Linknet.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 6. LINKNET"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "GeEDd3r5Pr1q"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from PIL import Image\n",
    "import matplotlib.pyplot as plt\n",
    "import cv2\n",
    "import os\n",
    "from tqdm import tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "L2auScnCrGkC"
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "from tensorflow.keras.layers import Conv2D, BatchNormalization, Activation, Dropout, MaxPooling2D, Conv2DTranspose\n",
    "from tensorflow.keras.layers import Concatenate, Add\n",
    "from tensorflow.keras import Input, Model\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from tensorflow.keras.losses import binary_crossentropy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.1 Loading Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "RxDfZ84IPx2P"
   },
   "outputs": [],
   "source": [
    "path = '/content/drive/MyDrive/cs2/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "YK9fL3MpQKdJ"
   },
   "outputs": [],
   "source": [
    "#loading train and validation data\n",
    "train_data = pd.read_csv(path + \"data/train_data.csv\").fillna('')\n",
    "validtn_data = pd.read_csv(path + \"data/validtn_data.csv\").fillna('')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.2 Data Generator Implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "IYpZvFq2P4-e"
   },
   "outputs": [],
   "source": [
    "# Implementing custom data generator\n",
    "#https://towardsdatascience.com/implementing-custom-data-generators-in-keras-de56f013581c\n",
    "class DataGenerator(tf.keras.utils.Sequence):\n",
    "    def __init__(self, dataframe, batch_size=32, num_classes=None, shuffle=True, preprocess=None):\n",
    "        self.batch_size = batch_size\n",
    "        self.df = dataframe\n",
    "        self.indices = self.df.index.tolist()\n",
    "        self.preprocess = preprocess\n",
    "        self.shuffle = shuffle\n",
    "        self.on_epoch_end()\n",
    "\n",
    "    def __len__(self):\n",
    "        return len(self.indices) // (self.batch_size)\n",
    "\n",
    "    def __getitem__(self, index):\n",
    "        index = self.index[index * self.batch_size:(index + 1) * self.batch_size]\n",
    "        batch = [self.indices[k] for k in index]\n",
    "        \n",
    "        X, y = self.__get_data(batch)\n",
    "        return X, y\n",
    "\n",
    "    def on_epoch_end(self):\n",
    "        self.index = np.arange(len(self.indices))\n",
    "        if self.shuffle == True:\n",
    "            np.random.shuffle(self.index)\n",
    "\n",
    "    def __get_data(self, batch):\n",
    "        train_datagen = ImageDataGenerator()\n",
    "        param = {'flip_horizontal':True, 'samplewise_std_normalization' : True,\n",
    "                 'width_shift_range':0.1, 'height_shift_range':0.1,}\n",
    "        X = np.empty((self.batch_size,256,1600,3),dtype=np.float32) # image place-holders\n",
    "        Y = np.empty((self.batch_size,256,1600,4),dtype=np.float32)# 4 masks place-holders\n",
    "              \n",
    "        for i, id in enumerate(batch):\n",
    "          img = Image.open(path+'data/train_images/' + str(self.df['ImageId'].loc[id]))\n",
    "          X[i,] = train_datagen.apply_transform(x = img, transform_parameters = param)#input image\n",
    "          for j in range(4): #looping for each class\n",
    "                mask = rle2mask(self.df['rle_'+str(j+1)].loc[id])\n",
    "                Y[i,:,:,j] = train_datagen.apply_transform(x = mask, transform_parameters = param)#mask for each class\n",
    "                \n",
    "        # preprocess input\n",
    "        if self.preprocess!=None: X = self.preprocess(X)\n",
    "\n",
    "        return X, Y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.3 Utility Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "zUZbT_k2QF5U"
   },
   "outputs": [],
   "source": [
    "#https://www.kaggle.com/titericz/building-and-visualizing-masks\n",
    "#https://www.kaggle.com/paulorzp/rle-functions-run-lenght-encode-decode\n",
    "\n",
    "#defining function for converting EncodedPixels(rle: run length encoding) to mask\n",
    "def rle2mask(rle_string, img_shape=(256,1600)):\n",
    "    '''\n",
    "    input: EncodedPixels (run-length-encoded) string & image shape:-(width,height)\n",
    "    output: mask in numpy.ndarray format with shape (256,1600)\n",
    "    '''\n",
    "    rle_array = np.array([int(s)for s in rle_string.split()])\n",
    "    starts_array = rle_array[::2]-1\n",
    "    lengths_array = rle_array[1::2]\n",
    "    mask_array = np.zeros(img_shape[0]*img_shape[1],dtype=np.uint8)\n",
    "    #print(starts_array,lengths_array)\n",
    "    for i in range(len(starts_array)):\n",
    "        mask_array[starts_array[i]:starts_array[i]+lengths_array[i]] = 1\n",
    "    #order='F' because encoded pixels are numbered from top to bottom, then left to right\n",
    "    return mask_array.reshape(img_shape, order = 'F')\n",
    "\n",
    "#defining function for converting given mask to EncodedPixels(rle: run length encoding)\n",
    "def mask2rle(mask_array):\n",
    "    '''\n",
    "    input: mask in numpy.ndarray format\n",
    "    output: EncodedPixels (run-length-encoded) string\n",
    "    '''\n",
    "    mask_array = mask_array.T.flatten()\n",
    "    mask_array = np.concatenate([[0], mask_array, [0]])\n",
    "    rle_array = np.where(mask_array[1:]!=mask_array[:-1])[0]+1\n",
    "    rle_array[1::2] -= rle_array[::2]\n",
    "    rle_string = ' '.join(map(str,rle_array))\n",
    "    return rle_string\n",
    "\n",
    "#defining function for calculation of metric dice coefficient\n",
    "def dice_coefficient(y_true, y_pred):\n",
    "    y_true_f = tf.reshape(y_true, [-1])\n",
    "    y_pred_f = tf.reshape(y_pred, [-1])\n",
    "    intersection = tf.math.reduce_sum(y_true_f * y_pred_f)\n",
    "    smoothing_const = 1e-9\n",
    "    return (2. * intersection + smoothing_const) / (tf.math.reduce_sum(y_true_f) + tf.math.reduce_sum(y_pred_f) + smoothing_const)\n",
    "\n",
    "#defining function for calculation of dice coefficient\n",
    "def dice_loss(y_true, y_pred):\n",
    "    y_true_f = tf.reshape(y_true, [-1])\n",
    "    y_pred_f = tf.reshape(y_pred, [-1])\n",
    "    return (1-dice_coefficient(y_true, y_pred))\n",
    "\n",
    "#defining function for calculation of loss function: binary cross entropy + dice loss\n",
    "def bce_dice_loss(y_true, y_pred):\n",
    "    y_true_f = tf.reshape(y_true, [-1])\n",
    "    y_pred_f = tf.reshape(y_pred, [-1])\n",
    "    return binary_crossentropy(y_true, y_pred) + (1-dice_coefficient(y_true, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.4 Defining LinkNet Architecture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "4B3RaRGcUCJv"
   },
   "outputs": [],
   "source": [
    "# defining LinkNet architecture\n",
    "# https://arxiv.org/abs/1707.03718\n",
    "# https://github.com/qubvel/segmentation_models/blob/master/segmentation_models/models/linknet.py\n",
    "# https://towardsdatascience.com/understanding-semantic-segmentation-with-unet-6be4f42d4b47\n",
    "# https://github.com/hlamba28/UNET-TGS\n",
    "\n",
    "def conv2D_block(input_tensor, n_filters, kernel_size = 3, batchnorm = True ):\n",
    "    \"\"\"function to pass Input_tensor through 2- Conv2D layers configured as per the input parameters\"\"\"\n",
    "    # first Conv2D layer\n",
    "    x = Conv2D(filters = n_filters, kernel_size = (kernel_size, kernel_size),\\\n",
    "            kernel_initializer = 'he_normal', padding = 'same')(input_tensor)\n",
