Created
April 21, 2018 02:51
-
-
Save Superdanby/78b0cc0d564d134300557195f362d415 to your computer and use it in GitHub Desktop.
temp
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| { | |
| "cells": [ | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "e61ef2d8-f315-4f7f-b07e-1de0f4e8441a", | |
| "_uuid": "1677fddbb95f7545b6540e9201f3339a0fdbfc5d" | |
| }, | |
| "source": [ | |
| "# Intro\n", | |
| "Hello! This rather quick and dirty kernel shows how to get started on segmenting nuclei using a neural network in Keras. \n", | |
| "\n", | |
| "The architecture used is the so-called [U-Net](https://arxiv.org/abs/1505.04597), which is very common for image segmentation problems such as this. I believe they also have a tendency to work quite well even on small datasets.\n", | |
| "\n", | |
| "Let's get started importing everything we need!" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 1, | |
| "metadata": { | |
| "_cell_guid": "c332549b-8d23-4bb5-8497-e7a8eb8b21d2", | |
| "_uuid": "5c38504af3a84bee68c66d3cde74443c58df422f" | |
| }, | |
| "outputs": [ | |
| { | |
| "name": "stderr", | |
| "output_type": "stream", | |
| "text": [ | |
| "Using TensorFlow backend.\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "import os\n", | |
| "import sys\n", | |
| "import random\n", | |
| "import warnings\n", | |
| "\n", | |
| "import numpy as np\n", | |
| "import pandas as pd\n", | |
| "\n", | |
| "import matplotlib.pyplot as plt\n", | |
| "\n", | |
| "from tqdm import tqdm\n", | |
| "from itertools import chain\n", | |
| "from skimage.io import imread, imshow, imread_collection, concatenate_images\n", | |
| "from skimage.transform import resize\n", | |
| "from skimage.morphology import label\n", | |
| "\n", | |
| "from keras.models import Model, load_model\n", | |
| "from keras.layers import Input\n", | |
| "from keras.layers.core import 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 keras import backend as K\n", | |
| "\n", | |
| "import tensorflow as tf\n", | |
| "\n", | |
| "# Set some parameters\n", | |
| "TRAIN_PATH = '/hdd/dataset/nuclei_dataset/stage1_train/'\n", | |
| "TEST_PATH = '/hdd/dataset/nuclei_dataset/stage1_test/'\n", | |
| "IMG_HEIGHT = 256\n", | |
| "IMG_WIDTH = 256\n", | |
| "IMG_CHANNELS = 3\n", | |
| "\n", | |
| "warnings.filterwarnings('ignore', category=UserWarning, module='skimage')" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 2, | |
| "metadata": { | |
| "_cell_guid": "ffa0caf0-2d1b-40f2-865b-8e6db88526b6", | |
| "_uuid": "3fb9d6530fbbd0e22e41fc4fd9fd9fc0bff027ac" | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "# Get train and test IDs\n", | |
| "train_ids = next(os.walk(TRAIN_PATH))[1]\n", | |
| "test_ids = next(os.walk(TEST_PATH))[1]" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "59c4a25d-645f-4b74-9c53-145ac78cc481", | |
| "_uuid": "875af74f980236825de3a650825b46e25632422c" | |
| }, | |
| "source": [ | |
| "# Get the data\n", | |
| "Let's first import all the images and associated masks. I downsample both the training and test images to keep things light and manageable, but we need to keep a record of the original sizes of the test images to upsample our predicted masks and create correct run-length encodings later on. There are definitely better ways to handle this, but it works fine for now!" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "# Data Generation" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 3, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "Getting and resizing test images ... \n" | |
| ] | |
| }, | |
| { | |
| "name": "stderr", | |
| "output_type": "stream", | |
