        "        x.append(a)\n",
        "        y1 = np.zeros((10))\n",
        "        y1[j-1] = 1\n",
        "        y.append(y1)\n",
        "        k =k+1\n",
        "print('k='+str(k))        "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "HyNjLnC6K8Be",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "outputId": "4fdbfb4a-6bcb-47fa-c902-d46338eb2156",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525873944925,
          "user_tz": -330,
          "elapsed": 2238,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "from matplotlib import pyplot as plt\n",
        "plt.imshow(np.uint8(x[0]))\n",
        "plt.show()"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeQAAAD/CAYAAAAgyKUVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7ff4c1f7de80>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "3LdDSMDeK8Iw",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "hf = h5py.File('data_1.h5', 'w')\n",
        "hf.create_dataset('dataset_1', data=x)\n",
        "hf.create_dataset('dataset_2', data=y)\n",
        "hf.close()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "iI4hACAbLNVh",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Dataset for siamese networks"
      ]
    },
    {
      "metadata": {
        "id": "pacJPWOdKnON",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "x = []\n",
        "\n",
        "a = b = 0 \n",
        "path = os.listdir('trainingSet/OfflineSignatures/Chinese/TrainingSet/Offline Genuine/')\n",
        "path.sort(key = str.lower)\n",
        "\n",
        "for i in range(10):\n",
        "  y = []\n",
        "  for j in range(b,len(path)):\n",
        "    if(path[j][3:6] == str(i+1).zfill(2)):\n",
        "        img = cv2.imread('trainingSet/OfflineSignatures/Chinese/TrainingSet/Offline Genuine/'+path[j]+'')\n",
        "        #img  = cv2.cvtColor(img,cv2.COLOR_RGB2GRAY)\n",
        "        arr = np.asarray(img)\n",
        "        arr = Center(arr,551,1117)\n",
        "        y.append(arr)\n",
        "        a = a + 1\n",
        "        print(a)\n",
        "        if(a==235):\n",
        "          x.append(y)\n",
        "    else:\n",
        "      print('...............')\n",
        "      x.append(y)\n",
        "      b = a\n",
        "      break"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Iuo59QSdLUu7",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "hf = h5py.File('data_1.h5', 'w')\n",
        "hf.create_dataset('dataset_1', data=x)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "jAZkdNfvMjRj",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Check above folder has the files**"
      ]
    },
    {
      "metadata": {
        "id": "AkkUpgBSM2YX",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "import statements"
      ]
    },
    {
      "metadata": {
        "id": "qbkcf052KLzb",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "31135a70-29cd-46b5-f59e-cb3b810ee2f7",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525867922657,
          "user_tz": -330,
          "elapsed": 9485,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "import numpy as np # linear algebra\n",
        "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
        "import tensorflow as tf\n",
        "import h5py\n",
        "\n",
        "import keras\n",
        "from keras.models import  Sequential\n",
        "from keras.models import Model\n",
        "from keras.layers import Dense, Dropout, Flatten, Activation, GlobalAveragePooling2D\n",
        "from keras.callbacks import EarlyStopping\n",
        "from keras.layers.convolutional import Conv2D, MaxPooling2D\n",
        "from keras import regularizers\n",
        "\n",
        "from keras.utils import np_utils\n",
        "\n",
        "# InceptionV3 model imports\n",
        "from keras.applications.inception_v3 import InceptionV3\n",
        "\n",
        "from keras.models import load_model"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "metadata": {
        "id": "IfDFUL6fNSnI",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Load data**"
      ]
    },
    {
      "metadata": {
        "id": "NOZ4cuDtNPl8",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "hf = h5py.File('data_1.h5', 'r')\n",
        "x = hf.get('dataset_1')\n",
        "y = hf.get('dataset_2')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "7x29Kln-OtKv",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "check if data is loaded"
      ]
    },
    {
      "metadata": {
        "id": "Ksy6IOSoN3Ic",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "6042e72a-e696-4b52-9be0-10e6be2520f8",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525793609342,
          "user_tz": -330,
          "elapsed": 1388,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "x[0].shape"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(551, 1117, 3)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "metadata": {
        "id": "AF5WoCMTXnOb",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "x = np.asarray(x)\n",
        "y = np.asarray(y)\n",
        "p = np.random.permutation(x.shape[0])\n",
        "x = x[p]\n",
        "y = y[p]\n",
        "X_train = x[0:200]\n",
        "y_train = y[0:200]\n",
        "X_val = x[200:226]\n",
        "y_val = y[200:226]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "AbLEgxVELuDw",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Define model(Optional)** (example for tranfer learning and small model created by me for comparison)"
      ]
    },
    {
      "metadata": {
        "id": "A4AzdmnlMZjP",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "inceptionv3 for transfer learning"
      ]
    },
    {
      "metadata": {
        "id": "V_1gzl-9MDLd",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "bb1f13e9-3287-4699-84ca-5c9ea22ad3c7",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525874422545,
          "user_tz": -330,
          "elapsed": 23698,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "# InceptionV3 model imports\n",
        "from keras.applications.inception_v3 import InceptionV3\n",
        "base_model = InceptionV3(include_top = False, weights='imagenet', input_shape= (551,1117,3))"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://github.com/fchollet/deep-learning-models/releases/download/v0.5/inception_v3_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
            "87916544/87910968 [==============================] - 5s 0us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "MBIGT02aMh8I",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "# add a global spatial average pooling layer\n",
        "x = base_model.output\n",
        "x = GlobalAveragePooling2D()(x)\n",
        "# let's add a fully-connected layer\n",
        "x = Dense(128, activation='relu')(x)\n",
        "# and a logistic layer -- let's say we have 10 classes\n",
        "predictions = Dense(10, activation='softmax')(x)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "woOP6CYIMiAP",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 11407
        },
        "outputId": "68070058-135f-4b75-ba11-1e2877fc4d53",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525874425122,
          "user_tz": -330,
          "elapsed": 1684,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "m = Model(inputs=base_model.input, outputs=predictions)\n",
        "\n",
        "for layer in base_model.layers:\n",
        "    layer.trainable = False\n",
        "\n",
        "m.compile(loss=\"categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"])  \n",
        "m.summary()"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "__________________________________________________________________________________________________\n",
            "Layer (type)                    Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            "input_1 (InputLayer)            (None, 551, 1117, 3) 0                                            \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_1 (Conv2D)               (None, 275, 558, 32) 864         input_1[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_1 (BatchNor (None, 275, 558, 32) 96          conv2d_1[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "activation_1 (Activation)       (None, 275, 558, 32) 0           batch_normalization_1[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_2 (Conv2D)               (None, 273, 556, 32) 9216        activation_1[0][0]               \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_2 (BatchNor (None, 273, 556, 32) 96          conv2d_2[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "activation_2 (Activation)       (None, 273, 556, 32) 0           batch_normalization_2[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_3 (Conv2D)               (None, 273, 556, 64) 18432       activation_2[0][0]               \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_3 (BatchNor (None, 273, 556, 64) 192         conv2d_3[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "activation_3 (Activation)       (None, 273, 556, 64) 0           batch_normalization_3[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2D)  (None, 136, 277, 64) 0           activation_3[0][0]               \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_4 (Conv2D)               (None, 136, 277, 80) 5120        max_pooling2d_1[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_4 (BatchNor (None, 136, 277, 80) 240         conv2d_4[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "activation_4 (Activation)       (None, 136, 277, 80) 0           batch_normalization_4[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_5 (Conv2D)               (None, 134, 275, 192 138240      activation_4[0][0]               \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_5 (BatchNor (None, 134, 275, 192 576         conv2d_5[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "activation_5 (Activation)       (None, 134, 275, 192 0           batch_normalization_5[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2D)  (None, 66, 137, 192) 0           activation_5[0][0]               \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_9 (Conv2D)               (None, 66, 137, 64)  12288       max_pooling2d_2[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_9 (BatchNor (None, 66, 137, 64)  192         conv2d_9[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "activation_9 (Activation)       (None, 66, 137, 64)  0           batch_normalization_9[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_7 (Conv2D)               (None, 66, 137, 48)  9216        max_pooling2d_2[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_10 (Conv2D)              (None, 66, 137, 96)  55296       activation_9[0][0]               \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_7 (BatchNor (None, 66, 137, 48)  144         conv2d_7[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_10 (BatchNo (None, 66, 137, 96)  288         conv2d_10[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_7 (Activation)       (None, 66, 137, 48)  0           batch_normalization_7[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "activation_10 (Activation)      (None, 66, 137, 96)  0           batch_normalization_10[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "average_pooling2d_1 (AveragePoo (None, 66, 137, 192) 0           max_pooling2d_2[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_6 (Conv2D)               (None, 66, 137, 64)  12288       max_pooling2d_2[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_8 (Conv2D)               (None, 66, 137, 64)  76800       activation_7[0][0]               \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_11 (Conv2D)              (None, 66, 137, 96)  82944       activation_10[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_12 (Conv2D)              (None, 66, 137, 32)  6144        average_pooling2d_1[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_6 (BatchNor (None, 66, 137, 64)  192         conv2d_6[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_8 (BatchNor (None, 66, 137, 64)  192         conv2d_8[0][0]                   \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_11 (BatchNo (None, 66, 137, 96)  288         conv2d_11[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_12 (BatchNo (None, 66, 137, 32)  96          conv2d_12[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_6 (Activation)       (None, 66, 137, 64)  0           batch_normalization_6[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "activation_8 (Activation)       (None, 66, 137, 64)  0           batch_normalization_8[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "activation_11 (Activation)      (None, 66, 137, 96)  0           batch_normalization_11[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_12 (Activation)      (None, 66, 137, 32)  0           batch_normalization_12[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "mixed0 (Concatenate)            (None, 66, 137, 256) 0           activation_6[0][0]               \n",
            "                                                                 activation_8[0][0]               \n",
            "                                                                 activation_11[0][0]              \n",
            "                                                                 activation_12[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_16 (Conv2D)              (None, 66, 137, 64)  16384       mixed0[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_16 (BatchNo (None, 66, 137, 64)  192         conv2d_16[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_16 (Activation)      (None, 66, 137, 64)  0           batch_normalization_16[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_14 (Conv2D)              (None, 66, 137, 48)  12288       mixed0[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_17 (Conv2D)              (None, 66, 137, 96)  55296       activation_16[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_14 (BatchNo (None, 66, 137, 48)  144         conv2d_14[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_17 (BatchNo (None, 66, 137, 96)  288         conv2d_17[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_14 (Activation)      (None, 66, 137, 48)  0           batch_normalization_14[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_17 (Activation)      (None, 66, 137, 96)  0           batch_normalization_17[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "average_pooling2d_2 (AveragePoo (None, 66, 137, 256) 0           mixed0[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_13 (Conv2D)              (None, 66, 137, 64)  16384       mixed0[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_15 (Conv2D)              (None, 66, 137, 64)  76800       activation_14[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_18 (Conv2D)              (None, 66, 137, 96)  82944       activation_17[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_19 (Conv2D)              (None, 66, 137, 64)  16384       average_pooling2d_2[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_13 (BatchNo (None, 66, 137, 64)  192         conv2d_13[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_15 (BatchNo (None, 66, 137, 64)  192         conv2d_15[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_18 (BatchNo (None, 66, 137, 96)  288         conv2d_18[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_19 (BatchNo (None, 66, 137, 64)  192         conv2d_19[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_13 (Activation)      (None, 66, 137, 64)  0           batch_normalization_13[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_15 (Activation)      (None, 66, 137, 64)  0           batch_normalization_15[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_18 (Activation)      (None, 66, 137, 96)  0           batch_normalization_18[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_19 (Activation)      (None, 66, 137, 64)  0           batch_normalization_19[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "mixed1 (Concatenate)            (None, 66, 137, 288) 0           activation_13[0][0]              \n",
            "                                                                 activation_15[0][0]              \n",
            "                                                                 activation_18[0][0]              \n",
            "                                                                 activation_19[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_23 (Conv2D)              (None, 66, 137, 64)  18432       mixed1[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_23 (BatchNo (None, 66, 137, 64)  192         conv2d_23[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_23 (Activation)      (None, 66, 137, 64)  0           batch_normalization_23[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_21 (Conv2D)              (None, 66, 137, 48)  13824       mixed1[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_24 (Conv2D)              (None, 66, 137, 96)  55296       activation_23[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_21 (BatchNo (None, 66, 137, 48)  144         conv2d_21[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_24 (BatchNo (None, 66, 137, 96)  288         conv2d_24[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_21 (Activation)      (None, 66, 137, 48)  0           batch_normalization_21[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_24 (Activation)      (None, 66, 137, 96)  0           batch_normalization_24[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "average_pooling2d_3 (AveragePoo (None, 66, 137, 288) 0           mixed1[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_20 (Conv2D)              (None, 66, 137, 64)  18432       mixed1[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_22 (Conv2D)              (None, 66, 137, 64)  76800       activation_21[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_25 (Conv2D)              (None, 66, 137, 96)  82944       activation_24[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_26 (Conv2D)              (None, 66, 137, 64)  18432       average_pooling2d_3[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_20 (BatchNo (None, 66, 137, 64)  192         conv2d_20[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_22 (BatchNo (None, 66, 137, 64)  192         conv2d_22[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_25 (BatchNo (None, 66, 137, 96)  288         conv2d_25[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_26 (BatchNo (None, 66, 137, 64)  192         conv2d_26[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_20 (Activation)      (None, 66, 137, 64)  0           batch_normalization_20[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_22 (Activation)      (None, 66, 137, 64)  0           batch_normalization_22[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_25 (Activation)      (None, 66, 137, 96)  0           batch_normalization_25[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_26 (Activation)      (None, 66, 137, 64)  0           batch_normalization_26[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "mixed2 (Concatenate)            (None, 66, 137, 288) 0           activation_20[0][0]              \n",
