Choose_U++net.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 81
    },
    "colab_type": "code",
    "id": "jVLmkAtHQg1z",
    "outputId": "f2aa91a1-d400-477e-a9e9-6877261a0b42"
   },
   "outputs": [],
   "source": [
    "# Import Libraries\n",
    "import os\n",
    "import numpy as np\n",
    "from utils import *\n",
    "import cv2\n",
    "from random import randint\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "path=os.getenv('HOME') + '/repo/Portrait/data/Celeb/data_100/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "WgKNFVdtZrkD"
   },
   "outputs": [],
   "source": [
    "# 이미지와 마스크를 load.\n",
    "X = np.load(os.path.join(path,\"Images.npy\"))\n",
    "Y = np.load(os.path.join(path,\"Masks.npy\"))\n",
    "# 마스크를 binary format으로 변경.\n",
    "Y = np.where(Y > 1, 1, 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 87
    },
    "colab_type": "code",
    "id": "-x1_KfgYgLML",
    "outputId": "0d2a4c6f-8801-4a91-bd8d-897f16c428d6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "이미지의 개수: 300\n",
      "이미지의 모양: 512 x 512\n",
      "마스크의 개수: 300\n",
      "마스크의 모양: 512 x 512\n"
     ]
    }
   ],
   "source": [
    "data_info(X, Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 720x720 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_image(X, Y, 6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "3hPAtvoPiqyW"
   },
   "outputs": [],
   "source": [
    "x_train, y_train, x_valid, y_valid = train_val_split(X,Y)\n",
    "del X\n",
    "del Y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "4MMlAmPCQg2K"
   },
   "outputs": [],
   "source": [
    "# Data generators\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "\n",
    "image_datagen = ImageDataGenerator()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_crop_shape(target, refer):\n",
    "    # width, the 3rd dimension\n",
    "    cw = target[2] - refer[2]\n",
    "    assert (cw >= 0)\n",
    "    if cw % 2 != 0:\n",
    "        cw1, cw2 = int(cw/2), int(cw/2) + 1\n",
    "    else:\n",
    "        cw1, cw2 = int(cw/2), int(cw/2)\n",
    "    # height, the 2nd dimension\n",
    "    ch = target[1] - refer[1]\n",
    "    assert (ch >= 0)\n",
    "    if ch % 2 != 0:\n",
    "        ch1, ch2 = int(ch/2), int(ch/2) + 1\n",
    "    else:\n",
    "        ch1, ch2 = int(ch/2), int(ch/2)\n",
    "\n",
    "    return (ch1, ch2), (cw1, cw2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Unet 구현하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "8Yx_efEEQg2i"
   },
   "outputs": [],
   "source": [
    "from keras.models import Model, load_model\n",
    "from keras.layers import Input, UpSampling2D, BatchNormalization, Cropping2D\n",
    "from keras.layers.core import Dropout, Lambda\n",
    "from keras.layers.convolutional import Conv2D, Conv2DTranspose\n",
    "from keras.layers.pooling import MaxPooling2D\n",
    "from keras.layers.merge import concatenate\n",
    "from keras.callbacks import EarlyStopping, ModelCheckpoint\n",
    "from tensorflow.keras.backend import int_shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def upp_model(input_size=(512,512,3), filters= [0, 32, 64, 128, 256], kernel=3):\n",
    "    inputs = Input(input_size)\n",
    "    s = Lambda(lambda x: x / 255) (inputs)\n",
    "    \n",
    "    x1_0 = MaxPooling2D(pool_size=(2,2), name='x1_0_pool')(inputs)\n",
    "    x1_0 = Conv2D(filters[1], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x1_0_conv1')(s)\n",
    "    x1_0 = BatchNormalization()(x1_0)\n",
    "    x1_0 = Conv2D(filters[1], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x1_0_conv2')(x1_0)\n",
    "    x1_0 = BatchNormalization()(x1_0)\n",
    "    \n",
    "    x2_0 = MaxPooling2D(pool_size=(2,2), name='x2_0_pool')(x1_0)\n",
    "    x2_0 = Conv2D(filters[2], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x2_0_conv1')(x2_0)\n",
    "    x2_0 = BatchNormalization()(x2_0)\n",
    "    x2_0 = Conv2D(filters[2], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x2_0_conv2')(x2_0)\n",
    "    x2_0 = BatchNormalization()(x2_0)\n",
    "    \n",
    "    x3_0 = MaxPooling2D(pool_size=(2,2), name='x3_0_pool')(x2_0)\n",
    "    x3_0 = Conv2D(filters[3], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x3_0_conv1')(x3_0)\n",