| "text": [ | |
| "100%|██████████| 65/65 [00:00<00:00, 99.58it/s]\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "from keras.utils import Sequence\n", | |
| "import cv2\n", | |
| "from sklearn.utils import shuffle\n", | |
| "from skimage import color\n", | |
| "from skimage.transform import AffineTransform, warp\n", | |
| "from scipy.ndimage import morphology\n", | |
| "import copy\n", | |
| "\n", | |
| "Y_train_label = []\n", | |
| "\n", | |
| "class data_generator(Sequence):\n", | |
| " '''Generates data for Keras'''\n", | |
| " def __init__(self, ids, IMG_WIDTH, IMG_HEIGHT, IMG_CHANNELS, batch_size=10, training=True):\n", | |
| " '''Initialization'''\n", | |
| " self.ids = ids\n", | |
| " self.batch_size = batch_size\n", | |
| " self.training = training\n", | |
| " self.IMG_WIDTH = IMG_WIDTH\n", | |
| " self.IMG_HEIGHT = IMG_HEIGHT\n", | |
| " self.IMG_CHANNELS = IMG_CHANNELS\n", | |
| " def __len__(self):\n", | |
| " '''Denotes the number of batches per epoch'''\n", | |
| " return int(np.ceil(float(len(self.ids))/self.batch_size))\n", | |
| " def on_epoch_end(self):\n", | |
| " '''Updates indexes after each epoch'''\n", | |
| " if self.training: self.ids = shuffle(self.ids)\n", | |
| " def __getitem__(self, i): \n", | |
| " X = []\n", | |
| " for n, id_ in tqdm(enumerate(self.ids), total=self.batch_size):\n", | |
| " path = TRAIN_PATH + id_\n", | |
| " img = imread(path + '/images/' + id_ + '.png')[:,:,:self.IMG_CHANNELS]\n", | |
| " img = img.astype(int)\n", | |
| " img = resize(img, (self.IMG_HEIGHT, self.IMG_WIDTH), mode='constant', preserve_range=True) # skimage 0.13.1\n", | |
| "# img = resize(img, (IMG_HEIGHT, IMG_WIDTH), mode='constant', preserve_range=True, anti_aliasing=True) # wait for skimage 0.14.0\n", | |
| " img = color.rgb2hsv(img) # should I keep hue only? Maybe it can learn to invert purple images?\n", | |
| " mask_ = []\n", | |
| " for mask_file in next(os.walk(path + '/masks/'))[2]:\n", | |
| " msk = imread(path + '/masks/' + mask_file)\n", | |
| " msk = resize(msk, (self.IMG_HEIGHT, self.IMG_WIDTH), mode='constant', preserve_range=True)\n", | |
| " mask_.append(msk)\n", | |
| " \n", | |
| " if self.training:\n", | |
| " # flip vertical\n", | |
| " if np.random.rand() < .5:\n", | |
| " img = np.flip(img, 0)\n", | |
| " for mask in mask_:\n", | |
| " mask = np.flip(mask, 0)\n", | |
| " \n", | |
| " # flip horizontal\n", | |
| " if np.random.rand() < .5:\n", | |
| " img = np.flip(img, 1)\n", | |
| " for mask in mask_:\n", | |
| " mask = np.flip(mask, 1)\n", | |
| " \n", | |
| " # rotation, shearing\n", | |
| "# if np.random.rand() < 0.5:\n", | |
| "# y, x, _ = img.shape\n", | |
| "# h, w = y, x\n", | |
| " \n", | |
| "# rotT = np.random.uniform(-45,45)\n", | |
| "# shearT = np.random.uniform(-5,5)\n", | |
| "# translationX = np.random.uniform(-w,w) * .1\n", | |
| "# translationY = np.random.uniform(-h,h) * .1\n", | |
| "# atmtx = AffineTransform(rotation=rotT, shear=shearT, translation=(translationX, translationY))\n", | |
| "\n", | |
| "# img = warp(img, atmtx)\n", | |
| "# for mask in mask_:\n", | |
| "# mask = warp(mask, atmtx)\n", | |
| "\n", | |
| " ### end of data augmentation ###\n", | |
| " \n", | |
| " X.append(img)\n", | |
| "# imshow(color.hsv2rgb(img))\n", | |
| "# plt.show()\n", | |
| " labeled_mask = np.zeros(img.shape[:2], dtype=np.uint8) # semantic segmentation ground truth\n", | |
| " dt = np.zeros(img.shape[:2], dtype=np.float32) # distance transform ground truth\n", | |
| " dir_grad = [np.zeros(img.shape[:2], dtype=np.float32)]\n", | |
| " dir_grad.append(np.zeros(img.shape[:2], dtype=np.float32))\n", | |