            "                                                                 activation_22[0][0]              \n",
            "                                                                 activation_25[0][0]              \n",
            "                                                                 activation_26[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_28 (Conv2D)              (None, 66, 137, 64)  18432       mixed2[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_28 (BatchNo (None, 66, 137, 64)  192         conv2d_28[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_28 (Activation)      (None, 66, 137, 64)  0           batch_normalization_28[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_29 (Conv2D)              (None, 66, 137, 96)  55296       activation_28[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_29 (BatchNo (None, 66, 137, 96)  288         conv2d_29[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_29 (Activation)      (None, 66, 137, 96)  0           batch_normalization_29[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_27 (Conv2D)              (None, 32, 68, 384)  995328      mixed2[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_30 (Conv2D)              (None, 32, 68, 96)   82944       activation_29[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_27 (BatchNo (None, 32, 68, 384)  1152        conv2d_27[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_30 (BatchNo (None, 32, 68, 96)   288         conv2d_30[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_27 (Activation)      (None, 32, 68, 384)  0           batch_normalization_27[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_30 (Activation)      (None, 32, 68, 96)   0           batch_normalization_30[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "max_pooling2d_3 (MaxPooling2D)  (None, 32, 68, 288)  0           mixed2[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "mixed3 (Concatenate)            (None, 32, 68, 768)  0           activation_27[0][0]              \n",
            "                                                                 activation_30[0][0]              \n",
            "                                                                 max_pooling2d_3[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_35 (Conv2D)              (None, 32, 68, 128)  98304       mixed3[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_35 (BatchNo (None, 32, 68, 128)  384         conv2d_35[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_35 (Activation)      (None, 32, 68, 128)  0           batch_normalization_35[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_36 (Conv2D)              (None, 32, 68, 128)  114688      activation_35[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_36 (BatchNo (None, 32, 68, 128)  384         conv2d_36[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_36 (Activation)      (None, 32, 68, 128)  0           batch_normalization_36[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_32 (Conv2D)              (None, 32, 68, 128)  98304       mixed3[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_37 (Conv2D)              (None, 32, 68, 128)  114688      activation_36[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_32 (BatchNo (None, 32, 68, 128)  384         conv2d_32[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_37 (BatchNo (None, 32, 68, 128)  384         conv2d_37[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_32 (Activation)      (None, 32, 68, 128)  0           batch_normalization_32[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_37 (Activation)      (None, 32, 68, 128)  0           batch_normalization_37[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_33 (Conv2D)              (None, 32, 68, 128)  114688      activation_32[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_38 (Conv2D)              (None, 32, 68, 128)  114688      activation_37[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_33 (BatchNo (None, 32, 68, 128)  384         conv2d_33[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_38 (BatchNo (None, 32, 68, 128)  384         conv2d_38[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_33 (Activation)      (None, 32, 68, 128)  0           batch_normalization_33[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_38 (Activation)      (None, 32, 68, 128)  0           batch_normalization_38[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "average_pooling2d_4 (AveragePoo (None, 32, 68, 768)  0           mixed3[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_31 (Conv2D)              (None, 32, 68, 192)  147456      mixed3[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_34 (Conv2D)              (None, 32, 68, 192)  172032      activation_33[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_39 (Conv2D)              (None, 32, 68, 192)  172032      activation_38[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_40 (Conv2D)              (None, 32, 68, 192)  147456      average_pooling2d_4[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_31 (BatchNo (None, 32, 68, 192)  576         conv2d_31[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_34 (BatchNo (None, 32, 68, 192)  576         conv2d_34[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_39 (BatchNo (None, 32, 68, 192)  576         conv2d_39[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_40 (BatchNo (None, 32, 68, 192)  576         conv2d_40[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_31 (Activation)      (None, 32, 68, 192)  0           batch_normalization_31[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_34 (Activation)      (None, 32, 68, 192)  0           batch_normalization_34[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_39 (Activation)      (None, 32, 68, 192)  0           batch_normalization_39[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_40 (Activation)      (None, 32, 68, 192)  0           batch_normalization_40[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "mixed4 (Concatenate)            (None, 32, 68, 768)  0           activation_31[0][0]              \n",
            "                                                                 activation_34[0][0]              \n",
            "                                                                 activation_39[0][0]              \n",
            "                                                                 activation_40[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_45 (Conv2D)              (None, 32, 68, 160)  122880      mixed4[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_45 (BatchNo (None, 32, 68, 160)  480         conv2d_45[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_45 (Activation)      (None, 32, 68, 160)  0           batch_normalization_45[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_46 (Conv2D)              (None, 32, 68, 160)  179200      activation_45[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_46 (BatchNo (None, 32, 68, 160)  480         conv2d_46[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_46 (Activation)      (None, 32, 68, 160)  0           batch_normalization_46[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_42 (Conv2D)              (None, 32, 68, 160)  122880      mixed4[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_47 (Conv2D)              (None, 32, 68, 160)  179200      activation_46[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_42 (BatchNo (None, 32, 68, 160)  480         conv2d_42[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_47 (BatchNo (None, 32, 68, 160)  480         conv2d_47[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_42 (Activation)      (None, 32, 68, 160)  0           batch_normalization_42[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_47 (Activation)      (None, 32, 68, 160)  0           batch_normalization_47[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_43 (Conv2D)              (None, 32, 68, 160)  179200      activation_42[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_48 (Conv2D)              (None, 32, 68, 160)  179200      activation_47[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_43 (BatchNo (None, 32, 68, 160)  480         conv2d_43[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_48 (BatchNo (None, 32, 68, 160)  480         conv2d_48[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_43 (Activation)      (None, 32, 68, 160)  0           batch_normalization_43[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_48 (Activation)      (None, 32, 68, 160)  0           batch_normalization_48[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "average_pooling2d_5 (AveragePoo (None, 32, 68, 768)  0           mixed4[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_41 (Conv2D)              (None, 32, 68, 192)  147456      mixed4[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_44 (Conv2D)              (None, 32, 68, 192)  215040      activation_43[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_49 (Conv2D)              (None, 32, 68, 192)  215040      activation_48[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_50 (Conv2D)              (None, 32, 68, 192)  147456      average_pooling2d_5[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_41 (BatchNo (None, 32, 68, 192)  576         conv2d_41[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_44 (BatchNo (None, 32, 68, 192)  576         conv2d_44[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_49 (BatchNo (None, 32, 68, 192)  576         conv2d_49[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_50 (BatchNo (None, 32, 68, 192)  576         conv2d_50[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_41 (Activation)      (None, 32, 68, 192)  0           batch_normalization_41[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_44 (Activation)      (None, 32, 68, 192)  0           batch_normalization_44[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_49 (Activation)      (None, 32, 68, 192)  0           batch_normalization_49[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_50 (Activation)      (None, 32, 68, 192)  0           batch_normalization_50[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "mixed5 (Concatenate)            (None, 32, 68, 768)  0           activation_41[0][0]              \n",