    "    if batchnorm:\n",
    "        x = BatchNormalization()(x)\n",
    "        x = Activation('relu')(x)\n",
    "\n",
    "    # second Conv2D layer\n",
    "    x = Conv2D(filters = n_filters, kernel_size = (kernel_size, kernel_size),\\\n",
    "            kernel_initializer = 'he_normal', padding = 'same')(input_tensor)\n",
    "    if batchnorm:\n",
    "        x = BatchNormalization()(x)\n",
    "        x = Activation('relu')(x)\n",
    "\n",
    "    return x\n",
    "\n",
    "def Linknet_Model(input_image, n_filters = 16, dropout = 0.1, batchnorm = True):\n",
    "    # Encoder (Contraction Path)\n",
    "    E1 = conv2D_block(input_img, n_filters * 1, kernel_size = 3, batchnorm = batchnorm)\n",
    "    p1 = MaxPooling2D((2, 2))(E1)\n",
    "    p1 = Dropout(dropout)(p1)\n",
    "\n",
    "    E2 = conv2D_block(p1, n_filters * 2, kernel_size = 3, batchnorm = batchnorm)\n",
    "    p2 = MaxPooling2D((2, 2))(E2)\n",
    "    p2 = Dropout(dropout)(p2)\n",
    "\n",
    "    E3 = conv2D_block(p2, n_filters * 4, kernel_size = 3, batchnorm = batchnorm)\n",
    "    p3 = MaxPooling2D((2, 2))(E3)\n",
    "    p3 = Dropout(dropout)(p3)\n",
    "\n",
    "    E4 = conv2D_block(p3, n_filters * 8, kernel_size = 3, batchnorm = batchnorm)\n",
    "    p4 = MaxPooling2D((2, 2))(E4)\n",
    "    p4 = Dropout(dropout)(p4)\n",
    "\n",
    "    E5 = conv2D_block(p4, n_filters = n_filters * 16, kernel_size = 3, batchnorm = batchnorm)\n",
    "\n",
    "    # Decoder (Expansive Path)\n",
    "    D6 = Conv2DTranspose(n_filters * 8, kernel_size = (3, 3), strides = (2, 2), padding = 'same')(E5)\n",
    "    D6 = Add()([D6, E4])\n",
    "    D6 = Dropout(dropout)(D6)\n",
    "    E6 = conv2D_block(D6, n_filters * 8, kernel_size = 3, batchnorm = batchnorm)\n",
    "\n",
    "    D7 = Conv2DTranspose(n_filters * 4, (3, 3), strides = (2, 2), padding = 'same')(E6)\n",
    "    D7 = Add()([D7, E3])\n",
    "    D7 = Dropout(dropout)(D7)\n",
    "    E7 = conv2D_block(D7, n_filters * 4, kernel_size = 3, batchnorm = batchnorm)\n",
    "\n",
    "    D8 = Conv2DTranspose(n_filters * 2, (3, 3), strides = (2, 2), padding = 'same')(E7)\n",
    "    D8 = Add()([D8, E2])\n",
    "    D8 = Dropout(dropout)(D8)\n",
    "    E8 = conv2D_block(D8, n_filters * 2, kernel_size = 3, batchnorm = batchnorm)\n",
    "\n",
    "    D9 = Conv2DTranspose(n_filters * 1, (3, 3), strides = (2, 2), padding = 'same')(E8)\n",
    "    D9 = Add()([D9, E1])\n",
    "    D9 = Dropout(dropout)(D9)\n",
    "    E9 = conv2D_block(D9, n_filters * 1, kernel_size = 3, batchnorm = batchnorm)\n",
    "\n",
    "    outputs = Conv2D(4, (1, 1), activation='sigmoid')(E9)\n",
    "    model = Model(inputs=[input_img], outputs=[outputs])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "EI2OgYwyU3DW",
    "outputId": "5d44aa22-0eab-490b-c861-e4126a46ba2d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "img (InputLayer)                [(None, 256, 1600, 3 0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_1 (Conv2D)               (None, 256, 1600, 8) 224         img[0][0]                        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_1 (BatchNor (None, 256, 1600, 8) 32          conv2d_1[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "activation_1 (Activation)       (None, 256, 1600, 8) 0           batch_normalization_1[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d (MaxPooling2D)    (None, 128, 800, 8)  0           activation_1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "dropout (Dropout)               (None, 128, 800, 8)  0           max_pooling2d[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_3 (Conv2D)               (None, 128, 800, 16) 1168        dropout[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_3 (BatchNor (None, 128, 800, 16) 64          conv2d_3[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "activation_3 (Activation)       (None, 128, 800, 16) 0           batch_normalization_3[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2D)  (None, 64, 400, 16)  0           activation_3[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "dropout_1 (Dropout)             (None, 64, 400, 16)  0           max_pooling2d_1[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_5 (Conv2D)               (None, 64, 400, 32)  4640        dropout_1[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_5 (BatchNor (None, 64, 400, 32)  128         conv2d_5[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "activation_5 (Activation)       (None, 64, 400, 32)  0           batch_normalization_5[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2D)  (None, 32, 200, 32)  0           activation_5[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "dropout_2 (Dropout)             (None, 32, 200, 32)  0           max_pooling2d_2[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_7 (Conv2D)               (None, 32, 200, 64)  18496       dropout_2[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_7 (BatchNor (None, 32, 200, 64)  256         conv2d_7[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "activation_7 (Activation)       (None, 32, 200, 64)  0           batch_normalization_7[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_3 (MaxPooling2D)  (None, 16, 100, 64)  0           activation_7[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "dropout_3 (Dropout)             (None, 16, 100, 64)  0           max_pooling2d_3[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_9 (Conv2D)               (None, 16, 100, 128) 73856       dropout_3[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_9 (BatchNor (None, 16, 100, 128) 512         conv2d_9[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "activation_9 (Activation)       (None, 16, 100, 128) 0           batch_normalization_9[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_transpose (Conv2DTranspo (None, 32, 200, 64)  73792       activation_9[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "add (Add)                       (None, 32, 200, 64)  0           conv2d_transpose[0][0]           \n",
      "                                                                 activation_7[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "dropout_4 (Dropout)             (None, 32, 200, 64)  0           add[0][0]                        \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_11 (Conv2D)              (None, 32, 200, 64)  36928       dropout_4[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_11 (BatchNo (None, 32, 200, 64)  256         conv2d_11[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_11 (Activation)      (None, 32, 200, 64)  0           batch_normalization_11[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_transpose_1 (Conv2DTrans (None, 64, 400, 32)  18464       activation_11[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "add_1 (Add)                     (None, 64, 400, 32)  0           conv2d_transpose_1[0][0]         \n",
      "                                                                 activation_5[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "dropout_5 (Dropout)             (None, 64, 400, 32)  0           add_1[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_13 (Conv2D)              (None, 64, 400, 32)  9248        dropout_5[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_13 (BatchNo (None, 64, 400, 32)  128         conv2d_13[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_13 (Activation)      (None, 64, 400, 32)  0           batch_normalization_13[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_transpose_2 (Conv2DTrans (None, 128, 800, 16) 4624        activation_13[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "add_2 (Add)                     (None, 128, 800, 16) 0           conv2d_transpose_2[0][0]         \n",