| " dir_grad = np.asarray(dir_grad) # direction ground truth (gradient of distance transform)\n", | |
| " for i, mask in enumerate(mask_):\n", | |
| " dt_ = -morphology.distance_transform_edt(mask).astype(np.float32) # basin\n", | |
| " dir_grad_y, dir_grad_x = np.array(mask).astype(int) * np.gradient(dt_)\n", | |
| "\n", | |
| " # normalize direction\n", | |
| " dir_grad_ = np.sqrt(dir_grad_x * dir_grad_x + dir_grad_y * dir_grad_y)\n", | |
| " dir_grad_ = np.where(dir_grad_ < 1e-5, 1e-5, dir_grad_)\n", | |
| " dir_grad_x = dir_grad_x / dir_grad_\n", | |
| " dir_grad_y = dir_grad_y / dir_grad_\n", | |
| " # print(dt_)\n", | |
| " # sys.exit()\n", | |
| "\n", | |
| " # superpositioning\n", | |
| "# print(f'{dir_grad[0].shape} {dir_grad_x.shape}')\n", | |
| " dir_grad[0] = np.where(dir_grad_x != 0, dir_grad_x, dir_grad[0])\n", | |
| " dir_grad[1] = np.where(dir_grad_y != 0, dir_grad_y, dir_grad[1])\n", | |
| " dt = np.where(dt_ != 0, dt_, dt)\n", | |
| " mask = np.where(mask > 0, i, 0)\n", | |
| " labeled_mask = labeled_mask | mask\n", | |
| " gt = np.maximum(np.where(dt == -1, 0, dt), -1) # ground truth\n", | |
| " Y_train_label.append([labeled_mask, dir_grad[0], dir_grad[1], dt, gt])\n", | |
| "# print(f'shape {len(Y_train_label)}')\n", | |
| " \n", | |
| " # Check if training data looks all right\n", | |
| " ix = random.randint(0, self.batch_size)\n", | |
| " print(X[ix])\n", | |
| " rgb = color.hsv2rgb(X[ix])\n", | |
| " imshow(rgb.astype(int))\n", | |
| " plt.show()\n", | |
| " imshow(Y_train_label[ix][0])\n", | |
| " plt.show()\n", | |
| " imshow(Y_train_label[ix][1])\n", | |
| " plt.show()\n", | |
| " imshow(Y_train_label[ix][2])\n", | |
| " plt.show()\n", | |
| " imshow(Y_train_label[ix][3])\n", | |
| " plt.show()\n", | |
| " imshow(Y_train_label[ix][4])\n", | |
| " plt.show()\n", | |
| " \n", | |
| " \n", | |
| " return np.asarray(X), np.asarray([row[4] for row in Y_train_label])\n", | |
| " \n", | |
| "\n", | |
| "# Get and resize test images\n", | |
| "X_test = np.zeros((len(test_ids), IMG_HEIGHT, IMG_WIDTH, IMG_CHANNELS), dtype=np.uint8)\n", | |
| "sizes_test = []\n", | |
| "print('Getting and resizing test images ... ')\n", | |
| "sys.stdout.flush()\n", | |
| "for n, id_ in tqdm(enumerate(test_ids), total=len(test_ids)):\n", | |
| " path = TEST_PATH + id_\n", | |
| " img = imread(path + '/images/' + id_ + '.png')[:,:,:IMG_CHANNELS]\n", | |
| " sizes_test.append([img.shape[0], img.shape[1]])\n", | |
| " img = resize(img, (IMG_HEIGHT, IMG_WIDTH), mode='constant', preserve_range=True)\n", | |
| " X_test[n] = img\n" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "c0523b03-1fc5-4505-a1b8-eb35ee617c8a", | |
| "_uuid": "d4f8327802a1ec6139ce0585953986272ba62ce1" | |
| }, | |
| "source": [ | |
| "Let's see if things look all right by drawing some random images and their associated masks." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 4, | |
| "metadata": { | |
| "_cell_guid": "88829b53-50ce-45d9-9540-77dd7384ad4c", | |
| "_uuid": "283af26f0860b7069bdfd133c746e5d20971542c" | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "# # Check if training data looks all right\n", | |
| "# ix = random.randint(0, len(train_ids))\n", | |
| "# imshow(X_train[ix])\n", | |
| "# plt.show()\n", | |
| "# imshow(np.squeeze(Y_train[ix]))\n", | |
| "# plt.show()" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "2574ffe9-b911-4bfd-a00f-9ba5c25f45de", | |
| "_uuid": "938648da705689a0f940ff462477c801db3f0737" | |
| }, | |
| "source": [ | |
| "Seems good!\n", | |
| "\n", | |
| "# Create our Keras metric\n", | |
| "\n", | |