            "                                                                 activation_44[0][0]              \n",
            "                                                                 activation_49[0][0]              \n",
            "                                                                 activation_50[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_55 (Conv2D)              (None, 32, 68, 160)  122880      mixed5[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_55 (BatchNo (None, 32, 68, 160)  480         conv2d_55[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_55 (Activation)      (None, 32, 68, 160)  0           batch_normalization_55[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_56 (Conv2D)              (None, 32, 68, 160)  179200      activation_55[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_56 (BatchNo (None, 32, 68, 160)  480         conv2d_56[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_56 (Activation)      (None, 32, 68, 160)  0           batch_normalization_56[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_52 (Conv2D)              (None, 32, 68, 160)  122880      mixed5[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_57 (Conv2D)              (None, 32, 68, 160)  179200      activation_56[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_52 (BatchNo (None, 32, 68, 160)  480         conv2d_52[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_57 (BatchNo (None, 32, 68, 160)  480         conv2d_57[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_52 (Activation)      (None, 32, 68, 160)  0           batch_normalization_52[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_57 (Activation)      (None, 32, 68, 160)  0           batch_normalization_57[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_53 (Conv2D)              (None, 32, 68, 160)  179200      activation_52[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_58 (Conv2D)              (None, 32, 68, 160)  179200      activation_57[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_53 (BatchNo (None, 32, 68, 160)  480         conv2d_53[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_58 (BatchNo (None, 32, 68, 160)  480         conv2d_58[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_53 (Activation)      (None, 32, 68, 160)  0           batch_normalization_53[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_58 (Activation)      (None, 32, 68, 160)  0           batch_normalization_58[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "average_pooling2d_6 (AveragePoo (None, 32, 68, 768)  0           mixed5[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_51 (Conv2D)              (None, 32, 68, 192)  147456      mixed5[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_54 (Conv2D)              (None, 32, 68, 192)  215040      activation_53[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_59 (Conv2D)              (None, 32, 68, 192)  215040      activation_58[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_60 (Conv2D)              (None, 32, 68, 192)  147456      average_pooling2d_6[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_51 (BatchNo (None, 32, 68, 192)  576         conv2d_51[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_54 (BatchNo (None, 32, 68, 192)  576         conv2d_54[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_59 (BatchNo (None, 32, 68, 192)  576         conv2d_59[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_60 (BatchNo (None, 32, 68, 192)  576         conv2d_60[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_51 (Activation)      (None, 32, 68, 192)  0           batch_normalization_51[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_54 (Activation)      (None, 32, 68, 192)  0           batch_normalization_54[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_59 (Activation)      (None, 32, 68, 192)  0           batch_normalization_59[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_60 (Activation)      (None, 32, 68, 192)  0           batch_normalization_60[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "mixed6 (Concatenate)            (None, 32, 68, 768)  0           activation_51[0][0]              \n",
            "                                                                 activation_54[0][0]              \n",
            "                                                                 activation_59[0][0]              \n",
            "                                                                 activation_60[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_65 (Conv2D)              (None, 32, 68, 192)  147456      mixed6[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_65 (BatchNo (None, 32, 68, 192)  576         conv2d_65[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_65 (Activation)      (None, 32, 68, 192)  0           batch_normalization_65[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_66 (Conv2D)              (None, 32, 68, 192)  258048      activation_65[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_66 (BatchNo (None, 32, 68, 192)  576         conv2d_66[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_66 (Activation)      (None, 32, 68, 192)  0           batch_normalization_66[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_62 (Conv2D)              (None, 32, 68, 192)  147456      mixed6[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_67 (Conv2D)              (None, 32, 68, 192)  258048      activation_66[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_62 (BatchNo (None, 32, 68, 192)  576         conv2d_62[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_67 (BatchNo (None, 32, 68, 192)  576         conv2d_67[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_62 (Activation)      (None, 32, 68, 192)  0           batch_normalization_62[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_67 (Activation)      (None, 32, 68, 192)  0           batch_normalization_67[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_63 (Conv2D)              (None, 32, 68, 192)  258048      activation_62[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_68 (Conv2D)              (None, 32, 68, 192)  258048      activation_67[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_63 (BatchNo (None, 32, 68, 192)  576         conv2d_63[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_68 (BatchNo (None, 32, 68, 192)  576         conv2d_68[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_63 (Activation)      (None, 32, 68, 192)  0           batch_normalization_63[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_68 (Activation)      (None, 32, 68, 192)  0           batch_normalization_68[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "average_pooling2d_7 (AveragePoo (None, 32, 68, 768)  0           mixed6[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_61 (Conv2D)              (None, 32, 68, 192)  147456      mixed6[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_64 (Conv2D)              (None, 32, 68, 192)  258048      activation_63[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_69 (Conv2D)              (None, 32, 68, 192)  258048      activation_68[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_70 (Conv2D)              (None, 32, 68, 192)  147456      average_pooling2d_7[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_61 (BatchNo (None, 32, 68, 192)  576         conv2d_61[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_64 (BatchNo (None, 32, 68, 192)  576         conv2d_64[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_69 (BatchNo (None, 32, 68, 192)  576         conv2d_69[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_70 (BatchNo (None, 32, 68, 192)  576         conv2d_70[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_61 (Activation)      (None, 32, 68, 192)  0           batch_normalization_61[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_64 (Activation)      (None, 32, 68, 192)  0           batch_normalization_64[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_69 (Activation)      (None, 32, 68, 192)  0           batch_normalization_69[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_70 (Activation)      (None, 32, 68, 192)  0           batch_normalization_70[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "mixed7 (Concatenate)            (None, 32, 68, 768)  0           activation_61[0][0]              \n",
            "                                                                 activation_64[0][0]              \n",
            "                                                                 activation_69[0][0]              \n",