    "    x3_0 = BatchNormalization()(x3_0)\n",
    "    x3_0 = Conv2D(filters[3], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x3_0_conv2')(x3_0)\n",
    "    x3_0 = BatchNormalization()(x3_0)\n",
    "    \n",
    "    x4_0 = MaxPooling2D(pool_size=(2,2), name='x4_0_pool')(x3_0)\n",
    "    x4_0 = Conv2D(filters[4], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x4_0_conv1')(x4_0)\n",
    "    x4_0 = BatchNormalization()(x4_0)\n",
    "    x4_0 = Conv2D(filters[4], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x4_0_conv2')(x4_0)\n",
    "    x4_0 = BatchNormalization()(x4_0)\n",
    "    \n",
    "    u4_0 = Conv2DTranspose(filters[3], (2, 2), strides=(2, 2), padding='valid', name='u4_0_convTran')(x4_0)\n",
    "    r3_0 = tf.image.resize(x3_0, (u4_0.shape[1], u4_0.shape[2]), name='x3_0_resize')\n",
    "    x3_1 = concatenate([r3_0, u4_0], axis = 3, name='x3_1_merge')\n",
    "    x3_1 = Conv2D(filters[3], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x3_1_conv1')(x3_1)\n",
    "    x3_1 = Conv2D(filters[3], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x3_1_conv2')(x3_1)\n",
    "    \n",
    "    # 2-2로 감\n",
    "    u3_0 = Conv2DTranspose(filters[2], (2, 2), strides=(2, 2), padding='valid', name='u3_0_convTran')(x3_0)\n",
    "    r2_0 = tf.image.resize(x2_0, (u3_0.shape[1], u3_0.shape[2]), name='a_r2_0_resize')\n",
    "    x2_1 = concatenate([r2_0,u3_0], axis = 3, name='x2_1_merge')\n",
    "    x2_1 = Conv2D(filters[2], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x2_1_conv1')(x2_1)\n",
    "    x2_1 = Conv2D(filters[2], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x2_1_conv2')(x2_1)\n",
    "    \n",
    "    # origin\n",
    "    u3_1 = Conv2DTranspose(filters[2], (2, 2), strides=(2, 2), padding='valid', name='u3_1_convTran')(x3_1)\n",
    "    r2_0 = tf.image.resize(x2_0, (u3_1.shape[1], u3_1.shape[2]), name='b_r2_0_resize')\n",
    "    r2_1 = tf.image.resize(x2_1, (u3_1.shape[1], u3_1.shape[2]), name='b_r2_1_resize')\n",
    "    x2_2 = concatenate([r2_0,r2_1, u3_1], axis = 3, name='x2_2_merge')\n",
    "    x2_2 = Conv2D(filters[2], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x2_2_conv1')(x2_2)\n",
    "    x2_2 = Conv2D(filters[2], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x2_2_conv2')(x2_2)\n",
    "    \n",
    "    # 1-3로 감\n",
    "    u2_0 = Conv2DTranspose(filters[2], (2, 2), strides=(2, 2), padding='valid', name='u2_0_convTran')(x2_0)\n",
    "    r1_0 = tf.image.resize(x1_0, (u2_0.shape[1], u2_0.shape[2]), name='a_r1_0_resize')\n",
    "    x1_1 = concatenate([r1_0,u2_0], axis = 3, name='x1_1_merge')\n",
    "    x1_1 = Conv2D(filters[1], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x1_1_conv1')(x1_1)\n",
    "    x1_1 = Conv2D(filters[1], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x1_1_conv2')(x1_1)\n",
    "    \n",
    "    # 1-3로 감\n",
    "    u2_1 = Conv2DTranspose(filters[2], (2, 2), strides=(2, 2), padding='valid', name='u2_1_convTran')(x2_1)\n",
    "    r1_0 = tf.image.resize(x1_0, (u2_1.shape[1], u2_1.shape[2]), name='b_r1_0_resize')\n",
    "    r1_1 = tf.image.resize(x1_1, (u2_1.shape[1], u2_1.shape[2]), name='b_r1_1_resize')\n",
    "    x1_2 = concatenate([r1_0, r1_1, u2_1], axis = 3, name='x1_2_merge')\n",
    "    x1_2 = Conv2D(filters[1], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x1_2_conv1')(x1_2)\n",
    "    x1_2 = Conv2D(filters[1], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x1_2_conv2')(x1_2)\n",
    "    \n",
    "    # origin\n",
    "    u2_2 = Conv2DTranspose(filters[2], (2, 2), strides=(2, 2), padding='valid', name='u2_2_convTran')(x2_2)\n",
    "    r1_0 = tf.image.resize(x1_0, (u2_2.shape[1], u2_2.shape[2]), name='c_r1_0_resize')\n",
    "    r1_1 = tf.image.resize(x1_1, (u2_2.shape[1], u2_2.shape[2]), name='c_r1_1_resize')\n",
    "    r1_2 = tf.image.resize(x1_2, (u2_2.shape[1], u2_2.shape[2]), name='c_r1_2_resize')\n",
    "    x1_3 = concatenate([r1_0, r1_1, r1_2, u2_2], axis = 3, name='x1_3_merge')\n",
    "    x1_3 = Conv2D(filters[1], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x1_3_conv1')(x1_3)\n",