| "Now we try to define the *mean average precision at different intersection over union (IoU) thresholds* metric in Keras. TensorFlow has a mean IoU metric, but it doesn't have any native support for the mean over multiple thresholds, so I tried to implement this. **I'm by no means certain that this implementation is correct, though!** Any assistance in verifying this would be most welcome! \n", | |
| "\n", | |
| "*Update: This implementation is most definitely not correct due to the very large discrepancy between the results reported here and the LB results. It also seems to just increase over time no matter what when you train ... *" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 5, | |
| "metadata": { | |
| "_cell_guid": "c1df6f3a-d58f-434b-9216-ef7be38637d4", | |
| "_uuid": "5abd38950ae99b60f8afec7656eb654a48d449fe" | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "# Define IoU metric\n", | |
| "def mean_iou_unet(y_true, y_pred):\n", | |
| " prec = []\n", | |
| " for t in np.arange(0.5, 1.0, 0.05):\n", | |
| " y_pred_ = tf.to_int32(y_pred > t)\n", | |
| " score, up_opt = tf.metrics.mean_iou(y_true, y_pred_, 2)\n", | |
| " K.get_session().run(tf.local_variables_initializer())\n", | |
| " with tf.control_dependencies([up_opt]):\n", | |
| " score = tf.identity(score)\n", | |
| " prec.append(score)\n", | |
| " return K.mean(K.stack(prec), axis=0)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "c3b9f148-1dba-4b6a-981b-6cdbf394fc3c", | |
| "_uuid": "986488a4c5223576be370e224426a30431911eb2" | |
| }, | |
| "source": [ | |
| "# Build and train our neural network\n", | |
| "Next we build our U-Net model, loosely based on [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://arxiv.org/pdf/1505.04597.pdf) and very similar to [this repo](https://github.com/jocicmarko/ultrasound-nerve-segmentation) from the Kaggle Ultrasound Nerve Segmentation competition.\n", | |
| "\n", | |
| "" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 6, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "# custom loss\n", | |
| "\n", | |
| "def unet_custom(y_true, y_pred):\n", | |
| " y_true_ = y_true\n", | |
| " y_pred_ = y_pred\n", | |
| " return K.mean(K.binary_crossentropy(y_true_, y_pred_)) " | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 7, | |
| "metadata": { | |
| "_cell_guid": "c1dbc57c-b497-4ccb-b077-2053203ab7ed", | |
| "_uuid": "0aa97d66c29f45dfac9b0f45fcf74ba0e778ba5d" | |
| }, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "__________________________________________________________________________________________________\n", | |
| "Layer (type) Output Shape Param # Connected to \n", | |
| "==================================================================================================\n", | |
| "input_1 (InputLayer) (None, 256, 256, 3) 0 \n", | |
| "__________________________________________________________________________________________________\n", | |
| "lambda_1 (Lambda) (None, 256, 256, 3) 0 input_1[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_1 (Conv2D) (None, 256, 256, 8) 224 lambda_1[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_2 (Conv2D) (None, 256, 256, 8) 584 conv2d_1[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "max_pooling2d_1 (MaxPooling2D) (None, 128, 128, 8) 0 conv2d_2[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_3 (Conv2D) (None, 128, 128, 16) 1168 max_pooling2d_1[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_4 (Conv2D) (None, 128, 128, 16) 2320 conv2d_3[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "max_pooling2d_2 (MaxPooling2D) (None, 64, 64, 16) 0 conv2d_4[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_5 (Conv2D) (None, 64, 64, 32) 4640 max_pooling2d_2[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_6 (Conv2D) (None, 64, 64, 32) 9248 conv2d_5[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "max_pooling2d_3 (MaxPooling2D) (None, 32, 32, 32) 0 conv2d_6[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_7 (Conv2D) (None, 32, 32, 64) 18496 max_pooling2d_3[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_8 (Conv2D) (None, 32, 32, 64) 36928 conv2d_7[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "max_pooling2d_4 (MaxPooling2D) (None, 16, 16, 64) 0 conv2d_8[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_9 (Conv2D) (None, 16, 16, 128) 73856 max_pooling2d_4[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_10 (Conv2D) (None, 16, 16, 128) 147584 conv2d_9[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_transpose_1 (Conv2DTrans (None, 32, 32, 64) 32832 conv2d_10[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "concatenate_1 (Concatenate) (None, 32, 32, 128) 0 conv2d_transpose_1[0][0] \n", | |
| " conv2d_8[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_11 (Conv2D) (None, 32, 32, 64) 73792 concatenate_1[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_12 (Conv2D) (None, 32, 32, 64) 36928 conv2d_11[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_transpose_2 (Conv2DTrans (None, 64, 64, 32) 8224 conv2d_12[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "concatenate_2 (Concatenate) (None, 64, 64, 64) 0 conv2d_transpose_2[0][0] \n", | |
| " conv2d_6[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_13 (Conv2D) (None, 64, 64, 32) 18464 concatenate_2[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_14 (Conv2D) (None, 64, 64, 32) 9248 conv2d_13[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_transpose_3 (Conv2DTrans (None, 128, 128, 16) 2064 conv2d_14[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "concatenate_3 (Concatenate) (None, 128, 128, 32) 0 conv2d_transpose_3[0][0] \n", | |
| " conv2d_4[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_15 (Conv2D) (None, 128, 128, 16) 4624 concatenate_3[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_16 (Conv2D) (None, 128, 128, 16) 2320 conv2d_15[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_transpose_4 (Conv2DTrans (None, 256, 256, 8) 520 conv2d_16[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "concatenate_4 (Concatenate) (None, 256, 256, 16) 0 conv2d_transpose_4[0][0] \n", | |
| " conv2d_2[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_17 (Conv2D) (None, 256, 256, 8) 1160 concatenate_4[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_18 (Conv2D) (None, 256, 256, 8) 584 conv2d_17[0][0] \n", | |
| "__________________________________________________________________________________________________\n", | |
| "conv2d_19 (Conv2D) (None, 256, 256, 1) 9 conv2d_18[0][0] \n", | |
| "==================================================================================================\n", | |
| "Total params: 485,817\n", | |
| "Trainable params: 485,817\n", | |
| "Non-trainable params: 0\n", | |
| "__________________________________________________________________________________________________\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "# Build U-Net model\n", | |
| "inputs = Input((IMG_HEIGHT, IMG_WIDTH, IMG_CHANNELS))\n", | |
| "s = Lambda(lambda x: x / 255) (inputs)\n", | |
| "\n", | |
| "c1 = Conv2D(8, (3, 3), activation='relu', padding='same') (s)\n", | |
| "c1 = Conv2D(8, (3, 3), activation='relu', padding='same') (c1)\n", | |
| "p1 = MaxPooling2D((2, 2)) (c1)\n", | |
| "\n", | |