      "                                                                 activation_3[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "dropout_6 (Dropout)             (None, 128, 800, 16) 0           add_2[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_15 (Conv2D)              (None, 128, 800, 16) 2320        dropout_6[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_15 (BatchNo (None, 128, 800, 16) 64          conv2d_15[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_15 (Activation)      (None, 128, 800, 16) 0           batch_normalization_15[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_transpose_3 (Conv2DTrans (None, 256, 1600, 8) 1160        activation_15[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "add_3 (Add)                     (None, 256, 1600, 8) 0           conv2d_transpose_3[0][0]         \n",
      "                                                                 activation_1[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "dropout_7 (Dropout)             (None, 256, 1600, 8) 0           add_3[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_17 (Conv2D)              (None, 256, 1600, 8) 584         dropout_7[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_17 (BatchNo (None, 256, 1600, 8) 32          conv2d_17[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "activation_17 (Activation)      (None, 256, 1600, 8) 0           batch_normalization_17[0][0]     \n",
      "__________________________________________________________________________________________________\n",
      "conv2d_18 (Conv2D)              (None, 256, 1600, 4) 36          activation_17[0][0]              \n",
      "==================================================================================================\n",
      "Total params: 247,012\n",
      "Trainable params: 246,276\n",
      "Non-trainable params: 736\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "#configuring input tensor\n",
    "input_img = Input((256, 1600, 3), name='img')\n",
    "\n",
    "model = Linknet_Model(input_img, n_filters=8, dropout=0.2, batchnorm=True)\n",
    "model.compile(optimizer=Adam(), loss=bce_dice_loss, metrics=[dice_coefficient])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "5B1lE7l2XVWI"
   },
   "outputs": [],
   "source": [
    "#loading train & validation data using DataGenerator class defined\n",
    "train_batches = DataGenerator(train_data,shuffle=True)\n",
    "validtn_batches = DataGenerator(validtn_data,shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "75eCO0A1XWx_"
   },
   "outputs": [],
   "source": [
    "from tensorflow.keras.callbacks import ModelCheckpoint, TensorBoard\n",
    "import datetime, os\n",
    "\n",
    "# #tensorboard for visualizing loss & metric summary \n",
    "# logdir = path+\"logs/\" + datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\")\n",
    "# tensorboard_callback = TensorBoard(logdir)\n",
    "\n",
    "#defining ModelCheckpoint to make a checkpoint when validation dice coefficient improves\n",
    "checkpoints = ModelCheckpoint(path+'linknet.h5', \\\n",
    "                              monitor='val_dice_coefficient', verbose=1, save_best_only=True, mode='max')\n",
    "\n",
    "#adding ModelCheckpoint & TensorBoard to callback functions list\n",
    "callbacks_list = [checkpoints]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.5 Traing Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "C6U5HMm9XZoh",
    "outputId": "3f2ee0fa-bff7-4d76-89cb-5e485e0e4d7a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/60\n",
      "333/333 [==============================] - 5202s 16s/step - loss: 1.5593 - dice_coefficient: 0.0142 - val_loss: 1.1646 - val_dice_coefficient: 0.0187\n",
      "\n",
      "Epoch 00001: val_dice_coefficient improved from -inf to 0.01875, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 2/60\n",
      "333/333 [==============================] - 337s 1s/step - loss: 1.1302 - dice_coefficient: 0.0287 - val_loss: 1.0261 - val_dice_coefficient: 0.0380\n",
      "\n",
      "Epoch 00002: val_dice_coefficient improved from 0.01875 to 0.03799, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 3/60\n",
      "333/333 [==============================] - 329s 988ms/step - loss: 0.8595 - dice_coefficient: 0.1935 - val_loss: 0.6708 - val_dice_coefficient: 0.3774\n",
      "\n",
      "Epoch 00003: val_dice_coefficient improved from 0.03799 to 0.37742, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 4/60\n",
      "333/333 [==============================] - 328s 983ms/step - loss: 0.5978 - dice_coefficient: 0.4415 - val_loss: 0.6224 - val_dice_coefficient: 0.4388\n",
      "\n",
      "Epoch 00004: val_dice_coefficient improved from 0.37742 to 0.43883, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 5/60\n",
      "333/333 [==============================] - 328s 983ms/step - loss: 0.5564 - dice_coefficient: 0.4842 - val_loss: 0.5790 - val_dice_coefficient: 0.4701\n",
      "\n",
      "Epoch 00005: val_dice_coefficient improved from 0.43883 to 0.47006, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 6/60\n",
      "333/333 [==============================] - 327s 981ms/step - loss: 0.5596 - dice_coefficient: 0.4823 - val_loss: 0.5911 - val_dice_coefficient: 0.4627\n",
      "\n",
      "Epoch 00006: val_dice_coefficient did not improve from 0.47006\n",
      "Epoch 7/60\n",
      "333/333 [==============================] - 328s 983ms/step - loss: 0.5380 - dice_coefficient: 0.5024 - val_loss: 0.5599 - val_dice_coefficient: 0.4886\n",
      "\n",
      "Epoch 00007: val_dice_coefficient improved from 0.47006 to 0.48862, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 8/60\n",
      "333/333 [==============================] - 328s 983ms/step - loss: 0.5241 - dice_coefficient: 0.5151 - val_loss: 0.5590 - val_dice_coefficient: 0.4878\n",
      "\n",
      "Epoch 00008: val_dice_coefficient did not improve from 0.48862\n",
      "Epoch 9/60\n",
      "333/333 [==============================] - 329s 985ms/step - loss: 0.5180 - dice_coefficient: 0.5201 - val_loss: 0.5691 - val_dice_coefficient: 0.4810\n",
      "\n",
      "Epoch 00009: val_dice_coefficient did not improve from 0.48862\n",
      "Epoch 10/60\n",
      "333/333 [==============================] - 327s 982ms/step - loss: 0.5083 - dice_coefficient: 0.5302 - val_loss: 0.5352 - val_dice_coefficient: 0.5124\n",
      "\n",
      "Epoch 00010: val_dice_coefficient improved from 0.48862 to 0.51241, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 11/60\n",
      "333/333 [==============================] - 324s 971ms/step - loss: 0.5107 - dice_coefficient: 0.5269 - val_loss: 0.5426 - val_dice_coefficient: 0.5033\n",
      "\n",
      "Epoch 00011: val_dice_coefficient did not improve from 0.51241\n",
      "Epoch 12/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.5133 - dice_coefficient: 0.5236 - val_loss: 0.5377 - val_dice_coefficient: 0.5105\n",
      "\n",
      "Epoch 00012: val_dice_coefficient did not improve from 0.51241\n",
      "Epoch 13/60\n",
      "333/333 [==============================] - 322s 967ms/step - loss: 0.5037 - dice_coefficient: 0.5330 - val_loss: 0.5468 - val_dice_coefficient: 0.4983\n",
      "\n",
      "Epoch 00013: val_dice_coefficient did not improve from 0.51241\n",
      "Epoch 14/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.5067 - dice_coefficient: 0.5292 - val_loss: 0.5178 - val_dice_coefficient: 0.5265\n",
      "\n",