    "    x1_3 = Conv2D(filters[1], kernel, activation='relu', padding='valid',kernel_initializer='he_normal', name='x1_3_conv2')(x1_3)\n",
    "    x1_3 = Conv2D(2, kernel, activation='relu', padding='same',kernel_initializer='he_normal', name='x1_3_conv3')(x1_3)\n",
    "    x1_3 = Conv2D(1, 1, activation='sigmoid')(x1_3)\n",
    "    outputs = tf.image.resize(x1_3, (inputs.shape[1], inputs.shape[2]), name='output')\n",
    "    \n",
    "    model = Model(inputs = inputs, outputs = outputs)\n",
    "    \n",
    "    model.compile(optimizer='adam', loss='binary_crossentropy')\n",
    "\n",
    "    model.summary()\n",
    "    \n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"functional_1\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            [(None, 512, 512, 3) 0                                            \n",
      "__________________________________________________________________________________________________\n",
      "lambda (Lambda)                 (None, 512, 512, 3)  0           input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "x1_0_conv1 (Conv2D)             (None, 510, 510, 32) 896         lambda[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization (BatchNorma (None, 510, 510, 32) 128         x1_0_conv1[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x1_0_conv2 (Conv2D)             (None, 508, 508, 32) 9248        batch_normalization[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_1 (BatchNor (None, 508, 508, 32) 128         x1_0_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x2_0_pool (MaxPooling2D)        (None, 254, 254, 32) 0           batch_normalization_1[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "x2_0_conv1 (Conv2D)             (None, 252, 252, 64) 18496       x2_0_pool[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_2 (BatchNor (None, 252, 252, 64) 256         x2_0_conv1[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x2_0_conv2 (Conv2D)             (None, 250, 250, 64) 36928       batch_normalization_2[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_3 (BatchNor (None, 250, 250, 64) 256         x2_0_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x3_0_pool (MaxPooling2D)        (None, 125, 125, 64) 0           batch_normalization_3[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "x3_0_conv1 (Conv2D)             (None, 123, 123, 128 73856       x3_0_pool[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_4 (BatchNor (None, 123, 123, 128 512         x3_0_conv1[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x3_0_conv2 (Conv2D)             (None, 121, 121, 128 147584      batch_normalization_4[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_5 (BatchNor (None, 121, 121, 128 512         x3_0_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x4_0_pool (MaxPooling2D)        (None, 60, 60, 128)  0           batch_normalization_5[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "x4_0_conv1 (Conv2D)             (None, 58, 58, 256)  295168      x4_0_pool[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_6 (BatchNor (None, 58, 58, 256)  1024        x4_0_conv1[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x4_0_conv2 (Conv2D)             (None, 56, 56, 256)  590080      batch_normalization_6[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "batch_normalization_7 (BatchNor (None, 56, 56, 256)  1024        x4_0_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_ResizeBilinear_4 (T [(None, 500, 500, 32 0           batch_normalization_1[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "u2_0_convTran (Conv2DTranspose) (None, 500, 500, 64) 16448       batch_normalization_3[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_ResizeBilinear_1 (T [(None, 242, 242, 64 0           batch_normalization_3[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "u3_0_convTran (Conv2DTranspose) (None, 242, 242, 64) 32832       batch_normalization_5[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_ResizeBilinear (Ten [(None, 112, 112, 12 0           batch_normalization_5[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "u4_0_convTran (Conv2DTranspose) (None, 112, 112, 128 131200      batch_normalization_7[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "x1_1_merge (Concatenate)        (None, 500, 500, 96) 0           tf_op_layer_ResizeBilinear_4[0][0\n",