| "c2 = Conv2D(16, (3, 3), activation='relu', padding='same') (p1)\n", | |
| "c2 = Conv2D(16, (3, 3), activation='relu', padding='same') (c2)\n", | |
| "p2 = MaxPooling2D((2, 2)) (c2)\n", | |
| "\n", | |
| "c3 = Conv2D(32, (3, 3), activation='relu', padding='same') (p2)\n", | |
| "c3 = Conv2D(32, (3, 3), activation='relu', padding='same') (c3)\n", | |
| "p3 = MaxPooling2D((2, 2)) (c3)\n", | |
| "\n", | |
| "c4 = Conv2D(64, (3, 3), activation='relu', padding='same') (p3)\n", | |
| "c4 = Conv2D(64, (3, 3), activation='relu', padding='same') (c4)\n", | |
| "p4 = MaxPooling2D(pool_size=(2, 2)) (c4)\n", | |
| "\n", | |
| "c5 = Conv2D(128, (3, 3), activation='relu', padding='same') (p4)\n", | |
| "c5 = Conv2D(128, (3, 3), activation='relu', padding='same') (c5)\n", | |
| "\n", | |
| "u6 = Conv2DTranspose(64, (2, 2), strides=(2, 2), padding='same') (c5)\n", | |
| "u6 = concatenate([u6, c4])\n", | |
| "c6 = Conv2D(64, (3, 3), activation='relu', padding='same') (u6)\n", | |
| "c6 = Conv2D(64, (3, 3), activation='relu', padding='same') (c6)\n", | |
| "\n", | |
| "u7 = Conv2DTranspose(32, (2, 2), strides=(2, 2), padding='same') (c6)\n", | |
| "u7 = concatenate([u7, c3])\n", | |
| "c7 = Conv2D(32, (3, 3), activation='relu', padding='same') (u7)\n", | |
| "c7 = Conv2D(32, (3, 3), activation='relu', padding='same') (c7)\n", | |
| "\n", | |
| "u8 = Conv2DTranspose(16, (2, 2), strides=(2, 2), padding='same') (c7)\n", | |
| "u8 = concatenate([u8, c2])\n", | |
| "c8 = Conv2D(16, (3, 3), activation='relu', padding='same') (u8)\n", | |
| "c8 = Conv2D(16, (3, 3), activation='relu', padding='same') (c8)\n", | |
| "\n", | |
| "u9 = Conv2DTranspose(8, (2, 2), strides=(2, 2), padding='same') (c8)\n", | |
| "u9 = concatenate([u9, c1], axis=3)\n", | |
| "c9 = Conv2D(8, (3, 3), activation='relu', padding='same') (u9)\n", | |
| "c9 = Conv2D(8, (3, 3), activation='relu', padding='same') (c9)\n", | |
| "\n", | |
| "outputs = Conv2D(1, (1, 1), activation='sigmoid') (c9)\n", | |
| "\n", | |
| "model = Model(inputs=[inputs], outputs=[outputs])\n", | |
| "model.compile(optimizer='adam', loss=unet_custom, metrics=[mean_iou_unet])\n", | |
| "model.summary()" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "72330944-6ce7-4070-b276-c3c4b20c4fe5", | |
| "_uuid": "92350b6e18cc50f3fa7b6e9a02d39fcbff8238f7" | |
| }, | |
| "source": [ | |
| "Next we fit the model on the training data, using a validation split of 0.1. We use a small batch size because we have so little data. I recommend using checkpointing and early stopping when training your model. I won't do it here to make things a bit more reproducible (although it's very likely that your results will be different anyway). I'll just train for 10 epochs, which takes around 10 minutes in the Kaggle kernel with the current parameters. \n", | |
| "\n", | |
| "*Update: Added early stopping and checkpointing and increased to 30 epochs.*" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 8, | |
| "metadata": { | |
| "_cell_guid": "9415b1c4-aa69-41b9-a1e3-d6053dbd4f64", | |
| "_uuid": "c060db22daa2abf12b28240cd81bbcbf1ce1bf87" | |
| }, | |
| "outputs": [ | |
| { | |
| "name": "stderr", | |
| "output_type": "stream", | |
| "text": [ | |
| "\r", | |
| " 0%| | 0/10 [00:00<?, ?it/s]" | |
| ] | |
| }, | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "Epoch 1/1\n" | |
| ] | |
| }, | |
| { | |
| "name": "stderr", | |
| "output_type": "stream", | |
| "text": [ | |
| "603it [03:15, 3.08it/s] \n", | |
| "/hdd/home/superdanby/.conda/envs/yee/lib/python3.6/site-packages/skimage/io/_plugins/matplotlib_plugin.py:51: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n", | |