      "Epoch 00014: val_dice_coefficient improved from 0.51241 to 0.52652, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 15/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4936 - dice_coefficient: 0.5417 - val_loss: 0.5243 - val_dice_coefficient: 0.5196\n",
      "\n",
      "Epoch 00015: val_dice_coefficient did not improve from 0.52652\n",
      "Epoch 16/60\n",
      "333/333 [==============================] - 323s 970ms/step - loss: 0.4855 - dice_coefficient: 0.5518 - val_loss: 0.5104 - val_dice_coefficient: 0.5324\n",
      "\n",
      "Epoch 00016: val_dice_coefficient improved from 0.52652 to 0.53241, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 17/60\n",
      "333/333 [==============================] - 323s 970ms/step - loss: 0.4912 - dice_coefficient: 0.5449 - val_loss: 0.5198 - val_dice_coefficient: 0.5255\n",
      "\n",
      "Epoch 00017: val_dice_coefficient did not improve from 0.53241\n",
      "Epoch 18/60\n",
      "333/333 [==============================] - 323s 970ms/step - loss: 0.4788 - dice_coefficient: 0.5569 - val_loss: 0.5025 - val_dice_coefficient: 0.5362\n",
      "\n",
      "Epoch 00018: val_dice_coefficient improved from 0.53241 to 0.53623, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 19/60\n",
      "333/333 [==============================] - 322s 967ms/step - loss: 0.4718 - dice_coefficient: 0.5633 - val_loss: 0.5574 - val_dice_coefficient: 0.4936\n",
      "\n",
      "Epoch 00019: val_dice_coefficient did not improve from 0.53623\n",
      "Epoch 20/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4787 - dice_coefficient: 0.5557 - val_loss: 0.4887 - val_dice_coefficient: 0.5556\n",
      "\n",
      "Epoch 00020: val_dice_coefficient improved from 0.53623 to 0.55564, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 21/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4679 - dice_coefficient: 0.5649 - val_loss: 0.4952 - val_dice_coefficient: 0.5478\n",
      "\n",
      "Epoch 00021: val_dice_coefficient did not improve from 0.55564\n",
      "Epoch 22/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4482 - dice_coefficient: 0.5845 - val_loss: 0.4948 - val_dice_coefficient: 0.5481\n",
      "\n",
      "Epoch 00022: val_dice_coefficient did not improve from 0.55564\n",
      "Epoch 23/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4588 - dice_coefficient: 0.5750 - val_loss: 0.4918 - val_dice_coefficient: 0.5547\n",
      "\n",
      "Epoch 00023: val_dice_coefficient did not improve from 0.55564\n",
      "Epoch 24/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4470 - dice_coefficient: 0.5859 - val_loss: 0.4435 - val_dice_coefficient: 0.5927\n",
      "\n",
      "Epoch 00024: val_dice_coefficient improved from 0.55564 to 0.59272, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 25/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4510 - dice_coefficient: 0.5829 - val_loss: 0.4594 - val_dice_coefficient: 0.5801\n",
      "\n",
      "Epoch 00025: val_dice_coefficient did not improve from 0.59272\n",
      "Epoch 26/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4496 - dice_coefficient: 0.5838 - val_loss: 0.4539 - val_dice_coefficient: 0.5853\n",
      "\n",
      "Epoch 00026: val_dice_coefficient did not improve from 0.59272\n",
      "Epoch 27/60\n",
      "333/333 [==============================] - 323s 969ms/step - loss: 0.4444 - dice_coefficient: 0.5883 - val_loss: 0.4536 - val_dice_coefficient: 0.5841\n",
      "\n",
      "Epoch 00027: val_dice_coefficient did not improve from 0.59272\n",
      "Epoch 28/60\n",
      "333/333 [==============================] - 323s 969ms/step - loss: 0.4338 - dice_coefficient: 0.5979 - val_loss: 0.4415 - val_dice_coefficient: 0.5970\n",
      "\n",
      "Epoch 00028: val_dice_coefficient improved from 0.59272 to 0.59702, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 29/60\n",
      "333/333 [==============================] - 326s 977ms/step - loss: 0.4367 - dice_coefficient: 0.5956 - val_loss: 0.4698 - val_dice_coefficient: 0.5714\n",
      "\n",
      "Epoch 00029: val_dice_coefficient did not improve from 0.59702\n",
      "Epoch 30/60\n",
      "333/333 [==============================] - 327s 980ms/step - loss: 0.4263 - dice_coefficient: 0.6054 - val_loss: 0.4706 - val_dice_coefficient: 0.5701\n",
      "\n",
      "Epoch 00030: val_dice_coefficient did not improve from 0.59702\n",
      "Epoch 31/60\n",
      "333/333 [==============================] - 323s 969ms/step - loss: 0.4251 - dice_coefficient: 0.6061 - val_loss: 0.4468 - val_dice_coefficient: 0.5924\n",
      "\n",
      "Epoch 00031: val_dice_coefficient did not improve from 0.59702\n",
      "Epoch 32/60\n",
      "333/333 [==============================] - 323s 969ms/step - loss: 0.4247 - dice_coefficient: 0.6072 - val_loss: 0.4437 - val_dice_coefficient: 0.5951\n",
      "\n",
      "Epoch 00032: val_dice_coefficient did not improve from 0.59702\n",
      "Epoch 33/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4201 - dice_coefficient: 0.6105 - val_loss: 0.4607 - val_dice_coefficient: 0.5775\n",
      "\n",
      "Epoch 00033: val_dice_coefficient did not improve from 0.59702\n",
      "Epoch 34/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4183 - dice_coefficient: 0.6131 - val_loss: 0.4939 - val_dice_coefficient: 0.5475\n",
      "\n",
      "Epoch 00034: val_dice_coefficient did not improve from 0.59702\n",
      "Epoch 35/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4216 - dice_coefficient: 0.6089 - val_loss: 0.4723 - val_dice_coefficient: 0.5711\n",
      "\n",
      "Epoch 00035: val_dice_coefficient did not improve from 0.59702\n",
      "Epoch 36/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4139 - dice_coefficient: 0.6171 - val_loss: 0.4699 - val_dice_coefficient: 0.5746\n",
      "\n",
      "Epoch 00036: val_dice_coefficient did not improve from 0.59702\n",
      "Epoch 37/60\n",
      "333/333 [==============================] - 322s 967ms/step - loss: 0.4282 - dice_coefficient: 0.6027 - val_loss: 0.4360 - val_dice_coefficient: 0.6022\n",
      "\n",
      "Epoch 00037: val_dice_coefficient improved from 0.59702 to 0.60222, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 38/60\n",
      "333/333 [==============================] - 322s 967ms/step - loss: 0.4129 - dice_coefficient: 0.6176 - val_loss: 0.4276 - val_dice_coefficient: 0.6092\n",
      "\n",
      "Epoch 00038: val_dice_coefficient improved from 0.60222 to 0.60917, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 39/60\n",
      "333/333 [==============================] - 323s 967ms/step - loss: 0.4184 - dice_coefficient: 0.6113 - val_loss: 0.4669 - val_dice_coefficient: 0.5755\n",
      "\n",
      "Epoch 00039: val_dice_coefficient did not improve from 0.60917\n",
      "Epoch 40/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4132 - dice_coefficient: 0.6172 - val_loss: 0.4693 - val_dice_coefficient: 0.5703\n",
      "\n",
      "Epoch 00040: val_dice_coefficient did not improve from 0.60917\n",
      "Epoch 41/60\n",
      "333/333 [==============================] - 322s 967ms/step - loss: 0.4172 - dice_coefficient: 0.6131 - val_loss: 0.4522 - val_dice_coefficient: 0.5888\n",
      "\n",
      "Epoch 00041: val_dice_coefficient did not improve from 0.60917\n",
      "Epoch 42/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4089 - dice_coefficient: 0.6209 - val_loss: 0.4251 - val_dice_coefficient: 0.6124\n",
      "\n",
      "Epoch 00042: val_dice_coefficient improved from 0.60917 to 0.61243, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 43/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4142 - dice_coefficient: 0.6160 - val_loss: 0.4313 - val_dice_coefficient: 0.6067\n",