      "                                                                 u2_0_convTran[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "x2_1_merge (Concatenate)        (None, 242, 242, 128 0           tf_op_layer_ResizeBilinear_1[0][0\n",
      "                                                                 u3_0_convTran[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "x3_1_merge (Concatenate)        (None, 112, 112, 256 0           tf_op_layer_ResizeBilinear[0][0] \n",
      "                                                                 u4_0_convTran[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "x1_1_conv1 (Conv2D)             (None, 498, 498, 32) 27680       x1_1_merge[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x2_1_conv1 (Conv2D)             (None, 240, 240, 64) 73792       x2_1_merge[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x3_1_conv1 (Conv2D)             (None, 110, 110, 128 295040      x3_1_merge[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x1_1_conv2 (Conv2D)             (None, 496, 496, 32) 9248        x1_1_conv1[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x2_1_conv2 (Conv2D)             (None, 238, 238, 64) 36928       x2_1_conv1[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x3_1_conv2 (Conv2D)             (None, 108, 108, 128 147584      x3_1_conv1[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_ResizeBilinear_5 (T [(None, 476, 476, 32 0           batch_normalization_1[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_ResizeBilinear_6 (T [(None, 476, 476, 32 0           x1_1_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "u2_1_convTran (Conv2DTranspose) (None, 476, 476, 64) 16448       x2_1_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_ResizeBilinear_2 (T [(None, 216, 216, 64 0           batch_normalization_3[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_ResizeBilinear_3 (T [(None, 216, 216, 64 0           x2_1_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "u3_1_convTran (Conv2DTranspose) (None, 216, 216, 64) 32832       x3_1_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x1_2_merge (Concatenate)        (None, 476, 476, 128 0           tf_op_layer_ResizeBilinear_5[0][0\n",
      "                                                                 tf_op_layer_ResizeBilinear_6[0][0\n",
      "                                                                 u2_1_convTran[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "x2_2_merge (Concatenate)        (None, 216, 216, 192 0           tf_op_layer_ResizeBilinear_2[0][0\n",
      "                                                                 tf_op_layer_ResizeBilinear_3[0][0\n",
      "                                                                 u3_1_convTran[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "x1_2_conv1 (Conv2D)             (None, 474, 474, 32) 36896       x1_2_merge[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x2_2_conv1 (Conv2D)             (None, 214, 214, 64) 110656      x2_2_merge[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x1_2_conv2 (Conv2D)             (None, 472, 472, 32) 9248        x1_2_conv1[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x2_2_conv2 (Conv2D)             (None, 212, 212, 64) 36928       x2_2_conv1[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_ResizeBilinear_7 (T [(None, 424, 424, 32 0           batch_normalization_1[0][0]      \n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_ResizeBilinear_8 (T [(None, 424, 424, 32 0           x1_1_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_ResizeBilinear_9 (T [(None, 424, 424, 32 0           x1_2_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "u2_2_convTran (Conv2DTranspose) (None, 424, 424, 64) 16448       x2_2_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x1_3_merge (Concatenate)        (None, 424, 424, 160 0           tf_op_layer_ResizeBilinear_7[0][0\n",
      "                                                                 tf_op_layer_ResizeBilinear_8[0][0\n",