            "                                                                 activation_70[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_73 (Conv2D)              (None, 32, 68, 192)  147456      mixed7[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_73 (BatchNo (None, 32, 68, 192)  576         conv2d_73[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_73 (Activation)      (None, 32, 68, 192)  0           batch_normalization_73[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_74 (Conv2D)              (None, 32, 68, 192)  258048      activation_73[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_74 (BatchNo (None, 32, 68, 192)  576         conv2d_74[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_74 (Activation)      (None, 32, 68, 192)  0           batch_normalization_74[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_71 (Conv2D)              (None, 32, 68, 192)  147456      mixed7[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_75 (Conv2D)              (None, 32, 68, 192)  258048      activation_74[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_71 (BatchNo (None, 32, 68, 192)  576         conv2d_71[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_75 (BatchNo (None, 32, 68, 192)  576         conv2d_75[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_71 (Activation)      (None, 32, 68, 192)  0           batch_normalization_71[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_75 (Activation)      (None, 32, 68, 192)  0           batch_normalization_75[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_72 (Conv2D)              (None, 15, 33, 320)  552960      activation_71[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_76 (Conv2D)              (None, 15, 33, 192)  331776      activation_75[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_72 (BatchNo (None, 15, 33, 320)  960         conv2d_72[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_76 (BatchNo (None, 15, 33, 192)  576         conv2d_76[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_72 (Activation)      (None, 15, 33, 320)  0           batch_normalization_72[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_76 (Activation)      (None, 15, 33, 192)  0           batch_normalization_76[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "max_pooling2d_4 (MaxPooling2D)  (None, 15, 33, 768)  0           mixed7[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "mixed8 (Concatenate)            (None, 15, 33, 1280) 0           activation_72[0][0]              \n",
            "                                                                 activation_76[0][0]              \n",
            "                                                                 max_pooling2d_4[0][0]            \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_81 (Conv2D)              (None, 15, 33, 448)  573440      mixed8[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_81 (BatchNo (None, 15, 33, 448)  1344        conv2d_81[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_81 (Activation)      (None, 15, 33, 448)  0           batch_normalization_81[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_78 (Conv2D)              (None, 15, 33, 384)  491520      mixed8[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_82 (Conv2D)              (None, 15, 33, 384)  1548288     activation_81[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_78 (BatchNo (None, 15, 33, 384)  1152        conv2d_78[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_82 (BatchNo (None, 15, 33, 384)  1152        conv2d_82[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_78 (Activation)      (None, 15, 33, 384)  0           batch_normalization_78[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_82 (Activation)      (None, 15, 33, 384)  0           batch_normalization_82[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_79 (Conv2D)              (None, 15, 33, 384)  442368      activation_78[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_80 (Conv2D)              (None, 15, 33, 384)  442368      activation_78[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_83 (Conv2D)              (None, 15, 33, 384)  442368      activation_82[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_84 (Conv2D)              (None, 15, 33, 384)  442368      activation_82[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "average_pooling2d_8 (AveragePoo (None, 15, 33, 1280) 0           mixed8[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_77 (Conv2D)              (None, 15, 33, 320)  409600      mixed8[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_79 (BatchNo (None, 15, 33, 384)  1152        conv2d_79[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_80 (BatchNo (None, 15, 33, 384)  1152        conv2d_80[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_83 (BatchNo (None, 15, 33, 384)  1152        conv2d_83[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_84 (BatchNo (None, 15, 33, 384)  1152        conv2d_84[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_85 (Conv2D)              (None, 15, 33, 192)  245760      average_pooling2d_8[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_77 (BatchNo (None, 15, 33, 320)  960         conv2d_77[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_79 (Activation)      (None, 15, 33, 384)  0           batch_normalization_79[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_80 (Activation)      (None, 15, 33, 384)  0           batch_normalization_80[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_83 (Activation)      (None, 15, 33, 384)  0           batch_normalization_83[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_84 (Activation)      (None, 15, 33, 384)  0           batch_normalization_84[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_85 (BatchNo (None, 15, 33, 192)  576         conv2d_85[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_77 (Activation)      (None, 15, 33, 320)  0           batch_normalization_77[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "mixed9_0 (Concatenate)          (None, 15, 33, 768)  0           activation_79[0][0]              \n",
            "                                                                 activation_80[0][0]              "
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "__________________________________________________________________________________________________\n",
            "concatenate_1 (Concatenate)     (None, 15, 33, 768)  0           activation_83[0][0]              \n",
            "                                                                 activation_84[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "activation_85 (Activation)      (None, 15, 33, 192)  0           batch_normalization_85[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "mixed9 (Concatenate)            (None, 15, 33, 2048) 0           activation_77[0][0]              \n",
            "                                                                 mixed9_0[0][0]                   \n",
            "                                                                 concatenate_1[0][0]              \n",
            "                                                                 activation_85[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_90 (Conv2D)              (None, 15, 33, 448)  917504      mixed9[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_90 (BatchNo (None, 15, 33, 448)  1344        conv2d_90[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_90 (Activation)      (None, 15, 33, 448)  0           batch_normalization_90[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_87 (Conv2D)              (None, 15, 33, 384)  786432      mixed9[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_91 (Conv2D)              (None, 15, 33, 384)  1548288     activation_90[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_87 (BatchNo (None, 15, 33, 384)  1152        conv2d_87[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_91 (BatchNo (None, 15, 33, 384)  1152        conv2d_91[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_87 (Activation)      (None, 15, 33, 384)  0           batch_normalization_87[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_91 (Activation)      (None, 15, 33, 384)  0           batch_normalization_91[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_88 (Conv2D)              (None, 15, 33, 384)  442368      activation_87[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_89 (Conv2D)              (None, 15, 33, 384)  442368      activation_87[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_92 (Conv2D)              (None, 15, 33, 384)  442368      activation_91[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_93 (Conv2D)              (None, 15, 33, 384)  442368      activation_91[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "average_pooling2d_9 (AveragePoo (None, 15, 33, 2048) 0           mixed9[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_86 (Conv2D)              (None, 15, 33, 320)  655360      mixed9[0][0]                     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_88 (BatchNo (None, 15, 33, 384)  1152        conv2d_88[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_89 (BatchNo (None, 15, 33, 384)  1152        conv2d_89[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_92 (BatchNo (None, 15, 33, 384)  1152        conv2d_92[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_93 (BatchNo (None, 15, 33, 384)  1152        conv2d_93[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "conv2d_94 (Conv2D)              (None, 15, 33, 192)  393216      average_pooling2d_9[0][0]        \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_86 (BatchNo (None, 15, 33, 320)  960         conv2d_86[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_88 (Activation)      (None, 15, 33, 384)  0           batch_normalization_88[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_89 (Activation)      (None, 15, 33, 384)  0           batch_normalization_89[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_92 (Activation)      (None, 15, 33, 384)  0           batch_normalization_92[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "activation_93 (Activation)      (None, 15, 33, 384)  0           batch_normalization_93[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "batch_normalization_94 (BatchNo (None, 15, 33, 192)  576         conv2d_94[0][0]                  \n",