      "\n",
      "Epoch 00043: val_dice_coefficient did not improve from 0.61243\n",
      "Epoch 44/60\n",
      "333/333 [==============================] - 323s 970ms/step - loss: 0.4068 - dice_coefficient: 0.6225 - val_loss: 0.5124 - val_dice_coefficient: 0.5389\n",
      "\n",
      "Epoch 00044: val_dice_coefficient did not improve from 0.61243\n",
      "Epoch 45/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4113 - dice_coefficient: 0.6182 - val_loss: 0.4255 - val_dice_coefficient: 0.6123\n",
      "\n",
      "Epoch 00045: val_dice_coefficient did not improve from 0.61243\n",
      "Epoch 46/60\n",
      "333/333 [==============================] - 323s 970ms/step - loss: 0.3963 - dice_coefficient: 0.6336 - val_loss: 0.4482 - val_dice_coefficient: 0.5917\n",
      "\n",
      "Epoch 00046: val_dice_coefficient did not improve from 0.61243\n",
      "Epoch 47/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.4081 - dice_coefficient: 0.6231 - val_loss: 0.4237 - val_dice_coefficient: 0.6139\n",
      "\n",
      "Epoch 00047: val_dice_coefficient improved from 0.61243 to 0.61390, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 48/60\n",
      "333/333 [==============================] - 323s 970ms/step - loss: 0.3986 - dice_coefficient: 0.6317 - val_loss: 0.4334 - val_dice_coefficient: 0.6046\n",
      "\n",
      "Epoch 00048: val_dice_coefficient did not improve from 0.61390\n",
      "Epoch 49/60\n",
      "333/333 [==============================] - 324s 971ms/step - loss: 0.3945 - dice_coefficient: 0.6351 - val_loss: 0.4158 - val_dice_coefficient: 0.6211\n",
      "\n",
      "Epoch 00049: val_dice_coefficient improved from 0.61390 to 0.62114, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 50/60\n",
      "333/333 [==============================] - 323s 969ms/step - loss: 0.3998 - dice_coefficient: 0.6292 - val_loss: 0.4169 - val_dice_coefficient: 0.6201\n",
      "\n",
      "Epoch 00050: val_dice_coefficient did not improve from 0.62114\n",
      "Epoch 51/60\n",
      "333/333 [==============================] - 323s 969ms/step - loss: 0.3987 - dice_coefficient: 0.6302 - val_loss: 0.4686 - val_dice_coefficient: 0.5733\n",
      "\n",
      "Epoch 00051: val_dice_coefficient did not improve from 0.62114\n",
      "Epoch 52/60\n",
      "333/333 [==============================] - 325s 975ms/step - loss: 0.3936 - dice_coefficient: 0.6349 - val_loss: 0.4323 - val_dice_coefficient: 0.6057\n",
      "\n",
      "Epoch 00052: val_dice_coefficient did not improve from 0.62114\n",
      "Epoch 53/60\n",
      "333/333 [==============================] - 323s 969ms/step - loss: 0.4049 - dice_coefficient: 0.6248 - val_loss: 0.4107 - val_dice_coefficient: 0.6273\n",
      "\n",
      "Epoch 00053: val_dice_coefficient improved from 0.62114 to 0.62729, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 54/60\n",
      "333/333 [==============================] - 323s 969ms/step - loss: 0.3923 - dice_coefficient: 0.6361 - val_loss: 0.4052 - val_dice_coefficient: 0.6295\n",
      "\n",
      "Epoch 00054: val_dice_coefficient improved from 0.62729 to 0.62949, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 55/60\n",
      "333/333 [==============================] - 323s 969ms/step - loss: 0.3956 - dice_coefficient: 0.6333 - val_loss: 0.4056 - val_dice_coefficient: 0.6311\n",
      "\n",
      "Epoch 00055: val_dice_coefficient improved from 0.62949 to 0.63115, saving model to /content/drive/MyDrive/cs2/linknet.h5\n",
      "Epoch 56/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.3934 - dice_coefficient: 0.6356 - val_loss: 0.4328 - val_dice_coefficient: 0.6050\n",
      "\n",
      "Epoch 00056: val_dice_coefficient did not improve from 0.63115\n",
      "Epoch 57/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.3880 - dice_coefficient: 0.6402 - val_loss: 0.4270 - val_dice_coefficient: 0.6089\n",
      "\n",
      "Epoch 00057: val_dice_coefficient did not improve from 0.63115\n",
      "Epoch 58/60\n",
      "333/333 [==============================] - 323s 969ms/step - loss: 0.4040 - dice_coefficient: 0.6256 - val_loss: 0.4783 - val_dice_coefficient: 0.5636\n",
      "\n",
      "Epoch 00058: val_dice_coefficient did not improve from 0.63115\n",
      "Epoch 59/60\n",
      "333/333 [==============================] - 323s 969ms/step - loss: 0.3862 - dice_coefficient: 0.6427 - val_loss: 0.4167 - val_dice_coefficient: 0.6199\n",
      "\n",
      "Epoch 00059: val_dice_coefficient did not improve from 0.63115\n",
      "Epoch 60/60\n",
      "333/333 [==============================] - 323s 968ms/step - loss: 0.3905 - dice_coefficient: 0.6387 - val_loss: 0.4206 - val_dice_coefficient: 0.6178\n",
      "\n",
      "Epoch 00060: val_dice_coefficient did not improve from 0.63115\n"
     ]
    }
   ],
   "source": [
    "# fitting model\n",
    "history = model.fit(train_batches,validation_data=validtn_batches, epochs = 60, callbacks=callbacks_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "6pYAJ4ncfA8l",
    "outputId": "36c7f037-aed9-49a7-b7a6-2df2a0310801",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualizing history for dice_coefficient\n",
    "plt.figure(figsize=(12, 8))\n",
    "plt.plot(history.history['dice_coefficient'])\n",
    "plt.plot(history.history['val_dice_coefficient'])\n",
    "plt.title('model dice_coefficient')\n",
    "plt.ylabel('dice_coefficient')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'Validation'], loc='upper left')\n",
    "plt.show()\n",
    "# Visualizing history for loss\n",
    "plt.figure(figsize=(12, 8))\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('bce_dice_loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'Validation'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "y_oDaBHsfEP_"
   },
   "outputs": [],
   "source": [
    "# saving history data for first 60 epochs in csv file\n",
    "history1 = pd.DataFrame(history.history)\n",
    "history1.to_csv(path + \"data/history1-linknet.csv\", index=True)\n",
    "# saving model(weights upto 60 epochs)\n",
    "model.save(path+'linknet-60epochs.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.6 Visualizing Model Predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ZgE0kbY4gbnm",
    "outputId": "9b17d3dc-82bc-4341-db2c-235c4a6cad00"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "58/58 [==============================] - 38s 651ms/step - loss: 0.4056 - dice_coefficient: 0.6311\n",
      "loss: 0.40556466579437256 dice_coefficient: 0.631147027015686\n"
     ]
    }
   ],
   "source": [
    "# evaluating model(best weights) with the validation batch\n",
    "model.load_weights(path + 'linknet.h5')\n",
    "loss, dc= model.evaluate(validtn_batches,verbose=1)\n",
    "print('loss:',loss, 'dice_coefficient:',dc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "m7kmLfigQwlb"
   },
   "outputs": [],
   "source": [
    "#defining function to display GroundTruth & Predicted mask images for a given imageId\n",
    "def visualize_model_prediction(img_id):\n",
    "    fig, axs = plt.subplots(4, 3, figsize=(16,8))\n",
    "    img_obj = cv2.imread(path+'data/train_images/'+ img_id)\n",
    "    masks_actual = full_data[full_data['ImageId'] == img_id]\n",
    "    x = np.empty((1,256,1600,3),dtype=np.float32) # image place-holders\n",
    "    x[0,] = Image.open(path+'data/train_images/' + img_id)\n",
    "    masks_predicted = model.predict(x)\n",
    "    for i in range(4):\n",
    "        axs[i,0].imshow(img_obj)\n",
    "        axs[i,0].set_title(img_id)\n",
    "        axs[i,1].imshow(rle2mask(masks_actual['rle_'+str(i+1)].iloc[0]))\n",