      "                                                                 tf_op_layer_ResizeBilinear_9[0][0\n",
      "                                                                 u2_2_convTran[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "x1_3_conv1 (Conv2D)             (None, 422, 422, 32) 46112       x1_3_merge[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x1_3_conv2 (Conv2D)             (None, 420, 420, 32) 9248        x1_3_conv1[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "x1_3_conv3 (Conv2D)             (None, 420, 420, 2)  578         x1_3_conv2[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv2d (Conv2D)                 (None, 420, 420, 1)  3           x1_3_conv3[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "tf_op_layer_ResizeBilinear_10 ( [(None, 512, 512, 1) 0           conv2d[0][0]                     \n",
      "==================================================================================================\n",
      "Total params: 2,262,245\n",
      "Trainable params: 2,260,325\n",
      "Non-trainable params: 1,920\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "batch_size = 4\n",
    "model = upp_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "wwLLD73fQg2p",
    "outputId": "48b393df-d853-4027-e5e2-57e1a9141217"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      " 2/60 [>.............................] - ETA: 12s - loss: 0.7787WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.1472s vs `on_train_batch_end` time: 0.2919s). Check your callbacks.\n",
      "60/60 [==============================] - ETA: 0s - loss: 0.4658WARNING:tensorflow:Callbacks method `on_test_batch_end` is slow compared to the batch time (batch time: 0.0030s vs `on_test_batch_end` time: 0.1181s). Check your callbacks.\n",
      "60/60 [==============================] - 27s 453ms/step - loss: 0.4658 - val_loss: 0.7004\n",
      "Epoch 2/20\n",
      "60/60 [==============================] - 27s 455ms/step - loss: 0.4018 - val_loss: 0.5315\n",
      "Epoch 3/20\n",
      "60/60 [==============================] - 27s 456ms/step - loss: 0.3693 - val_loss: 0.5134\n",
      "Epoch 4/20\n",
      "60/60 [==============================] - 28s 464ms/step - loss: 0.3406 - val_loss: 0.7862\n",
      "Epoch 5/20\n",
      "60/60 [==============================] - 31s 523ms/step - loss: 0.3340 - val_loss: 0.5488\n",
      "Epoch 6/20\n",
      "60/60 [==============================] - 29s 488ms/step - loss: 0.3195 - val_loss: 0.4555\n",
      "Epoch 7/20\n",
      "60/60 [==============================] - 29s 481ms/step - loss: 0.2954 - val_loss: 0.3643\n",
      "Epoch 8/20\n",
      "60/60 [==============================] - 31s 514ms/step - loss: 0.3092 - val_loss: 0.4354\n",
      "Epoch 9/20\n",
      "60/60 [==============================] - 30s 497ms/step - loss: 0.2928 - val_loss: 0.4000\n",
      "Epoch 10/20\n",
      "60/60 [==============================] - 28s 474ms/step - loss: 0.2664 - val_loss: 0.4258\n",
      "Epoch 11/20\n",
      "60/60 [==============================] - 29s 479ms/step - loss: 0.2657 - val_loss: 0.2818\n",
      "Epoch 12/20\n",
      "60/60 [==============================] - 30s 499ms/step - loss: 0.2720 - val_loss: 0.2908\n",
      "Epoch 13/20\n",
      "60/60 [==============================] - 31s 514ms/step - loss: 0.2682 - val_loss: 0.2724\n",
      "Epoch 14/20\n",
      "60/60 [==============================] - 32s 532ms/step - loss: 0.2474 - val_loss: 0.3241\n",
      "Epoch 15/20\n",
      "60/60 [==============================] - 29s 480ms/step - loss: 0.2445 - val_loss: 0.2940\n",
      "Epoch 16/20\n",
      "60/60 [==============================] - 27s 458ms/step - loss: 0.2574 - val_loss: 0.2968\n",
      "Epoch 17/20\n",
      "60/60 [==============================] - 29s 484ms/step - loss: 0.2361 - val_loss: 0.2669\n",
      "Epoch 18/20\n",
      "60/60 [==============================] - 28s 467ms/step - loss: 0.2296 - val_loss: 0.2815\n",
      "Epoch 19/20\n",
      "60/60 [==============================] - 27s 458ms/step - loss: 0.2427 - val_loss: 0.2788\n",
      "Epoch 20/20\n",
      "60/60 [==============================] - 27s 456ms/step - loss: 0.2218 - val_loss: 0.2742\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(image_datagen.flow(x_train, y_train, batch_size = batch_size), \n",