            "__________________________________________________________________________________________________\n",
            "activation_86 (Activation)      (None, 15, 33, 320)  0           batch_normalization_86[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "mixed9_1 (Concatenate)          (None, 15, 33, 768)  0           activation_88[0][0]              \n",
            "                                                                 activation_89[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "concatenate_2 (Concatenate)     (None, 15, 33, 768)  0           activation_92[0][0]              \n",
            "                                                                 activation_93[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "activation_94 (Activation)      (None, 15, 33, 192)  0           batch_normalization_94[0][0]     \n",
            "__________________________________________________________________________________________________\n",
            "mixed10 (Concatenate)           (None, 15, 33, 2048) 0           activation_86[0][0]              \n",
            "                                                                 mixed9_1[0][0]                   \n",
            "                                                                 concatenate_2[0][0]              \n",
            "                                                                 activation_94[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "global_average_pooling2d_1 (Glo (None, 2048)         0           mixed10[0][0]                    \n",
            "__________________________________________________________________________________________________\n",
            "dense_1 (Dense)                 (None, 128)          262272      global_average_pooling2d_1[0][0] \n",
            "__________________________________________________________________________________________________\n",
            "dense_2 (Dense)                 (None, 10)           1290        dense_1[0][0]                    \n",
            "==================================================================================================\n",
            "Total params: 22,066,346\n",
            "Trainable params: 263,562\n",
            "Non-trainable params: 21,802,784\n",
            "__________________________________________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "mqIxnjqGMv7n",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "my model for comparison"
      ]
    },
    {
      "metadata": {
        "id": "_PTL7UUxMh6U",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 612
        },
        "outputId": "32334c4c-5347-436b-a604-1a567b45ef04",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525874429733,
          "user_tz": -330,
          "elapsed": 1461,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        " conv_activation = 'relu'\n",
        "deep_activation = 'relu'\n",
        "\n",
        "input_shape = (551, 1117, 3)\n",
        "num_classes = 10\n",
        "\n",
        "model = Sequential()\n",
        "\n",
        "model.add(\n",
        "\tConv2D(\n",
        "\t\t16,\n",
        "\t\tkernel_size=(8, 8),\n",
        "\t\tstrides=(1, 1),\n",
        "\t\tactivation=conv_activation,\n",
        "\t\tinput_shape=input_shape,\n",
        "\t\tdata_format='channels_last'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(16, (16, 16), activation=conv_activation))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(16, (16, 16), activation=conv_activation))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Conv2D(16, (16, 16), activation=conv_activation))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "model.add(Flatten())\n",
        "\n",
        "# model.add(Dense(100, activation=deep_activation))\n",
        "model.add(Dense(100, activation=deep_activation))\n",
        "model.add(Dense(50, activation=deep_activation))\n",
        "model.add(Dense(num_classes, activation='softmax'))\n",
        "\n",
        "model.compile(loss=\"categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"])  \n",
        "\n",
        "model.summary()\n"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_95 (Conv2D)           (None, 544, 1110, 16)     3088      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_5 (MaxPooling2 (None, 272, 555, 16)      0         \n",
            "_________________________________________________________________\n",
            "conv2d_96 (Conv2D)           (None, 257, 540, 16)      65552     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_6 (MaxPooling2 (None, 128, 270, 16)      0         \n",
            "_________________________________________________________________\n",
            "conv2d_97 (Conv2D)           (None, 113, 255, 16)      65552     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_7 (MaxPooling2 (None, 56, 127, 16)       0         \n",
            "_________________________________________________________________\n",
            "max_pooling2d_8 (MaxPooling2 (None, 28, 63, 16)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_98 (Conv2D)           (None, 13, 48, 16)        65552     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_9 (MaxPooling2 (None, 6, 24, 16)         0         \n",
            "_________________________________________________________________\n",
            "max_pooling2d_10 (MaxPooling (None, 3, 12, 16)         0         \n",
            "_________________________________________________________________\n",
            "flatten_1 (Flatten)          (None, 576)               0         \n",
            "_________________________________________________________________\n",
            "dense_3 (Dense)              (None, 100)               57700     \n",
            "_________________________________________________________________\n",
            "dense_4 (Dense)              (None, 50)                5050      \n",
            "_________________________________________________________________\n",
            "dense_5 (Dense)              (None, 10)                510       \n",
            "=================================================================\n",
            "Total params: 263,004\n",
            "Trainable params: 263,004\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "IHagfjJFOx6w",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**load model(provide path accordingly) **"
      ]
    },
    {
      "metadata": {
        "id": "ti6S8T_2WLuC",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "go to runtime and change the runtime type to GPU"
      ]
    },
    {
      "metadata": {
        "id": "FAhR1n_lN4PD",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "model = load_model('tranfer_20ep_l2+dropout.h5')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "0ScklxF1Zh8D",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**train model(optinal)**"
      ]
    },
    {
      "metadata": {
        "id": "86Qta1dpRBDk",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 714
        },
        "outputId": "6ee63d31-6a4d-4735-c93a-b88cdbd7af9e",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525794418618,
          "user_tz": -330,
          "elapsed": 390985,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "history = model.fit(X_train, y_train, validation_data=(X_val, y_val), epochs= 20, batch_size= 32, verbose=2)"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 200 samples, validate on 26 samples\n",
            "Epoch 1/20\n",
            " - 36s - loss: 1.3924 - acc: 0.5450 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 2/20\n",
            " - 19s - loss: 1.3239 - acc: 0.5950 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 3/20\n",
            " - 19s - loss: 1.4369 - acc: 0.4900 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 4/20\n",
            " - 19s - loss: 1.3536 - acc: 0.5550 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 5/20\n",
            " - 19s - loss: 1.3495 - acc: 0.5250 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 6/20\n",
            " - 19s - loss: 1.3491 - acc: 0.5850 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 7/20\n",
            " - 19s - loss: 1.3246 - acc: 0.5800 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 8/20\n",
            " - 19s - loss: 1.3612 - acc: 0.5100 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 9/20\n",
            " - 19s - loss: 1.3102 - acc: 0.5750 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 10/20\n",
            " - 19s - loss: 1.3434 - acc: 0.5200 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 11/20\n",
            " - 19s - loss: 1.2986 - acc: 0.5850 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 12/20\n",
            " - 19s - loss: 1.2407 - acc: 0.5850 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 13/20\n",
            " - 19s - loss: 1.2422 - acc: 0.6250 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 14/20\n",
            " - 19s - loss: 1.2179 - acc: 0.6150 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 15/20\n",
            " - 19s - loss: 1.2225 - acc: 0.6150 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 16/20\n",