    "        axs[i,1].set_title(\"Actual mask for Class '{}'\".format(i+1) )\n",
    "        axs[i,2].imshow(masks_predicted[0,:,:,i])\n",
    "        axs[i,2].set_title(\"Predicted mask for Class '{}'\".format(i+1))\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 446
    },
    "id": "ORYvFHwKQzgE",
    "outputId": "3d701dab-e5ef-44bc-8ae9-4d806cf5574f"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_model_prediction('0002cc93b.jpg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 446
    },
    "id": "07MuvASpQ8Oj",
    "outputId": "fa4f5e46-fe59-4a5e-d427-635918a947d7",
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "visualize_model_prediction('4d38c353e.jpg')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.7 Preparing Data for submission"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "id": "F5AK0b7aqRbk"
   },
   "outputs": [],
   "source": [
    "class PredictDataGenerator(tf.keras.utils.Sequence):\n",
    "    def __init__(self,dataframe, list_idcs, batch_size=32, ):\n",
    "        self.batch_size = batch_size\n",
    "        self.df = dataframe\n",
    "        self.list_idcs = list_idcs\n",
    "        self.indices = self.df.index.tolist()\n",
    "        self.rem = len(self.list_idcs) % (self.batch_size)\n",
    "        self.on_epoch_end()\n",
    "\n",
    "    def __len__(self):\n",
    "         return len(self.list_idcs) // (self.batch_size)\n",
    "#         if (self.rem) == 0:\n",
    "#             return len(self.list_idcs) // (self.batch_size)\n",
    "#         else:\n",
    "#             return (len(self.list_idcs) // (self.batch_size) )+1\n",
    "\n",
    "    def __getitem__(self, index):\n",
    "        index = self.indices[index * self.batch_size:(index + 1) * self.batch_size]\n",
    "#         if ((index + 1) * self.batch_size) < len(self.list_idcs):\n",
    "#             index = self.indices[index * self.batch_size:(index + 1) * self.batch_size]\n",
    "#         else:\n",
    "#             index = self.indices[index * self.batch_size: (index * self.batch_size)+ self.rem]\n",
    "        batch = [self.list_idcs[k] for k in index]\n",
    "        \n",
    "        X = self.__get_data(batch)\n",
    "         \n",
    "        return X\n",
    "    def on_epoch_end(self):\n",
    "        self.index = np.arange(len(self.indices))\n",
    "\n",
    "    def __get_data(self, batch):\n",
    "        X = np.empty((self.batch_size,256,1600,3),dtype=np.float32) # image place-holders\n",
    "              \n",
    "        for i, id in enumerate(batch):\n",
    "            img = Image.open(path + 'data/test_images/' + str(self.df['ImageId'].loc[id]))\n",
    "            X[i,] = img#input image\n",
    "\n",
    "        return X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 221
    },
    "id": "w1ue3zm5qidV",
    "outputId": "1246e9fb-1bcc-49f6-ae09-c1e5e99440d1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5506\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ImageId</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ede66fecf.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ec6434bac.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>f0bd6beec.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>edfba47f3.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>f27c44348.jpg</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         ImageId\n",
       "0  ede66fecf.jpg\n",
       "1  ec6434bac.jpg\n",
       "2  f0bd6beec.jpg\n",
       "3  edfba47f3.jpg\n",
       "4  f27c44348.jpg"
      ]
     },
     "execution_count": 23,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_img_IDs = list(os.listdir(path + 'data/test_images/'))\n",
    "test_imgsIds_df = pd.DataFrame({'ImageId': test_img_IDs})\n",
    "print(len(test_imgsIds_df))\n",
    "test_imgsIds_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "id": "4ABHK6oPqzlu"
   },
   "outputs": [],
   "source": [
    "SubmissionDf = pd.DataFrame(columns = ['ImageId','EncodedPixels','ClassId'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "y3aI_xFzqujN",
    "outputId": "94fb2ec9-a9e7-4510-c780-be1536c273c2"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "320it [00:03, 84.05it/s]\n",
      "320it [00:03, 83.25it/s]\n",
      "320it [00:03, 82.69it/s]\n",
      "320it [00:03, 84.22it/s]\n",
      "320it [00:03, 81.79it/s]\n",
      "320it [00:03, 80.39it/s]\n",
      "320it [00:03, 81.80it/s]\n",
      "320it [00:04, 80.00it/s]\n",
      "320it [00:03, 80.65it/s]\n",
      "320it [00:03, 80.12it/s]\n",
      "320it [00:04, 79.17it/s]\n",
      "320it [00:04, 78.81it/s]\n",
      "320it [00:04, 77.58it/s]\n",
      "320it [00:04, 77.95it/s]\n",
      "320it [00:04, 77.80it/s]\n",
      "320it [00:04, 76.83it/s]\n",
      "320it [00:04, 73.71it/s]\n",
      "66it [00:00, 77.02it/s]\n"
     ]
    }
   ],
   "source": [
    "for i in range(0,len(test_imgsIds_df),320):\n",
    "    batch_idcs =  list(range(i, min(test_imgsIds_df.shape[0], i + 320)))#.iloc[batch_idcs]\n",
    "    if len(batch_idcs)== 320:        \n",
    "        test_subbatch = PredictDataGenerator(dataframe = test_imgsIds_df,\n",
    "                                             list_idcs = batch_idcs)\n",
    "    else:\n",
    "        test_subbatch = PredictDataGenerator(dataframe = test_imgsIds_df,\n",
    "                                             list_idcs = batch_idcs,\n",
    "                                             batch_size= len(batch_idcs))\n",
    "    #print(len(test_subbatch))\n",
    "    subbatch_pred_masks = model.predict(test_subbatch)\n",
    "    #print(len(subbatch_pred_masks))\n",
    "    #break\n",
    "    for j, idx in tqdm(enumerate(batch_idcs)):\n",
    "        filename = test_imgsIds_df['ImageId'].iloc[idx]\n",
    "        rle1 = mask2rle(subbatch_pred_masks[j,:,:,0].round().astype(int))\n",
    "        rle2 = mask2rle(subbatch_pred_masks[j,:,:,1].round().astype(int))\n",
    "        rle3 = mask2rle(subbatch_pred_masks[j,:,:,2].round().astype(int))\n",
    "        rle4 = mask2rle(subbatch_pred_masks[j,:,:,3].round().astype(int))\n",
    "        df = pd.DataFrame({'ImageId':[filename]*4,\n",
    "                      'EncodedPixels': [rle1,rle2,rle3,rle4],\n",
    "                      'ClassId':['1', '2', '3', '4']})\n",
    "        SubmissionDf = SubmissionDf.append(df,ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "id": "N-eL3mWjrJha",
    "outputId": "cc667ee3-abf1-42a4-d993-5e8b6830fed5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(22024, 4)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ImageId</th>\n",
       "      <th>EncodedPixels</th>\n",
       "      <th>ClassId</th>\n",
       "      <th>ImageId_ClassId</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>20256</th>\n",
       "      <td>0000f269f.jpg</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>0000f269f.jpg_1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20257</th>\n",
       "      <td>0000f269f.jpg</td>\n",
       "      <td></td>\n",
       "      <td>2</td>\n",
       "      <td>0000f269f.jpg_2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20258</th>\n",
       "      <td>0000f269f.jpg</td>\n",
       "      <td>145858 12 145890 27 146082 93 146227 32 146329...</td>\n",
       "      <td>3</td>\n",
       "      <td>0000f269f.jpg_3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20259</th>\n",
       "      <td>0000f269f.jpg</td>\n",
       "      <td></td>\n",
       "      <td>4</td>\n",
       "      <td>0000f269f.jpg_4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20244</th>\n",