    "                      steps_per_epoch=x_train.shape[0]/batch_size,\n",
    "                      validation_data=(x_valid, y_valid), \n",
    "                      validation_steps=x_valid.shape[0]/batch_size,\n",
    "                      epochs=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 295
    },
    "colab_type": "code",
    "id": "Ylf8u2shmm5t",
    "outputId": "d556b901-d191-4c1e-ff6e-cd45fb7c61e5"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot training & validation loss values\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', 'Valid'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Plot training & validation loss values\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', 'Valid'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "K2jHslukmZS2"
   },
   "outputs": [],
   "source": [
    "# Saving the model\n",
    "model.save('model.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "QpQMJc-7WLCO"
   },
   "source": [
    "Testing Phase"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 696
    },
    "colab_type": "code",
    "id": "VGI2oBDc6pbf",
    "outputId": "bc47e05b-1350-44a8-e9d7-1b50e918ce3f"
   },
   "outputs": [],
   "source": [
    "# Load the model\n",
    "model = load_model('model.h5')\n",
    "\n",
    "# Read the Test Images and their respective masks\n",
    "x_test = np.load(os.path.join(os.getcwd(),\"Test_Images.npy\"))\n",
    "y_test = np.load(os.path.join(os.getcwd(),\"Test_Masks.npy\"))\n",
    "\n",
    "# Convert Masks to Binary Masks\n",
    "y_test = np.where(y_test > 1, 1, 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 52
    },
    "colab_type": "code",
    "id": "l6fgfBAqLnUK",
    "outputId": "2bd2383f-d3a0-480e-88d4-6db92fbf73df"
   },
   "outputs": [],
   "source": [
    "import timeit\n",
    "start = timeit.default_timer()\n",
    "\n",
    "output = []\n",
    "for test_img in x_test:\n",
    "  output.append(model.predict(test_img.reshape(1,512,512,3))[0,:,:,:])\n",
    "output = np.array(output)\n",
    "\n",
    "stop = timeit.default_timer()\n",
    "\n",
    "# Convert pixel values from float to integer\n",
    "output = np.where(output > 0.9, 1, 0)\n",
    "output = output.astype('uint8')\n",
    "\n",
    "print('Time to inference the model for a single image:', (stop-start)/test_img.shape[0],'s')\n",
    "print('A 30 FPS real-time engine will require ',1/30,'s inference time for a single Image')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "pKjJUI0l-DSu"
   },
   "source": [
    "Dilation Operation on the predicted masks to smooth out the masks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "_CjmAoC4RkyU"
   },
   "outputs": [],
   "source": [
    "  kernel = np.ones((15,15),np.uint8)\n",
    "  dilate_opt = []\n",
    "  for image in output:\n",
    "    dilate_opt.append(cv2.dilate(image,kernel,iterations = 1))\n",
    "  dilate_opt = np.array(dilate_opt)\n",
    "  dilate_opt = np.expand_dims(dilate_opt, axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 404
    },
    "colab_type": "code",
    "id": "fuejABgNM6DP",
    "outputId": "a04c6dfc-ab40-408b-cf63-9f37f01c266a"
   },
   "outputs": [],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "\n",
    "for i in list(np.linspace(1,13,4, dtype = int)):\n",
    "  idx = randint(0,24)\n",
    "  plt.subplot(6,4,i)\n",
    "  plt.title('Image')\n",
    "  plt.axis('off')\n",
    "  plt.imshow(x_test[idx,:,:,:])\n",
    "\n",
    "  plt.subplot(6,4,i+1)\n",
    "  plt.title('Predicted')\n",
    "  plt.axis('off')\n",
    "  plt.imshow(output[idx,:,:,0])\n",
    "\n",
    "  plt.subplot(6,4,i+2)\n",
    "  plt.title('Dilating Predicted')\n",
    "  plt.axis('off')\n",
    "  plt.imshow(dilate_opt[idx,:,:,0])\n",
    "\n",
    "  plt.subplot(6,4,i+3)\n",
    "  plt.title('Actual')\n",
    "  plt.axis('off')\n",
    "  plt.imshow(y_test[idx,:,:,0])\n",
    "\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "machine_shape": "hm",
   "name": "Portrait_Segmentation.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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}