            " - 19s - loss: 1.2373 - acc: 0.6400 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 17/20\n",
            " - 19s - loss: 1.3005 - acc: 0.5400 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 18/20\n",
            " - 19s - loss: 1.1873 - acc: 0.6350 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 19/20\n",
            " - 19s - loss: 1.1546 - acc: 0.6650 - val_loss: 15.7582 - val_acc: 0.0385\n",
            "Epoch 20/20\n",
            " - 19s - loss: 1.2039 - acc: 0.6200 - val_loss: 15.7582 - val_acc: 0.0385\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "WclzRdWqcEW5",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "model.save('tranfer_20ep_l2+dropout.h5')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "B8fTNkO_WcPl",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 376
        },
        "outputId": "31e9a7ff-5aa8-4f47-9516-ed81bcb7e743",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525794464814,
          "user_tz": -330,
          "elapsed": 2635,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "# summarize history for accuracy\n",
        "plt.plot(history.history['acc'])\n",
        "plt.plot(history.history['val_acc'])\n",
        "plt.title('model accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'validation'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nREQMRslbRETEYL4C2SFFO7reM7gAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f07d17425c0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "w5SWYBUOadLB",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 376
        },
        "outputId": "a3fcf3f6-f015-4efb-ed5b-8b1954f04668",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525794478693,
          "user_tz": -330,
          "elapsed": 1886,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "# summarize history for loss\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'validation'], loc='upper left')\n",
        "plt.show()"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe0AAAFnCAYAAACLnxFFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzt3Xl8VPW9//H3mZlMNkISIAmuqNSl\nCgpYUEAiIBik6nUrKAJS9eICgohlsVK5vxRtBCkaV2jRXpYrilwvdSGKoiAiV8CqSCki9QoIIYYk\nhOwzc35/JBkSSEJCM5l8D6/ngzxm5nvOnO/3wyzvc86cOWPZtm0LAAC0eq5wDwAAADQOoQ0AgCEI\nbQAADEFoAwBgCEIbAABDENoAABiC0AZOUr/97W+VmZnZ4DwrVqzQmDFjGt0OILQIbQAADEFoAwbY\ns2ePrrjiCi1YsEBpaWlKS0vT3/72N40dO1b9+vXT9OnTg/O+++67uvbaazVkyBCNHj1aP/zwgyQp\nLy9Pd955pwYOHKixY8eqsLAweJ+dO3dq5MiRSktL03XXXaevv/660WPLz8/XxIkTlZaWpqFDh2r+\n/PnBaX/84x+D4x09erSys7MbbAfQME+4BwCgcfLy8pSUlKSsrCxNmDBBkyZN0htvvCHLspSamqr7\n7rtPHo9HM2bM0BtvvKFOnTpp4cKF+t3vfqdXXnlFCxYsUGJiohYuXKg9e/bo+uuv17nnnqtAIKBx\n48bp7rvv1q9+9Stt3rxZ999/v9asWdOocc2dO1fx8fHKyspSfn6+brzxRvXo0UPx8fFatWqV3nrr\nLUVERGjRokXasGGDLrroojrbb7jhhhD/DwLmY0sbMITP59OQIUMkSeedd566du2qdu3aKTExUUlJ\nSTpw4IDWr1+vyy67TJ06dZIk/epXv9LGjRvl8/m0adMmXXPNNZKk008/Xb169ZIk7dq1S7m5ubrl\nllskSZdeeqnatWunL774olHj+vjjjzVixAhJUkJCggYPHqz169erbdu2OnjwoP7617+qoKBAo0aN\n0g033FBvO4DjI7QBQ7jdbkVFRUmSXC6XYmJiak3z+/3Ky8tT27Ztg+1xcXGybVt5eXkqKChQXFxc\ncFr1fIcOHVJpaamuueYaDRkyREOGDFFubq7y8/MbNa6DBw/W6rNt27bKzc1VSkqKMjMztWrVKvXv\n319jx47Vvn376m0HcHyENuAg7du3rxW2BQUFcrlcSkxMVNu2bWt9jn3w4EFJUnJysmJjY7Vq1arg\n3yeffKLBgwc3qs8OHTrU6jM/P18dOnSQJF1++eWaP3++1q9fr1NOOUVz5sxpsB1AwwhtwEH69u2r\nTZs2affu3ZKkV199VX379pXH41G3bt20evVqSdIPP/ygzZs3S5JOO+00dezYUatWrZJUGeYPPfSQ\niouLG9Vn//79tWzZsuB933//ffXv31+ffPKJ/uM//kOBQEAxMTG64IILZFlWve0Ajo8D0QAH6dix\no37/+9/r/vvvV0VFhU4//XSlp6dLku655x5NmjRJAwcOVOfOnXX11VdLkizL0ty5czVz5kzNmzdP\nLpdLv/71r2vtfm/Igw8+qJkzZ2rIkCFyuVwaO3asLr74YpWVlentt99WWlqavF6v2rVrp8cff1zJ\nycl1tgM4Povf0wYAwAzsHgcAwBCENgAAhiC0AQAwBKENAIAhCG0AAAzRqr/ylZNTePyZmigxMUZ5\neY37/qkpnFiT5My6qMkcTqzLiTVJzqsrKSmu3mkn3Za2x+MO9xCanRNrkpxZFzWZw4l1ObEmybl1\n1eWkC20AAExFaAMAYAhCGwAAQ4Q0tHfs2KFBgwZp8eLFkqSKigpNnjxZt9xyi+644w4VFBSEsnsA\nABwlZKFdXFys9PR09e7dO9j22muvKTExUcuXL9fQoUO1adOmUHUPAIDjhCy0vV6vFixYoOTk5GDb\nmjVrdP3110uShg8frquuuipU3QMA4DghC22Px6OoqKhabXv37tXatWs1atQoTZo0Sfn5+aHqHgAA\nx2nRk6vYtq2zzz5b48eP1/PPP6+XXnpJU6dOrXf+xMSYkHz/rqEvrjdWVlaW0tLSjjvfrFmzNHr0\naJ1xxhn/cp8NaY6aWiMn1kVN5nBiXU6sSXJuXUdr0dDu0KGDevbsKUm64oorlJmZ2eD8oTjDTVJS\n3L98prV9+37UihVvqkePPsedd+zYCZJCc3a3as1RU2vkxLqoyRxOrMuJNUnOq6uhFZAWDe3U1FSt\nW7dON998s7755hudffbZLdl9s5k7N0N///s36tevp66++hrt2/ej5s17Xk888f+Uk3NAJSUluvPO\nserbt5/Gjx+rhx6aojVrPlBR0WH98MP/ae/ePZowYbJ69+4b7lIAAAaxbNu2Q7HgrVu3KiMjQ3v3\n7pXH41FKSormzJmjWbNmKScnRzExMcrIyFCHDh3qXcbx1pxe+3CnPt9+oNFjKvYVq9xf0eA80Um5\nij9nd4Pz5H2Xo72ffqfYlLYqzinURbdfpvLDpcrdnq1TftFJJbmHtXXxRvWceJW+ePFjnXtDN+V8\ntVdF2YfUZdTlyv3Hfv342S51veP4W+qN4XZZ8gdC8jCGlRProiZzOLEuJ9Ykhb+u7sldddPPrm22\n5YVlS7tLly5atGjRMe3PPPNMqLoMi7gzEiVJnmivCvcc1I8b/ynLknzF5cfMG39We0lSZHy0fCW+\nFh0nAMB8rfpXvo5n2MCfadjAnzXpPs3x2ceWqE1a8Y/XdPYZnZWQkKCb+wzXu+++pYiYYv1uUboO\nHTqku+8epfQ+0zV+6T/1QLe7tSbvg+C8u3bt1NwPn1R6n+n/0jias6bWyIl1UZM5nFiXE2uSnFtX\nXTiN6QlwuVzy+/212vLz83XKKafK5XLp448/VEVFw7vhAQBoKkL7BHTqdLb+8Y/tKio6HGzr33+g\nPv10nSZOvE/R0dFKTk7Wyy8vCOMoAQBOE7ID0ZpDKHZ3OHE3ihNrkpxZFzWZw4l1ObEmyXl1NXQg\nGlvaAAAYgtAGAMAQhDYAAIYgtAEAMAShDQCAIQhtAAAMQWiHyC23XKfi4mItWvSKtm79qta04uJi\n3XLLdQ3e/6OPPpAkvfPOX/Xxx2tCNk4AgDmMPo2pCUaNGtPk++zb96NWr85S//5XaejQhsMdAHDy\nILSb6M47b9fjjz+ljh07av/+fZo+fbKSkpJVUlKi0tJSTZr0G114YZfg/LNmzVT//lepW7fu+u1v\np6i8vFwXX9wtOP29997V8uXL5Ha7dNZZnTV16m+DP/358ssLFAgEKs9ZfvNwPf/80/r66y/l8/l1\n883DNGTILzV+/Fj175+qdevWKz8/XxkZf1THjh3D8V8DAAgxo0N7xc639MWBr5t0n+P9hNvxfmIt\nNXWA1q9fq5tvHqZ16z5WauoAde58rlJT+2vz5s+1ZMlfNGvW7GPul5X1rs45p7MmTJisDz54T6tX\nZ0mSSkpK9NRTmYqLi9O4cf+u777bqdtuG6UVK17Tr3/97/rzn1+SJP3tb1u0a9d3euGFhSopKdEd\nd9yq1NT+kqQ2bdro6adf0AsvZGrt2g81bNiIJv2fAADMYHRoh0Nq6gA9++w83XzzMH3yyccaP36S\nXn11kf7rvxapoqJCUVFRdd7v++93qVu3SyVJ3btfGmxv27atpk+fLEn6v//7pwoK8uu8//bt29St\nWw9JUnR0tM466xzt3l35u9+/+MUvJEnJyckqKChonkIBAK2O0aF908+ubfIPj/+r56g955zOys3N\nUXb2fhUWFmrduo/UoUOyZsxI1/bt2/Tss/PqvJ9tSy6XJUkKVG3pV1RUaO7cJ/XKK0vVvn0HTZny\nYL39WpalmmeJ9/kqgstzu901+mm1p5IHAPyLOHr8BPTufYXmz39e/fpdqYKCfJ122umSpI8/XiOf\nz1fnfc48s5O2b/+7JGnLlk2SpOLiIrndbrVv30HZ2fu1ffvf5fP56vzpzwsuuEhffLG56n7F2rt3\nj04//cxQlQgAaIUI7RNw5ZUDgkd3DxnySy1btkSTJo3TRRd1UW5urt5+e+Ux9xky5Jf65puvNXHi\nfdq9+/9kWZbi4xPUs+dluvvu0Xr55QUaMWKUnnlmbvCnP5955qng/S+5pJvOP/8CjRv375o0aZzu\nvXe8oqOjW7JsAECY8dOcDuDEmiRn1kVN5nBiXU6sSXJeXfw0JwAADkBoAwBgCEIbAABDENoAABiC\n0AYAwBCENgAAhiC0AQAwBKENAIAhCG0AAAxBaAMAYIiQhvaOHTs0aNAgLV68uFb7unXrdP7554ey\nawAAHCdkoV1cXKz09HT17t27VntZWZnmz5+vpKSkUHUNAIAjhSy0vV6vFixYoOTk5FrtL774okaM\nGCGv1xuqrgEAcCRPyBbs8cjjqb34f/7zn9q+fbsmTpyo2bNnH3cZiYkx8njczT62hn5BxVROrEly\nZl3UZA4n1uXEmiTn1nW0kIV2XZ544gk9+uijjZ4/L6+42cfgtJ9wk5xZk+TMuqjJHE6sy4k1Sc6r\nq1X8NGd2drZ27dqlhx9+WMOGDdOBAwc0cuTIluoeAADjtdiWdkpKilavXh28PXDgwGOOKgcAAPUL\nWWhv3bpVGRkZ2rt3rzwej7KyspSZmamEhIRQdQkAgKOFLLS7dOmiRYsW1Tv9ww8/DFXXAAA4EmdE\nAwDAEIQ2AACGILQBADAEoQ0AgCEIbQAADEFoAwBgCEIbAABDENoAABiC0AYAwBCENgAAhiC0AQAw\nBKENAIAhCG0AAAxBaAMAYAhCGwAAQxDaAAAYgtAGAMAQhDYAAIYgtAEAMAShDQCAIQhtAAAMQWgD\nAGAIQhsAAEMQ2gAAGILQBgDAEIQ2AACGILQBADAEoQ0AgCFCGto7duzQoEGDtHjxYknSvn37NGbM\nGI0cOVJjxoxRTk5OKLsHAMBRQhbaxcXFSk9PV+/evYNt8+bN07Bhw7R48WINHjxYL7/8cqi6BwDA\ncUIW2l6vVwsWLFBycnKw7bHHHlNaWpokKTExUfn5+aHqHgAAx/GEbMEejzye2ouPiYmRJPn9fi1d\nulTjxo1rcBmJiTHyeNzNPrakpLhmX2a4ObEmyZl1UZM5nFiXE2uSnFvX0UIW2vXx+/2aMmWKLr/8\n8lq7zuuSl1fc7P0nJcUpJ6ew2ZcbTk6sSXJmXdRkDifW5cSaJOfV1dAKSIsfPT59+nR16tRJ48eP\nb+muAQAwWouG9sqVKxUREaEJEya0ZLcAADhCyHaPb926VRkZGdq7d688Ho+ysrKUm5uryMhIjRo1\nSpLUuXNnzZw5M1RDAADAUUIW2l26dNGiRYtCtXgAAE46nBENAABDENoAABiC0AYAwBCENgAAhiC0\nAQAwBKENAIAhCG0AAAxBaAMAYAhCGwAAQxDaAAAYgtAGAMAQhDYAAIYgtAEAMAShDQCAIQhtAAAM\nQWgDAGAIQhsAAEMQ2gAAGILQBgDAEIQ2AACGILQBADAEoQ0AgCEIbQAADEFoAwBgCEIbAABDENoA\nABiC0AYAwBCENgAAhghpaO/YsUODBg3S4sWLJUn79u3TqFGjNGLECE2cOFHl5eWh7B4AAEcJWWgX\nFxcrPT1dvXv3DrY988wzGjFihJYuXapOnTpp+fLloeoeAADHCVloe71eLViwQMnJycG2jRs36qqr\nrpIkDRgwQBs2bAhV9wAAOI4nZAv2eOTx1F58SUmJvF6vJKl9+/bKyckJVfcAADhOyEL7eGzbPu48\niYkx8njczd53UlJcsy8z3JxYk+TMuqjJHE6sy4k1Sc6t62gtGtoxMTEqLS1VVFSUsrOza+06r0te\nXnGzjyEpKU45OYXNvtxwcmJNkjProiZzOLEuJ9YkOa+uhlZAWvQrX3369FFWVpYk6b333lO/fv1a\nsnsAAIwWsi3trVu3KiMjQ3v37pXH41FWVpbmzJmjadOmadmyZTr11FN1ww03hKp7AAAcJ2Sh3aVL\nFy1atOiY9pdffjlUXQIA4GicEQ0AAEMQ2gAAGILQBgDAEIQ2AACGILQBADAEoQ0AgCEIbQAADEFo\nAwBgCEIbAABDENoAABiC0AYAwBCENgAAhiC0AQAwBKENAIAhCG0AAAxBaAMAYIgmh3Z5ebn27dsX\nirEAAIAGeBoz00svvaSYmBjdcsstuvnmmxUbG6u+ffvqwQcfDPX4AABAlUZtaa9Zs0YjR47UqlWr\nNGDAAL3++uvasmVLqMcGAABqaFRoezweWZaltWvXatCgQZKkQCAQ0oEBAIDaGrV7PC4uTmPHjtX+\n/fvVvXt3rVmzRpZlhXpsAACghkaF9lNPPaVPP/1UPXr0kCRFRkYqIyMjpAMDAAC1NWr3+MGDB5WY\nmKh27drptdde01tvvaWSkpJQjw0AANTQqNCePn26IiIitG3bNr3++utKS0vT73//+1CPDQAA1NCo\n0LYsSxdffLHef/993X777bryyitl23aoxwYAAGpoVGgXFxfrq6++UlZWllJTU1VeXq5Dhw6FemwA\nAKCGRoX2nXfeqRkzZmj48OFq166dMjMzde2114Z6bAAAoIZGHT0+dOhQDR06VPn5+SooKNBDDz3E\nV74AAGhhjQrtzZs3a+rUqSoqKlIgEFBiYqJmz56trl27NqmzoqIiTZ06VQUFBaqoqNC4cePUr1+/\nExo4AAAnm0aF9ty5c/X888/rvPPOkyRt27ZNs2bN0pIlS5rU2X//93/r7LPP1uTJk5Wdna077rhD\nq1atavqoAQA4CTXqM22XyxUMbEm68MIL5Xa7m9xZYmKi8vPzJUmHDh1SYmJik5cBAMDJqlFb2i6X\nS1lZWerbt68kae3atScU2r/85S+1YsUKDR48WIcOHdJLL73U5GUAAHCysuxGfOH6+++/V3p6ur7+\n+mtZlqVLLrlEM2bM0BlnnNGkzv7nf/5HmzZtUnp6urZv365HHnlEK1asqHd+n88vj6fpKwcAADhR\ng1vaI0aMCB4lbtu2fvazn0mSDh8+rGnTpjX5M+0tW7boiiuukCRdcMEFOnDggPx+f71b7Xl5xU1a\nfmMkJcUpJ6ew2ZcbTk6sSXJmXdRkDifW5cSaJOfVlZQUV++0BkP7wQcfbNaBdOrUSV9++aXS0tK0\nd+9excbGntBudgAATkYNhnavXr2atbPhw4frkUce0ciRI+Xz+TRz5sxmXT4AAE7WqAPRmktsbKye\nfvrpluwSAADHaNRXvgAAQPgR2gAAGILQBgDAEIQ2AACGILQBADAEoQ0AgCEIbQAADEFoAwBgCEIb\nAABDENoAABiC0AYAwBCENgAAhiC0AQAwBKENAIAhCG0AAAxBaAMAYAhCGwAAQxDaAAAYgtAGAMAQ\nhDYAAIYgtAEAMAShDQCAIQhtAAAMQWgDAGAIQhsAAEMQ2gAAGILQBgDAEIQ2AACGaPHQXrlypa6/\n/nrddNNN+uijj1q6ewAAjNWioZ2Xl6fnnntOS5cu1YsvvqgPPvigJbsHAMBonpbsbMOGDerdu7fa\ntGmjNm3aKD09vSW7BwDAaC26pb1nzx6Vlpbq3nvv1YgRI7Rhw4aW7B4AAKNZtm3bLdXZ/PnztWXL\nFj377LP68ccfNXr0aK1Zs0aWZdU5v8/nl8fjbqnhAQDQqrXo7vH27dure/fu8ng8OvPMMxUbG6uD\nBw