       "      <td>000ccc2ac.jpg</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>000ccc2ac.jpg_1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20245</th>\n",
       "      <td>000ccc2ac.jpg</td>\n",
       "      <td></td>\n",
       "      <td>2</td>\n",
       "      <td>000ccc2ac.jpg_2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20246</th>\n",
       "      <td>000ccc2ac.jpg</td>\n",
       "      <td></td>\n",
       "      <td>3</td>\n",
       "      <td>000ccc2ac.jpg_3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20247</th>\n",
       "      <td>000ccc2ac.jpg</td>\n",
       "      <td></td>\n",
       "      <td>4</td>\n",
       "      <td>000ccc2ac.jpg_4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20252</th>\n",
       "      <td>002451917.jpg</td>\n",
       "      <td></td>\n",
       "      <td>1</td>\n",
       "      <td>002451917.jpg_1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20253</th>\n",
       "      <td>002451917.jpg</td>\n",
       "      <td></td>\n",
       "      <td>2</td>\n",
       "      <td>002451917.jpg_2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             ImageId  ...  ImageId_ClassId\n",
       "20256  0000f269f.jpg  ...  0000f269f.jpg_1\n",
       "20257  0000f269f.jpg  ...  0000f269f.jpg_2\n",
       "20258  0000f269f.jpg  ...  0000f269f.jpg_3\n",
       "20259  0000f269f.jpg  ...  0000f269f.jpg_4\n",
       "20244  000ccc2ac.jpg  ...  000ccc2ac.jpg_1\n",
       "20245  000ccc2ac.jpg  ...  000ccc2ac.jpg_2\n",
       "20246  000ccc2ac.jpg  ...  000ccc2ac.jpg_3\n",
       "20247  000ccc2ac.jpg  ...  000ccc2ac.jpg_4\n",
       "20252  002451917.jpg  ...  002451917.jpg_1\n",
       "20253  002451917.jpg  ...  002451917.jpg_2\n",
       "\n",
       "[10 rows x 4 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "SubmissionDf.sort_values(by=['ImageId', 'ClassId'], inplace=True)\n",
    "SubmissionDf['ImageId_ClassId'] = SubmissionDf['ImageId'] + '_' + SubmissionDf['ClassId']\n",
    "print(SubmissionDf.shape)\n",
    "SubmissionDf.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "id": "4m4rVqpPrQF8"
   },
   "outputs": [],
   "source": [
    "SubmissionDf.to_csv(path+'submission-linknet.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.8 Kaggle Submission Score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/jpeg": "/9j/4AAQSkZJRgABAQEAYABgAAD/4RDuRXhpZgAATU0AKgAAAAgABAE7AAIAAAAMAAAISodpAAQAAAABAAAIVpydAAEAAAAYAAAQzuocAAcAAAgMAAAAPgAAAAAc6gAAAAgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAEFLSElMIFBFTlRBAAAFkAMAAgAAABQAABCkkAQAAgAAABQAABC4kpEAAgAAAAM5NAAAkpIAAgAAAAM5NAAA6hwABwAACAwAAAiYAAAAABzqAAAACAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMjAyMTowMzowNiAxNTo0ODoxNAAyMDIxOjAzOjA2IDE1OjQ4OjE0AAAAQQBLAEgASQBMACAAUABFAE4AVABBAAAA/+ELHmh0dHA6Ly9ucy5hZG9iZS5jb20veGFwLzEuMC8APD94cGFja2V0IGJlZ2luPSfvu78nIGlkPSdXNU0wTXBDZWhpSHpyZVN6TlRjemtjOWQnPz4NCjx4OnhtcG1ldGEgeG1sbnM6eD0iYWRvYmU6bnM6bWV0YS8iPjxyZGY6UkRGIHhtbG5zOnJkZj0iaHR0cDovL3d3dy53My5vcmcvMTk5OS8wMi8yMi1yZGYtc3ludGF4LW5zIyI+PHJkZjpEZXNjcmlwdGlvbiByZGY6YWJvdXQ9InV1aWQ6ZmFmNWJkZDUtYmEzZC0xMWRhLWFkMzEtZDMzZDc1MTgyZjFiIiB4bWxuczpkYz0iaHR0cDovL3B1cmwub3JnL2RjL2VsZW1lbnRzLzEuMS8iLz48cmRmOkRlc2NyaXB0aW9uIHJkZjphYm91dD0idXVpZDpmYWY1YmRkNS1iYTNkLTExZGEtYWQzMS1kMzNkNzUxODJmMWIiIHhtbG5zOnhtcD0iaHR0cDovL25zLmFkb2JlLmNvbS94YXAvMS4wLyI+PHhtcDpDcmVhdGVEYXRlPjIwMjEtMDMtMDZUMTU6NDg6MTQuOTM3PC94bXA6Q3JlYXRlRGF0ZT48L3JkZjpEZXNjcmlwdGlvbj48cmRmOkRlc2NyaXB0aW9uIHJkZjphYm91dD0idXVpZDpmYWY1YmRkNS1iYTNkLTExZGEtYWQzMS1kMzNkNzUxODJmMWIiIHhtbG5zOmRjPSJodHRwOi8vcHVybC5vcmcvZGMvZWxlbWVudHMvMS4xLyI+PGRjOmNyZWF0b3I+PHJkZjpTZXEgeG1sbnM6cmRmPSJodHRwOi8vd3d3LnczLm9yZy8xOTk5LzAyLzIyLXJkZi1zeW50YXgtbnMjIj48cmRmOmxpPkFLSElMIFBFTlRBPC9yZGY6bGk+PC9yZGY6U2VxPg0KCQkJPC9kYzpjcmVhdG9yPjwvcmRmOkRlc2NyaXB0aW9uPjwvcmRmOlJERj48L3g6eG1wbWV0YT4NCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAKICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIAogICAgICAgICAgICAgICAgICAgICAgICAgICAgPD94cGFja2V0IGVuZD0ndyc/Pv/bAEMABwUFBgUEBwYFBggHBwgKEQsKCQkKFQ8QDBEYFRoZGBUYFxseJyEbHSUdFxgiLiIlKCkrLCsaIC8zLyoyJyorKv/bAEMBBwgICgkKFAsLFCocGBwqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKv/AABEIAHIDuwMBIgACEQEDEQH/xAAfAAABBQEBAQEBAQAAAAAAAAAAAQIDBAUGBwgJCgv/xAC1EAACAQMDAgQDBQUEBAAAAX0BAgMABBEFEiExQQYTUWEHInEUMoGRoQgjQrHBFVLR8CQzYnKCCQoWFxgZGiUmJygpKjQ1Njc4OTpDREVGR0hJSlNUVVZXWFlaY2RlZmdoaWpzdHV2d3h5eoOEhYaHiImKkpOUlZaXmJmaoqOkpaanqKmqsrO0tba3uLm6wsPExcbHyMnK0tPU1dbX2Nna4eLj5OXm5+jp6vHy8/T19vf4+fr/xAAfAQADAQEBAQEBAQEBAAAAAAAAAQIDBAUGBwgJCgv/xAC1EQACAQIEBAMEBwUEBAABAncAAQIDEQQFITEGEkFRB2FxEyIygQgUQpGhscEJIzNS8BVictEKFiQ04SXxFxgZGiYnKCkqNTY3ODk6Q0RFRkdISUpTVFVWV1hZWmNkZWZnaGlqc3R1dnd4eXqCg4SFhoeIiYqSk5SVlpeYmZqio6Slpqeoqaqys7S1tre4ubrCw8TFxsfIycrS09TV1tfY2dri4+Tl5ufo6ery8/T19vf4+fr/2gAMAwEAAhEDEQA/APo23/1Z/wB9/wD0I1JUdv8A6s/77/8AoRqSgAoooyM47+lABRTfMTzPL3rvxu255x64pxOBk0AFFFFABRRRQAUUUUAFFFFABRRQSFUljgDkk9qACimRTxXEYkglSVG6MjAg/iKfQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFJvXft3DdjOM80ALRRRQAUUUUAFFFFABRRRQAUUA56VVutTsLGRUvb23t3f7qzSqhb6ZNGw0m3ZFqikVgyhlIIIyCD1paBBRRRQAUUUUAFFFFABRSblLFQw3AZIzzSkgdTj60AFFIGB6EH6UKytnaQcHBwelAC0Um5SxUEbhyRnkUbl3BcjcRkDPNAC0UhZR1IH40tABRRRQAUUUUAFFFFABRRRQAUUUgZWJCsCVOCAelAC0UhYDqQPrQCCMggigBaKRWVlypBHqDQGB6EH8aAFopNylioI3DkjPIo3KGClhuIyBnk0ALRRRQAUUUUABOBk1W/tGy/5/Lf/AL+r/jRqP/ILuv8Ari//AKCar2FhZvptsz2kDM0KEkxgknAoAsf2jZf8/lv/AN/V/wAasg5GRWdf2Fmmm3LJaQKywuQRGAQcGrGnf8gu1/64p/6CKALNFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQBHcf6sf76f+hCpKjuP9WP8AfT/0IVJQBHb/AOrP++//AKEakqO3/wBWf99//QjUlABXPt/yUqP/ALBLf+jlroKy9S8O6Zq12l1ewymdE8tXiuJIjtznHyMM81pTkk9fMyqxlKNo73X4NMxfEmqz6XqWpz2iQrNBo/nJKYgWyJCME9SPbpVfV7zVIItT03UruK7juNHlulKQ+X5TDClRycj5u/PFdE/hvS5YHhlt2kR7b7K2+ZyWiznaSTnr36+9TXWjWF7M0tzBvdrdrVjvYZiYgleD7DnrWiqQSSt/XQylTqNt37fpf8LnKQeJL+Twrqmo27i1/s61VIbWWL94GCA+Y4PY9gO3v0tza3quiXt2moTx36jTJL5FSHy9jIQNgwTkHPfnitu58O6Xd+Z59rnzbf7LJtkZd8f904POOxPIqy+mWct4LqSAPMIGt8sSR5ZIJXHQ8gU5VKbe39f8OKNKqlbm2t+eunpocvbahr1xCySyzPHcWru07WPlC3cLkbTn5lPI556HNWdA1K8sPBukST219q8k9uj7rdE3INoIDbmGfr1Natj4c0vTpvNtoH3CMxoJZnkEaHqqhiQo9hirWnaba6TZi1sEaOBTlUaRnC+w3E4HsOBSlUg00l26eo40ql02+/Xvbv6MwPNvR48eU3Uq240wTm0ZBwNxG3r1zzn8OlV9D13Wr+SwvHWWa3veZIfsZRIEKkqVkz82DgHPXPGK6aXTLSbU4dQeNvtUKFFkWRlyp7EA4YexzVW28NaVaXi3MFuweNmaNDM7RxFupVCdqk5PQDrR7SHLZrW3+f8AwNdxunU5rp6X7+n/AAdNtTA0rWdaaHRb+8vYZodRnaB7cQBdo+bDBs5z8v05/GprLW9Th1jy9ama3d3lC2jWn7uQKGK+VMDgtgAkN74AreXRtNt7S1iEISCxcyw5kbEZ55yTz1PWobTw5pNtdx3lrA25CzwqZ3aOMt1KISVXOTyB3pupTd9O/RfIXs6itr2vq/n/AFsY1jrOriDR9Uu7qGa21aVIzaJDt8kOCVKtnJI4zn36VnX9zrGqfDrVdRutSUI0c6LDHbgfKsjLy2c5IFdXa+GtKsr1bq3tmWSNmaNTK7JEW+8UQnapOT0A61N/Ymn/ANjSaV9n/wBClDh4t7c7iWbnOeST3p+1pp3S69lsSqNVq0n07vfv/wADocreTN4b1ZJwqTtZ6FPIAkQjViJUP3RwKtXN9r+ntFbNfwXk99ayywsluF8mVFDAAZ5Q8jnnpzXSSaXZzXn2mWBXl8hrfLEkGNiCVI6HOBVfTvD2m6VP51nA4kCeWrSTPJ5af3V3E7R7DHSl7WLWq1/4caozUtHpp18kvnszlpfHNx5k95FsNg9ntt1C/Mbry1k2/k+Meort7UTLZwi6cPOI1EjAYBbHJx9azl8L6MlnDaLYqIILn7VGm9vllzndnPr26VrVFSVNq0FYujCrF/vHf+v6/EKKKKxOgKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigArzv+zrXSv9P1TTre9ha981NfsZgbhd0vy7887RkKQrMMdgK9ErFXwhoq3guFtHGJfPEPnyeSJM53+Vu2ZzznHXmhb3B7WM278Q30Og69dI6CWxv/ACISU4C5j6+v3jU0d7rWp3uoT2N7a20Fjdm2W2miyJQoUsXfOVJycYHHGc1du/CWiX93Pc3VkXkuCrSgTOquy4wxUNt3DA+bGfen3XhjSb3UDeXFszSsyvIomdY5WX7peMHa5GBywPQelJaWv/W2v5/eLXX+u5nx3utaneX89je2ttBY3Zt1tposiUKFLF3zlScnGBxxnNUrnX9anvNSk0yOR1sbkwR2iWZdZ9oBbdJngnJxjgcE5rduvDGk3uoG8uLZmlZleRRM6xysv3S8YO1yMDlgeg9KLvwzpV7fNdXFu5eQhpUWeRY5iOheMMFfoPvA9BTWlr/1t/Vhv+v6/Uwr3X9an1LVRpaSf8S2QRRWy2hkFw2xXO98/LndgYxjqc9KTVdQ1TU7fxF5FxDZ22mwGI28kW4ykwh2LNn5RhsDHoSc9K37/wAM6VqV21xd27s8gAlVJ5ESYDoJEVgr+nzA8cUl94X0jUrp7i7tWZ5Y/LlVJnRJVAwA6qQGxnjIOO1TJXjYpOzuc5b6zqT/AGPTNPkkt1tdLtpi8doZzIzqQAeeFGz6nPUYq9barreralDZK0ekzQ2EV1cpJD5haRyw2ckYUbDnvyORWrdeGdLvEgWSGWPyIhCjQXMkTeWP4CyMCy+xzS3HhnSbk2261Mf2WPyovIleLEf9w7CNy8fdORVyd23/AF1M4q0UvJfoUvArMfA1g2FLbZDhTkE+Y3Q1kaLYR6lottfPpyahfX1xvvLmZI3EQEmGQhzkDaMAKPyrsNP0+10qwisrCIQ28QISMEkKCc9/c1l3fhO0nvJLm1vL/TmmbdMtlcGNZG9SMHn6YrOS1udVCahFxbtsQ+FgLXUNY022YtY2lwot8tkJuXLID6A10dVdN0210myW1sYvLiXnrksT1JPc1aqkrJIzqyU5uS/r/h9wooopmYUUUUAFI5IRivJA4paKT2A84tbO2g8KaF4gghQa1cXsBmugP3srSShZEY9SMFhtPA2j0rvNUs9PvbB01i2t7i1T94y3MYdBgdcHjiqkHhfSLfU/t8VqwmEjSqpmcxo7dXWMnYrHJ5AB5NTPoenyWN5ZyQM1veyNJcIZX+dmxnvwDjoOKfS39dB9b/11OU03Qgng3W7/AEmCDSp9YQvAkYECxRAYjBIA2kjJJ7F/amaeg0zxVYjSNAOixy2E/wBogPlqZimwqxCEg4JxuJydxrupraG4tXtp4kkgkQo8bLlWUjBBHpiqOl+H9O0eR5LGGQSOoQvLO8rBR0UFySF9hxQ/L+tBepxHhyDyLXw7rV9pdqJr+RA97FKRdNJIpOZDj51Jz8ufl49OGa9Zy2djrd6dI+1arHdPPa6wrRnywGGxA2d6kcJsAwT7Emu2tvC+kWmoLeQWpWVGZ41MztHEzdWSMnapOTyAOpok8MaRLqhv3tSZjIJWUSuI2cdHMedhYYHJGeKd9b/10/roH9fmY/jHR9Mv4YYG0qxm1fU2FvHPLbo7xrjLvkjPyqCR749a6iztIbCxgtLZdkMEaxxr6KBgD9Ka1hbNqSag0WbpIjCshY/KhIJAHQZIHPXgVYpdACiiigAooooAKKKKACiiigDN8SXE9p4X1S4syRcRWkrxkDJDBSRXNwaXYaJqnhebRII4ZLwtFcPGObmPyWcs5/jO5VO45PPvXasoZSrAEEYII61lab4Y0nSboXFjbMkiqUj3zPIIlJyVQMSEHsuKFo7/ANdQeqsHiS30dtJlu9esLa9gtEaRVnhWTB9FBHU8CuYn0RNE+HJtDAtvdahdRySWtugVJJXkU+RgYATACE+gJ5rrV0LTl0+3sfs5NtbyiaONpGOHDbgTk5ODzg5HT0qe/wBPtdUsntL+FZoXwSpJGCDkEEcgg8gjkUhnH6VaNa3niGxjtLXRJprFHis4XH2Zch183cAMEng/KOFHWpvD+nQ+HdYsrCTRtJtp7q0YRXlgo3koF3ByVBIOQc9yOR0robHw9pmnrcCG3aQ3S7J3uZXneRcY2lnJJXk8ZxzTdN8OaZpFwZ7KBxLs8tWlneUonXau9jtXpwMDin1/rzF/X5HDRWk2mQ6RNHo+zVhqUSXGrh4z9rV5MPh1JZwyknaQAoB9Bl3iGze1s/EF7JpP2rUkmee01dWjPkKACihs71K4xsA5J9zXa2vhjSLPUfttvalZQzOimV2jjZvvMkZO1ScnkAdT60TeGNIuNTN/LakzM6yOolcRu6/dZowdjMMDkgngelC/r8P8hvf+vP8AzNOIs0KGQYcqCw96fRRQStgooooGVtR/5Bd1/wBcX/8AQTVewv7NNNtle7gVlhQEGQAg4FaJGRg1W/s6y/587f8A79L/AIUAV7+/s3025VLuBmaFwAJASTg1Y07/AJBdr/1xT/0EUf2dZf8APnb/APfpf8KsgYGBQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFAEdx/qx/vp/6EKkqO4/1Y/wB9P/QhUlAEdv8A6s/77/8AoRqSo7f/AFZ/33/9CNSUAFYGozST+NtJsNxWGO3nvGAPDspRFz6geYxx649K36y73S5JfEGnapbFA9uskMwYkbonAPHuGVT9M0Lf+v63DoZtt4pnn+wQG2jW8lupobqPJxCkOd7juc/Jj/fFV9I8ZT6lcWEphtmtNQfbHHC7NLACCVZ8jBzgAgdCRya2bbw7ZWviK71mPzDcXaBHRmBRegJUY4LbVz67RUdh4bTTriLyNSv/ALJASYLEyKIo8544UMwGeAzED04FC8wfkZUXifWH8PPq7WFt5JlMUaoXdkAlKGR8A/KAM4GTx1HZTrn22DSJZXs7zzNVECzWNy+z/VsdxAPB4IKNkd614/D8UGhrplpeXluqSGRJ4pAJFYuX9MEZPQggjqDUEHhO0ieOWW5uri4W9F888hQNLIEKDcFUKBtPQAdKUfPy/T/g/wBbH/B/JlKPxTftbwaq9pbjSJ7oWy4kbzlBk8tZDxjBbHy9QD17VE3irVVSa9NlafYLfUzYOPMbzWHmiMOOMDBI4PXnpWjH4StI7lD9ru2so7j7THp7MvkpLu3bh8u7AbkLuwD2qVvDFm+mz2Rln8ue9+2sdw3B/NEmBx93I+uO9Nef9bf8Een9fP8A4BL4jtprnQ7kQ3s9psjd2a3ba7AKeN3VeccjB461kPPLd+FvDlmZZGm1A23mPuJYqqiVyT7hCD/vV09zAt1ay28hISVCjFeoBGKzLbQxa6hp0iSlrbT7JraFG5Yk7RuPb7qY/E0R3fy/C/8AwBPp8/xsULOG4t/iBPDPf3N1HNp5lMcr/u0Pm4AVBwMDjPU9zU3gp2Ph0wMdy2t1cW0Z/wBhJWVR+AAH4VpvpkZ1htTR2Fz9m+zAHBQDduzjrnPvSaLpaaNpEFkjmQxgl5CMGR2JZmx7sSfxojorf1u/0F1f9dP8y9RRRQMKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAjuP8AVj/fT/0IVJUdx/qx/vp/6EKkoAjt/wDVn/ff/wBCNSVHb/6s/wC+/wD6EakoAKgnvILZgsz7SRk4UnA9TjoPc1PWZqdyqSi3MM4EyYlmitnk+T+6CoPPJ+lAGhJKkShpGwCwUH3JwKfVPUIvOs40RWI82LhQQQN4/EcVSaNoZpEkimNmtznYis3ymMdhyV3Z6d6ANmiseF5YJreWSO4MBEyqNjMwBZSmR1HAPXp3xTbC3lMCvcJOrrZRBeu4Nhs4B43dKOlx2Nqis7Sdw8xTGwUBf3hjeMOec/K3Q+p759qqr9obU1MUbxnzJA4MT9MNglycEZAIAHFDEbdFYGJB5cdqlwlz9lkD78jdJ8vc8E9eRx71fsNpunNtHNHB5YBEqMuXz2Dc5x1P0oAvJKkm7Yc7W2t7GnVilCtzOFhuBcNdBo32Nt2/Lk56AYzkHr+VT2qyHUmt23eValnUk/e3/dH4DcPyoAt3d9bWCxNdyrEJZVhQn+J2OAKkFxE109sG/eoiuy4PCkkA/wDjp/KsDXrC+1nUjbW8Uf2eC2b559ygySZAZSByVAP/AH1VKUandXcV5c2s4gFrAL63SJg8pV5AwU9wD8xUcsuPXB2VNNLUxlUab0/rr/Xkzq4bmK4MohfcYnMb8HhgAcfqKjg1C3uJVijdvMdDIEeNlO0NtJwQO9c62nSxzzXsFvOtydXiKuA2fJJQNx/d27s9vyqxodvcx6jbvcQyqBazqWdSMEz5A59ufpR7ONr3/q1w9o+Zq39c1jSk17TobyS2lnZZI2CSMYX2IxAIBfG0dR371o1zQ0i8v77WoJZhBYXVwA6G3O+VfKQEq5OAOMfdPQ/hVS3vj4izLLMk/wBtLBlsZWzDnhfN3+WFK8YxkHsTzR7OLtZ9P8hOpJXbXX/M64kKpLHAAySe1JFLHPCksLrJHIoZHU5DA8gg+lcra6VILfT828++4SeO737vmUglQ2enIGM9OgrOa3ki8OadawWNxDJFYbQWtJ5D5+MMAqlQrZGQ59eOM0/ZLa/9a/5A6rTWn9af5/gd7RUNmZDYwGbd5hjXfuHOcc5qasWrOxsndXCiiikMKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKAI7j/Vj/AH0/9CFSVHcf6sf76f8AoQqSgCO3/wBWf99//QjUlR2/+rP++/8A6EakoAKKKKACiiigAooooAKKKKAIYbO3t33QxKhxtBHYeg9B7CpqKKACmRwxxbvLQKXbcxA+8fU0+igAooooAKKbLLHDGZJnWNF6s5wB+NCOksavGyujDKspyCKAHUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFAEdx/qx/vp/wChCpKjuP8AVj/fT/0IVJQBHb/6s/77/wDoRqSo7f8A1Z/33/8AQjUlABXJ61Gt14wWC4XU5oVsQ6x2N3JDhvMIydjrnj1zXWVVGnxDV21Hc/nNAICuRt2hi2enXJpdV/XRh0f9dTM+3XNtqlrpVjGqQrZi4eS7keSRVDAEZJJZuepP50y01y+kisL66t4EsdQdFjVWPmRbwShbsc8AgYxnvWsdOhOrjUSz+cIDb7cjbtLbumOuR61RtfDcNrPAftl1La2rbrazcp5cJwQMYUMcAnAZiB+Aw15/1r/lYXf+un+ZnQ+ItVnWylS2s/KvbqS0jBdtyMpk+c+oxGfl+nPps6PfT3tvOt4kaXFtO0EnlE7WIwQRnkZBBx2qOHQLWCCyiSSYrZ3T3UeWGSzb8g8dP3h/SrdpYx2T3LRM5NzMZn3EcMQBx7fKKF/X4f8ABG+lv63/AOAcp4c1O9tre4GpXMk9ve3d2ttNI2TDIk0gEWfQquV9wR6CrukapqV1p+n2lgIXmj063uLie6ZjuLggKMc5O1iSenHBzWkPDlj/AGFPpMnmSW88ksjFm+ZWeQyZBAGCGbg9RgVH/wAI3HFFbLYX95ZSQW6WplhMZaWNfuhtyEZGScgA8mn3/ruK2t/X80VItXvtQ1TRpLUxwW1zDMZ4XBYhkKhhn2OQDUVt4te4uoHX7M1tPceSsKFjMoLbQ54xjoSOwPU4xWsNCt4xp/2WWa2+wZWMxsDvU43K24HIOBk8H3psGgi1ulaDUb1LVJDItkroIwxOeu3fjJJ27sdsY4pLcfQx9M1fVLbSWvbhIpbJL6aJ97sZtv2hkDg9MDI+X0HXtVy5169jgvdRhgtzp1jI6SBmPmSBDh2B6DGGwDnOOozUkPhW3gIjW+vWtPtDXLWjMhjaQyGTJO3djcc7QcccjrmW48OQ3E83+mXUdpcSeZPZIU8qVuM5yu4A4GQGAP4nJ1/r+u/9bHQhuda1AtqE9hbQPa6axWRZGIechA7bSOFwCBznJz061nWHiFpPFGp6fbSGa4uJ4nt0mY7IovIjZm/UkKOSc9skbF54diu57hlvbu3hu/8Aj6t4WQJPwF5JUsuVAB2kcCiXw1ZStdOHmjkuJ451dGAMLoiopTjjhehznJHQ4oW/9fMYeKAG8PyBgCDNCCMdf3qVHBLIPHF1bB2ECabCyxA/KpMkoJA6ZwB+Qq7rGnyanpUlrDOsMjFGWR494BVgwyoIz09RVOfw/PcX66gNZurO8NskEzWcUQSQKWbO2RJCOWPel9q/9bMPstf1uinqWtyaZqepG3t1kljNlGN8zBW82Qp05C4z1A575qrr2t31tp2oW988cE1pLZzfaLZmUGKScAg55B+RgeeQfwrYm8NW1y0j3NzcyyS/ZjJISgLGBtynAUDk9cD6Yp9/4ftdQuLiaaSZWuBAGCMAB5MhkXHHcnn29KrS39eQRt1KU3iSf+wf7WtbeJ4JriKK2R2IZ1eRU3N/d6njGRjnnIDLnxJdaX9tg1KK3a4g8gxNGzLG4mdlXdnJGCpz14/Kr83hy0mFwnmTxxXFxHdNEjDasiOG3DI43FRkd+vBJNSXehWt7c3M8rSiS4jiQlWA8sxszIy8cMC2e44HFGlv68v+CHQwr3xFfyaHrC20tt9qtLYTR3UCt5ZB3cYPRht9SOQfat2TWIrKaC1vluWuZFXLW9jNJHknH31Uqv4nimDQhLYXlrqGoXl8LuPy3eZlUquCPlCKqg89cZPfoK0baJoLaOKSeS4ZFwZZQoZ/c7QBn6AUE9Tm/wC273T5vENxfPDNb2dwkdvGMrgtHHtBbnAy/Jwep+lMfxZcW9nqDOtrdSW1sbiN7feEODgo2eh5HOeeeBiteXQLaa4v3kln8u/C+dAGAUOoAEinG4NhV744zjPNRv4e+02N3a6hql/eLdR+UWlaNdi/7IVAufcgmkVLXYrz61qdjJeRXVlDcSpZtdW6WpYl8HBQjBJPI5A5z0psXiOT+y57gSWV9Ks0cMa2rsv7yRgoV1OSmCw/DnA6VqXWkxXd0bgzTxSm3a3DQvtKgsGyD1zlR7e1VF8NW7pdG9urm8nuljVriUorqIyWTbsVQMMSc4zmhef9a/5C6/12/wA7lS/8QX+kW94l9Bby3EMAuImiLKki7wpBByQRkc5Oc9qZrMepz3uh29y9rtlun82NN+1sRuwB5GQMd+4BxV1/DMNxa3Ud/fXl5LdRrE88pQOqKchVCqFHPPTJ/LGjc2Md1dWlxIzh7SRpECkYJKFefwY0Lcl31Oc0vVb4o9nbFZru41G9CSXLMVijjlI6dTjKgAEdfarsXiKRLhIr6GOIx3T2ly6MSqsI/MVgT2K+vQnHPWp/+Ebt0ixbXVzbzLcy3KXEZUujSMS4wVKleehB7dxmkn8MWd1ok2m3MtxItxIJZrguBLI+QdxIGB0AwABjgCktreS/T/gmkrOTfm/1/wCAVNM8Q3GqLZ289mkU9xLIs8RYny4gm4N9SHiH/Aj6Vc8MosVjdxpwkd9OqjOcAOcCrCaPBBrE+qwBjcyQCEIz4jGO444JwoJ54UccUzQrO7s7GUagsKTzXEszJBIZFXcxIG4qpP5Cj7Xy/VEvb5/5mnRRRTAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigAooooAKKKKACiiigCO4/wBWP99P/QhUlR3H+rH++n/oQqSgCO3/ANWf99//AEI1JUMAfy2wVA3v1H+0fepMP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADqKbh/wC8v/fP/wBejD/3l/75/wDr0AOopuH/ALy/98//AF6MP/eX/vn/AOvQA6im4f8AvL/3z/8AXow/95f++f8A69ADbj/Vj/fT/wBCFSVDOH8tclSN6dB/tD3qagCO3/1Z/wB9/wD0I1JRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAFFFFABRRRQAUUUUAR3H+rH++n/oQqSiigD//2Q==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Kaggle Score\n",
    "from IPython.display import Image\n",
    "Image(filename='linknet.JPG') "
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Linknet - Untitled21.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}