+qffv2dc6fl1fc7GNISopTTk5hsy83nJxYk+TMuqjJHE6sy4k1Sc6rKykprt5pLbp7/IorrtBn\nn32mQCCgvLw8FRcXKzExsSWHAACAsVp0SzslJUVpaWkaNmyYJOnRRx+Vy8VXxQEAaIwWDW1JuvXW\nW3Xrrbe2dLcAABiPzVwAAAxBaAMAYAhCGwAAQxDaAAAYgtAGAMAQhDYAAIYgtAEAMAShDQCAIQht\nAAAMQWgDAGAIQhsAAEMQ2gAAGILQBgDAEIQ2AACGILQBADAEoQ0AgCEIbQAADEFoAwBgCEIbAABD\nENoAABiC0AYAwBCENgAAhiC0AQAwBKENAIAhCG0AAAxBaAMAYAhCGwAAQxDaAAAYgtAGAMAQYQnt\n0tJSDRo0SCtWrAhH9wAAGCksof3CCy8oPj4+HF0DAGCsFg/t7777Tjt37lT//v1bumsAAIxm2bZt\nt2SHY8eO1YwZM/Tmm2/qtNNO00033VTvvD6fXx6PuwVHBwBA6+Vpyc7efPNNdevWTWeccUaj5s/L\nK272MSQlxSknp7DZlxtOTqxJcmZd1GQOJ9blxJok59WVlBRX77QWDe2PPvpIu3fv1kcffaT9+/fL\n6/WqY8eO6tOnT0sOAwAAI7VoaM+bNy94PTMzU6eddhqBDQBAI/E9bQAADNGiW9o1PfDAA+HqGgAA\nI7GlDQCAIQhtAAAMQWgDAGAIQhsAAEMQ2gAAGILQBgDAEIQ2AACGILQBADAEoQ0AgCEIbQAADEFo\nAwBgCEIbAABDENoAABiC0AYAwBCENgAAhiC0AQAwBKENAIAhCG0AAAxBaAMAYAhCGwAAQxDaAAAY\ngtAGAMAQhDYAAIYgtAEAMAShDQCAIQhtAAAMQWgDAGAIQhsAAEN4WrrDJ598Ups3b5bP59M999yj\nq6++uqWHAACAkVo0tD/77DN9++23WrZsmfLy8nTjjTcS2gAANFKLhnbPnj118cUXS5Latm2rkpIS\n+f1+ud3ulhwGAABGsmzbtsPR8bJly7Rp0ybNnj273nl8Pr88HgIdAAApDJ9pS9Lq1au1fPlyLVy4\nsMH58vKKm73vpKQ45eQUNvtyw8mJNUnOrIuazOHEupxYk+S8upKS4uqd1uKhvW7dOr344ov605/+\npLi4+gcGAABqa9HQLiws1JNPPqlXXnlFCQkJLdk1AADGa9HQfuedd5SXl6cHH3ww2JaRkaFTTz21\nJYcBAICRWjS0hw8fruHDh7dklwAAOAZnRAMAwBCENgAAhiC0AQAwBKENAIAhwnJylXCp8PlVXFoh\n27ZlWVbYxlFW7ld+UZnyC8tUUFSu/MPlKjhcpvzD5co/XNlWVFKhKK9bbaIjFBsdodioiKrrHrWJ\nrrpe1Wa73Sot8ynK6w5rXQCA0DppQrus3K+Hn1+volKfLEuKifQouuqv1vWomm1uxURFKDrSHWyr\nnjfC46oVkLZtq6TMr4KiGuFbI4QLDpcpryqcS8v9DY41NqoymEvK/fqpoFT+QOPONOt2WYqtCvQ2\nUZ7KsI+OUJuoI2HvcbtUPWxLlqr+VV1aR6ZVXamu0KqayTpqfgWvN07j1ykstTtYoqLDpXK7XHK7\nLbldljzuyuueqjaP21XVbsntcsnlavpKSyBgq8IXUIU/UHnp86vCF1C5L1Cr3ecLqLxqWs32QMBW\nZIRbkV535WUD1/3+QJPH11S2bctfoyZfjUvLsuSNcMkb4Vakx62ICJdcrOgBxjhpQtsb4VK/S07V\nT4fKlH+oVCVlPhWX+ZSTX3LcEK2Lx20Fg962bRUcLle5r+E35LiYCCUlRCu+jVcJsZGVl20ildDG\nq/g2kUqI9Sq+jVcRNc63btu2Ssv9Kiqp0OHSChWV+HS4pEJFpRU6XFL557el3PySyvaSChUcLtO+\nn4oUlpPKh5llSW6XqyrEjw35QMCuEc6Vf41dKWouHrd1bLAfddsb4VLArtw75PNXBrDPH6gziI+0\n28HrTRHhccnrqQzyyjCvvu6S1+OuFfLV12tOS2rXRuVl5YryehTlrayj+rr3qJVbmClg2yor96uk\nzKeScr9Ky3wqKfeptMyvsgq/4mK8wfexuJgIVgRDKGw/GNIYoTiXbF3nqA0EbJWU+1RSWhnk1YFe\nUuZTSVnlLvWSMn8d03wqLvVJluoN4YS4SMXHetU21iuPOzSHENRZk22ruNRXI+wrA97nP/Jw27Zd\nGey2qi4rb9d8RjRqnsauHjThmRawbUVHe1VwqHJPg89fGa7B635b/kDlZfW0I5c1pgVs+WtMd1mW\nIjyu2n9ulyI87mB4eWq1H/nzVs1T3V49n0tSmS+g8gq/Sssr38TKqi9rXC+vCCggqbCo/Jh5mhK0\nliqD1uOuGkONywhP5YpK9fTq8VbPK9tWWUXlHoPyisoxB6/XuCwrDyjQDG8NLsuqCvHqP0/t65FH\ntUe4FRVZeb36jb/W88uudVF5vdYw7WPaak52Va3U1Vyhc7ssud1Hrei5LKUkt1V+XpHcbksuy2oV\nKx8B21YgYAf3pgQCVW1V7YFAjeu2jrptq02baO07cCj4/lVS5ldpeeVlZQjXDGV/cL6mbNi4LEvx\nbbyKj639fli9sZIQ51V8bKTaxkbI7Wqe90TOPX6ScbksxUZVfkbsFC7LCn72nRLuwZwgp70Qpfpr\nCgTsIyFfFehut0sRbksRHrc87iNh7Ha1TID4/IEaYV4Z6GW1wr7yMiIyQj8dLFJp1ZZXaXllEJRW\nrcSUVgVDYXGFcvJL5WuBjwiamyVVBrzbJY/rSNBXX6rWyq0dXGmwqxprrujaNVd87WOnVd/P1pFg\ntqvDuoXqdbuq9yS6lZwQrahIj6K9lR8TRlW1R3uPfFRYWFx+zMeCe38q0vf763/9WpaCW+gJbSKP\nCfnoSE/VCql17ApocMW0eV8LPn9ApeWVz/faK92B2m1V7eUVAcmS0nqdqfhYb7ONoyGENtAKuFxH\nPm5pLarfHGOO8zbR1JWr6jfG0nJf1aX/mMAvq/Cr5qcWNd+W63qPrvnGbR1z5cjxFwG75p6XY/fI\n1Gxze9wqLi2v3KvjD1TNY8sXOLKnx+e3Ve7z1TjOo8bxIMFjPSy5rKrplmTJVXXdqpr/SA01l+Oy\nLFmuyq18l0tyW5ZcVStsrqoVh8o9AJUh63Id2SNQefvIMqrv2zYuSrbfX/lcq9rTUR2+R667a31E\nd6Js21Zxma/GgbbVgV5eeexPYZnyi8q1/2Cxfsg+fML9eNyWvBHu4F6SiBp7mjweq9aeKMuygqFb\nXnNvWFUon+hHZeeeHq/u5yadcA1N0XreIQCcFDxul9pEu9QmunXv2TqZ9vSEgmUd2YN5WofYeuer\nPm4nGOpFZcovrPwI6ejjOKpv+/xHjk2p3HNjqaSsouoA0sqPMqun1RXElqVax5O0ifYq0uuq1eYN\nHmviqn07+OeSt+obPimJMSECtVtJAAAKpUlEQVT8n6yN0AYAhI1lHdnLdEr7+sO9IQ2tjASqjmep\n8Adk21JkhKvqWzThP0bhRBDaAADHcrkseV2VW8pOwBnRAAAwBKENAIAhCG0AAAxBaAMAYAhCGwAA\nQxDaAAAYgtAGAMAQhDYAAIYgtAEAMAShDQCAIQhtAAAMYdl2M/zSPQAACDm2tAEAMAShDQCAIQht\nAAAMQWgDAGAIQhsAAEMQ2gAAGMIT7gGEyuOPP64vv/xSlmXpkUce0cUXXxyc9umnn2ru3Llyu91K\nTU3VuHHjwjjSpnnyySe1efNm+Xw+3XPPPbr66quD0wYOHKiOHTvK7XZLkubMmaOUlJRwDbVRNm7c\nqIkTJ+rcc8+VJJ133nmaMWNGcLqJj9Xrr7+ulStXBm9v3bpVX3zxRfD2RRddpB49egRvv/LKK8HH\nrDXasWOH7r//fo0ZM0YjR47Uvn37NGXKFPn9fiUlJWn27Nnyer217tPQ66+1qKuu6dOny+fzyePx\naPbs2UpKSgrOf7znamtwdE3Tpk3TN998o4SEBEnSXXfdpf79+9e6j4mP1YQJE5SXlydJys/PV7du\n3ZSenh6cf8WKFXr66ad15plnSpL69Omj++67Lyxjb3a2A23cuNEeO3asbdu2vXPnTnvYsGG1pl9z\nzTX2jz/+aPv9fvu2226zv/3223AMs8k2bNhg33333bZt2/bBgwftK6+8stb0AQMG2IcPHw7DyE7c\nZ599Zj/wwAP1Tjf1saq2ceNGe+bMmbXaevXqFabRNF1RUZE9cuRI+9FHH7UXLVpk27ZtT5s2zX7n\nnXds27btp556yl6yZEmt+xzv9dca1FXXlClT7Lffftu2bdtevHixnZGRUes+x3uuhltdNU2dOtX+\n8MMP672PqY9VTdOmTbO//PLLWm1vvPGG/Yc//KGlhtiiHLl7fMOGDRo0aJAkqXPnziooKNDhw4cl\nSbt371Z8fLxOOeUUuVwuXXnlldqwYUM4h9toPXv21NNPPy1Jatu2rUpKSuT3+8M8qtAx+bGq9txz\nz+n+++8P9zBOmNfr1YIFC5ScnBxs27hxo6666ipJ0oABA455TBp6/bUWddX12GOPKS0tTZKUmJio\n/Pz8cA3vhNRV0/GY+lhV27VrlwoLC1vl3oFQcWRo//TTT0pMTAzebteunXJyciRJOTk5ateuXZ3T\nWju3262YmBhJ0vLly5WamnrMbtXHHntMt912m+bMmSPbkJPd7dy5U/fee69uu+02rV+/Pthu8mMl\nSV999ZVOOeWUWrtYJam8vFyTJ0/WrbfeqpdffjlMo2scj8ejqKioWm0lJSXB3eHt27c/5jFp6PXX\nWtRVV0xMjNxut/x+v5YuXarrrrvumPvV91xtDeqqSZIWL16s0aNHa9KkSTp48GCtaaY+VtX+8z//\nUyNHjqxz2v/+7//qrrvu0h133KFt27aFcogtyrGfaddkSng11urVq7V8+XItXLiwVvuECRPUr18/\nxcfHa9y4ccrKytKQIUPCNMrGOeusszR+/Hhdc8012r17t0aPHq333nvvmM9ITbR8+XLdeOONx7RP\nmTJF119/vSzL0siRI/WLX/xCXbt2DcMI/3WNeW2Z9Prz+/2aMmWKLr/8cvXu3bvWNBOfq//2b/+m\nhIQE/fznP9f8+fP17LPP6ne/+12985v0WJWXl2vz5s2aOXPmMdMuueQStWvXTv3799cXX3yhqVOn\n6q9//WvLDzIEHLmlnZycrJ9++il4+8CBA8GtnaOnZWdnN2l3UritW7dOL774ohYsWKC4uLha0264\n4Qa1b99eHo9Hqamp2rFjR5hG2XgpKSkaOnSoLMvSmWeeqQ4dOig7O1uS+Y/Vxo0b1b1792Pab7vt\nNsXGxiomJkaXX365EY9TTTExMSotLZVU92PS0OuvtZs+fbo6deqk8ePHHzOtoedqa9W7d2/9/Oc/\nl1R5oOrRzzWTH6vPP/+83t3inTt3Dh5w1717dx08eNAxHyU6MrT79u2rrKwsSdI333yj5ORktWnT\nRpJ0+umn6/Dhw9qzZ498Pp/WrFmjvn37hnO4jVZYWKgnn3xSL730UvBo0JrT7rrrLpWXl0uqfEJX\nH+Xamq1cuVJ//vOfJVXuDs/NzQ0e8W7yY5Wdna3Y2NhjtsJ27dqlyZMny7Zt+Xw+bdmyxYjHqaY+\nffoEX1/vvfee+vXrV2t6Q6+/1mzlypWKiIjQhAkT6p1e33O1tXrggQe0e/duSZUrkUc/10x9rCTp\n66+/1gUXXFDntAULFuitt96SVHnkebt27Vr1NzSawrG/8jVnzhxt2rRJlmXpscce07Zt2xQXF6fB\ngwfr888/15w5cyRJV199te66664wj7Zxli1bpszMTJ199tnBtssuu0znn3++Bg8erL/85S968803\nFRkZqQsvvFAzZsyQZVlhHPHxHT58WA8//LAOHTqkiooKjR8/Xrm5ucY/Vlu3btW8efP0pz/9SZI0\nf/589ezZU927d9fs2bP12WefyeVyaeDAga36qyhbt25VRkaG9u7dK4/Ho5SUFM2ZM0fTpk1TWVmZ\nTj31VD3xxBOKiIjQpEmT9MQTTygqKuqY1199b67hUlddubm5ioyMDIZW586dNXPmzGBdPp/vmOfq\nlVdeGeZKjqirppEjR2r+/PmKjo5WTEyMnnjiCbVv3974xyozM1OZmZm69NJLNXTo0OC89913n154\n4QXt379fv/nNb4Irx631q2wnwrGhDQCA0zhy9zgAAE5EaAMAYAhCGwAAQxDaAAAYgtAGAMAQhDaA\nE7JixQo9/PDD4R4GcFIhtAEAMMRJce5x4GS2aNEivfvuu/L7/TrnnHN0991365577lFqaqq2b98u\nSfrjH/+olJQUffTRR3ruuecUFRWl6OhopaenKyUlRV9++aUef/xxRUREKD4+XhkZGZKOnBznu+++\n06mnnqpnn3221Z/QBzAZW9qAg3311Vd6//33tWTJEi1btkxxcXH69NNPtXv3bt10001aunSpevXq\npYULF6qkpESPPvqoMjMztWjRIqWmpmrevHmSpN/85jdKT0/X4sWL1bNnT3388ceSKn/1Kj09XStW\nrNC3336rb775JpzlAo7HljbgYBs3btQPP/yg0aNHS5KKi4uVnZ2thIQEdenSRZLUo0cP/eUvf9H3\n33+v9u3bq2PHjpKkXr166dVXX9XBgwd16NAhnXfeeZKkMWPGSKr8TLtr166Kjo6WVPmDGoWFhS1c\nIXByIbQBB/N6vRo4cGCtn2Pcs2ePbrrppuBt27ZlWdYxu7Vrttd3tuOjf4SBsyIDocXuccDBevTo\nobVr16qoqEiStGTJEuXk5KigoEDbtm2TJG3ZskXnn3++zjrrLOXm5urHH3+UJG3YsEGXXHKJEhMT\nlZCQoK+++kqStHDhQi1ZsiQ8BQEnOba0AQfr2rWrbr/9do0aNUqRkZFKTk7WZZddppSUFK1YsUJ/\n+MMfZNu25s6dq6ioKM2aNUuTJk2S1+tVTEyMZs2aJUmaPXu2Hn/8cXk8HsXFxWn27Nl67733wlwd\ncPLhV76Ak8yePXs0YsQIrV27NtxDAdBE7B4HAMAQbGkDAGAItrQBADAEoQ0AgCEIbQAADEFoAwBg\nCEIbAABDENoAABji/wPR2B2GtvKNBgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f07c52e3518>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "4teSZcQWatiM",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**testing model(load_model if not trained in above code snippet)**"
      ]
    },
    {
      "metadata": {
        "id": "KS-u5IHIa7wX",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "model = load_model('tranfer_20ep_l2+dropout.h5')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "HLVh3TXC0bOv",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "hf = h5py.File('data_1.h5', 'r')\n",
        "x = hf.get('dataset_1')\n",
        "y = hf.get('dataset_2')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "FQzscRgB0gTG",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          }
        }
      },
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "x = np.asarray(x)\n",
        "y = np.asarray(y)\n",
        "p = np.random.permutation(x.shape[0])\n",
        "x = x[p]\n",
        "y = y[p]\n",
        "X_test = x[0:100]\n",
        "y_test = y[0:100]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Y79E8s99bAwN",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "49f4c1a2-9008-47c2-df41-b8fe24c778d2",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525869864764,
          "user_tz": -330,
          "elapsed": 14186,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "history = model.evaluate(x=X_test, y=y_test, batch_size= 16 ,verbose=1)\n"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "100/100 [==============================] - 13s 131ms/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "Tuuk3sVd6B-a",
        "colab_type": "code",
        "colab": {
          "autoexec": {
            "startup": false,
            "wait_interval": 0
          },
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "52438dba-0dc7-43ba-d245-0f3f18aacd73",
        "executionInfo": {
          "status": "ok",
          "timestamp": 1525869912858,
          "user_tz": -330,
          "elapsed": 1109,
          "user": {
            "displayName": "Vaibhav Jade",
            "photoUrl": "https://lh3.googleusercontent.com/a/default-user=s128",
            "userId": "112426218802661198208"
          }
        }
      },
      "cell_type": "code",
      "source": [
        "history"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[14.822666625976563, 0.09]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    }
  ]
}