| " out_of_range_float = (np.issubdtype(image.dtype, np.float) and\n", | |
| "/hdd/home/superdanby/.conda/envs/yee/lib/python3.6/site-packages/matplotlib/axes/_base.py:1400: MatplotlibDeprecationWarning: The 'box-forced' keyword argument is deprecated since 2.2.\n", | |
| " \" since 2.2.\", cbook.mplDeprecation)\n" | |
| ] | |
| }, | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "[[[0. 0. 1.41625977]\n", | |
| " [0. 0. 1. ]\n", | |
| " [0. 0. 1. ]\n", | |
| " ...\n", | |
| " [0. 0. 8.72509766]\n", | |
| " [0. 0. 9.2175293 ]\n", | |
| " [0. 0. 9.484375 ]]\n", | |
| "\n", | |
| " [[0. 0. 1. ]\n", | |
| " [0. 0. 1. ]\n", | |
| " [0. 0. 1.38452148]\n", | |
| " ...\n", | |
| " [0. 0. 8.703125 ]\n", | |
| " [0. 0. 9.31616211]\n", | |
| " [0. 0. 8.0769043 ]]\n", | |
| "\n", | |
| " [[0. 0. 1.36254883]\n", | |
| " [0. 0. 1. ]\n", | |
| " [0. 0. 1.29663086]\n", | |
| " ...\n", | |
| " [0. 0. 9. ]\n", | |
| " [0. 0. 9.421875 ]\n", | |
| " [0. 0. 8.63745117]]\n", | |
| "\n", | |
| " ...\n", | |
| "\n", | |
| " [[0. 0. 2. ]\n", | |
| " [0. 0. 1.82202148]\n", | |
| " [0. 0. 1.59350586]\n", | |
| " ...\n", | |
| " [0. 0. 1. ]\n", | |
| " [0. 0. 1.82202148]\n", | |
| " [0. 0. 1.421875 ]]\n", | |
| "\n", | |
| " [[0. 0. 1.9362793 ]\n", | |
| " [0. 0. 1.80883789]\n", | |
| " [0. 0. 2. ]\n", | |
| " ...\n", | |
| " [0. 0. 1.453125 ]\n", | |
| " [0. 0. 1.421875 ]\n", | |
| " [0. 0. 1.9362793 ]]\n", | |
| "\n", | |
| " [[0. 0. 1.93188477]\n", | |
| " [0. 0. 1.7019043 ]\n", | |
| " [0. 0. 2. ]\n", | |
| " ...\n", | |
| " [0. 0. 1. ]\n", | |
| " [0. 0. 1.50244141]\n", | |
| " [0. 0. 1.92749023]]]\n" | |
| ] | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUgAAAEYCAYAAAA+mm/EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJztnX+UFdWV77+bRkFQlB+KCCioTaImjj8QBF0mDklEx3mYeT6fxmWI4wvJjE5MxpcBTVbicsaMzoxx4kvGBEeNZhmJMfFBEmM0PBOX4YeAIopEaPkh3fJDECWDCnb3fn9UVXd1ddWtU1XnVJ1Td3/WqtW3TtWte+DCp/c++9QpYmYIgiAI/RlQdQcEQRBsRQQpCIKQgAhSEAQhARGkIAhCAiJIQRCEBESQgiAICRgTJBHNJKJXiaiNiOaZ+hxBEARTkIl5kETUAmA9gE8CaAewAsAVzPyK9g8TBEEwhKkIcgqANmbeyMwHACwAMMvQZwmCIBhhoKHrjgWwNbTfDmBq+AQimgNgDgC0oOXMIRiWeLFBJ1G/tv3r9Ee+XSOG9mtreWtf4vkHxvY9/+CO5HOFZPYfO6Rf26DX362gJ9kYclLj4++uK6cfcdAhg1PP4ffeL/w5+yceknjswKY3djHzkcH+BecP5d1vdSlfe9Wa/b9h5pnFelgMU4JMhZnnA5gPAMNoBE+lGYnnTniw75ewecp7QH9nFmbvzLNj24c9vKxf26ZvTUu8zsSblmrrUzOw/sYp/dom/e1zFfQkG6f+uPE/wjVnVHcb74APn5x6TvfqYiNeG398WuPjV3x9S3h/11tdWP6bccrXP2jMa6Py9UwfplLsDgDjQ/vj/DZr2XtFvBzjaCTH4HjaOUIyLsgxjSrlqMqA09IlqhdGF3crbzZgSpArALQS0UQiOhjA5QAW5b3Y5invaetYmYgk602SBG2QY9Ho0AQMoBusvNmAkRSbmTuJ6DoAvwHQAuA+Zl5b5Jqbp7yHCc8dYkSWjaLHuPRaMIdr0aMNMqyK4z+zOjXNjtINOyJDVYyNQTLz4wAe13lNFyPJTd+aJmOSKbgmxWbizb/pzYKOvLvYv2MGo8ux5RXlThpBEGIJyzFuPw+SYgul0PL0MQCArvPfqLgnHptu8/7zTJwn0bIQDwPoskR8qoggLScQYdpxE6IcvbT/3NQd0/b2awvkKNSfN/9mWqFU25bIUBVJsRtQVYGm5eljerYs79FJnBzjiMpRZCkkwQC6mJU3GxBB5iBL0SVrgUa36PLQSI6q4hTcRsd4YxzdGbY0iOg+ItpJRC+H2v6ViP5IRGuI6DEiOiJ07EZ/8ZxXiegClf6KIJEvUlQRn1Svhbpz/GdWK5/LYHRl2BT4IYDorYhPAfgIM58Kb8GcGwGAiE6GNx/7FP89/+EvqtMQGYP0MSXJLBSJHnWMQapEh3FjkIKb6JpMHjcf8vjPrMbG6IkMdGnMnJn5GSKaEGl7MrS7DMCl/utZABYw834Am4ioDd6iOg3/E4sgBQD5UmcZb3QX3XfaqESS3p00pfLXAH7ivx4LT5gB7X5bQyTFtoi8UaAtU30EoTGErgwbgFFEtDK0zVH+JKKvAegE8FCRHksEKSgTTq+TokeZB2kP3atfqWBBimQYQHe2FHsXM0/O+jlE9DkAFwOYwb0rgudaQEcEaRld579hRSVbyM+WRz7aZ/+4y16qqCe9qXRYlFUuZNFlYp3CEEQ0E8A/APgYM4cXFV0E4MdE9G0AxwBoBZB6j6sI0kKySLKs9FqKM/nZ8shHK5UkYMfqPt6dNPoESUQPA/g4vFS8HcA34VWtBwF4iogAYBkzf5GZ1xLRIwBegZd6X8vMqav3iiAtxcVI0vX0+o2vTu/Xdsy/Lsl0jWj0GG6vWpJZOPLupbFzIYsuWNHN+gTJzFfENN/b4PxbAdya5TOkSGMxXee/0TBClOhRH3FyDNqTjmUlSZ62EpVh8dV8kLVIUzkSQTpAlVVqVTm6Hj2m8cZXp2eOJuNwMZLUBYPQ5VhM5lZvM3Jg5ll9NiEbjeQYFqLrclSNEKuMJAeOObpnc5luJuXNBmoryDghiiSTCctwx7S9SpHjxHlLnZdjVtIkaSI6dF2KAS6m2LUTZFq0KJJMRlWMdSNr6qwjkiwyHumuMAldPEB5swE7eqEJkZ9QFrrS7TSSZOiiJBnAB2hR3mygVoIUhLzkKcAkSVI1zXatql0UZokga8XAicf12QRBlTIq1S5Gkd0g5c0GaiNI3el1nBBFlPVGZxQp9Mcr0gxQ3mzAjl44hkjSfd6bNaVnC6NjrmOAShTZXGm2pNhNg0iyOLuvmdZvK4OoFKOi1FnVdmlSuGm89SAHKG82YEcvCpIlvT74iRXaPlckmZ+yZJiFIpIU1OhiUt5soBaCdJH1907G+nszL3VXCxrJsWpxhqNJkaReglsNZQzSUnRGjwF5osiwGJtNklULUJWwJIuKslGaXTQFN13J3vOr1n5bEbp5gPJmA3b0oslpFkm6IseA6FhlEY677KV+MrR9fLKoDKNIFdtiTESPAVmiyCQZ1l2SrskxQGckCfSKskw5tgwb1mcrSl5xMtTHH2UMUiMm5SfUj0MWpq6034do8SYqSpvHKuOEqCJJ3dFjgGtV7NqvB5lXnp2bttSmSk0HHdxnnz84UOrn2xg9HrLwuUwp9HuzpvQRq81SDGgkwpZhw9C1t9yFSZhhzfxGVWojSBNRZOemLQDSU+jgvKKsv3cyJl2zUsu1gP5iDLeXLUkbCYSnKsqoJMugc9v2hoWYzm3bc187SZKmokdYdAuhKm7pvCIaCVCXHAU30Fm4sYE8Y5LD/2JDrs9iwLk7aWoTQZpGRFhfsqbbZZMWRbqELdVpVdzqreOUWalOSq9tZ+S91axQfsjC50pPn7MQl0oXSa/DRKPIvBFiGgz1xy3Y8sgFiSAFIUTWccky0SXEKpEI0mL2/vqEfptt1H0+pCvYHE2qUqRKbSKKZMidNNaSJEMbJdmsVJVeJxFNu21Pw3UTJ8li4lR/YJctD+1qihRbJGg/tskxjMtS7Nq7t9DdM4EQ9/yqtXBUGUSQLlFIkES0GcCfAHQB6GTmyUQ0AsBPAEwAsBnAZcy8p1g386Mix72/PgHDLnythN4IJtn66Ed6Xo+/9OUKe1I/dKXctkSGquiIIM9n5l2h/XkAFjPzbUQ0z9+fq+FzMuNi5Khzorgr6Igew3KM7je7LItGkbpgJuciSBO9nQXgAf/1AwAuMfAZqWSVo4sydYVGAiwjtY7Ksxnp2rs3sWijWswZ9Puj+2y5+uHYRPGivWAATxLRKiKa47eNZuZt/uvtAEbHvZGI5hDRSiJa+QH2F+yGYDtxIixz3FEk6REWZSNpRokTYlZJMggfcIvyZgNFBXkuM58B4EIA1xLReeGDzMzwJNoPZp7PzJOZefJBGFSwG/VAZ3pt+l7rfU8c37OpEghx5L1LKynKiCR70bVQRRZJekUafRPFieg+ItpJRC+H2kYQ0VNEtMH/OdxvJyK6i4jaiGgNEZ2h0udCgmTmDv/nTgCPAZgCYAcRjfE7NQbAziKfURZlFGlsGV8sKs8sUoxic7XaFt6/eErsVgc0L5j7QwAzI21BDaQVwGJ/H/CCuFZ/mwPgbpUPyC1IIhpKRIcFrwF8CsDLABYBmO2fNhvAwryf0UyUJU8TkWURYZaN7VFkIxFWJcq8441RdN9qyMzPAHgr0pxUA5kF4EH2WAbgiCCQa0SRCHI0gGeJ6EUAzwH4FTM/AeA2AJ8kog0APuHvCz5xIjQlR/7gQI8Qw68FO1GVn8vRZMYFc0cFdQp/m5N2fSTXQMYC2Bo6r91va0juaT7MvBHAn8W07wYwI8u1uocPxb4ZU3v2h/5sed5u5aLsOZBlp9o6xWhrtDj+0peVo8Otj37Euqk/WaX3/sVTMPiXbk1g9xbMzTQPchcz5773lpmZiGJrIKrU9k6aotLb/fm+q2CPvEfGzgSPlkm9U8K61hf/5Wp7RLj/Y9u1pdklrNKzg4jGMPO2SA2kA8D40Hnj/LaG2DHZyDKickxqE3qxIbI0HRW2TDqhjxyDtiLYLkedeGOQxherSKqBLALwWb+afTaAd0KpeCJWCnLff5+afpLQFAwcN7bPloaqJLMWaxqJMK8km0mOAToXqyCihwEsBfAhImonomuQXAN5HMBGAG0A7gHwtyr9rW2KnReJFO0gSYZBe2d7anakjaJRYh3QkWYH8yB1wcxXJBzqVwPx52Rfm/UzrIwgq0Lk2Jiy0miVSLHRObYVYOIoGj26VqDxKCXF1oodvRAEHxU5qlC2JMuMMm2S4/6PZVvlvNt/sqHKZgNWpthlT/MBJHq0AV1yDMgy9ScJSa97ySrDKDmm+VSOdYKsQo6CHvY9cTyGztyY67265RhQJJLMIkcd033SsClyzIstqbMq1gmyClSiRxvmQXad3/f++pann6+oJ9VTZpHGBuogx+BWQ5ewSpB1jR43PNBXbK2zs4stKsegrS6S7GzvUI4im0WOdZBiFFvGFlWxSpB1IyrGcHseSVbN0JkbjVayVSRZFzkO/uVzsZXsOkoxQPc0nzKwQpAD9uyzOnrMk14nyTF83EVJmiZLJGkLeccf6yzDJGQMsmaYkGNW4tLrOlOXKFGIoLiMmU24pfMQA4YO7bOZwLQcdYu0CC0ntVbdBaHmMNybB+mkIOOEWESScSK0oWptmpaTWnu28H4j8k7jcRGV1LmM6T1Vs/6es7RdS+eCuWXgXIrdSIQDhg5F9759ua7bDEIMIxFjcZpJjuvvOQuTPr+i0LVcLNI4FUGaSqVdYvPFB3vbRYO8KT6Lx3mbRiSK7CVJgs0kR51IBCkYYfPFBwPdBHT3tv3mjdW4YJ0vx8XjgBntStfSET3GTfkxJc7t10+PbT/6O0uMfF4azSrHolGkixPFnYkgXYkeTUzdaXn6eU+OES5Yd7H2z+r5TAWJhoVoQo7br5+eKEeV47roWv9an03ICQOdPEB5swE7elEzypjf2HZlzFMrNafaKgydudGYHE2cKxSjSNqt+7nYZSCCNIRpSRaJHrvWbdDYE/3kEV5Z0aRQDBGkAVxJr6O0zn5emyhP+Gr5VXYXK90iSfPkjSJ1Pxe7DKRI4xB9JFlBOi3kp3PGmbHtAxevim3f/+SEhtcb9KnNBXvUmEmfX2Gkis2WiE8VJyJIVfLOgTSNjfdc255m14kkOQbHosfT5Bico3KebcidNEIscZK0UZw2UGT6TlVTf4oSSNIm6RWdGB6FWcYgK8PW6DGMbUK0OYrMIzpX5RjQOeNMHHvYW1V3Q4nc45BMypsN1GIM0gU5BmiT5Iz2/uOQihPFXeHo7yxRKrq4LsYwHV9vBebuVj5//5MTjI9H6sOeyFAVJwTZvW9fv0p2XFuzck/rAlx5/Q3Ap8diyGPZ1tXsWrfB6mp1neTnIknFmrzpty2RoSpOCBLoK8QgYtQZOe6ZHf9cmuEPWLyIhR8xXvnpG3qa3v301J7XqrJMkqTNKbjg4U706OZiFc4IEtCfSidJMXqO1ZJswMDjxve87tyyteG5UUmKHNNZ//3+j0yIY9IXn2tYybYdbVN+2CvUuIRTgtSJihyj57oqSsCTpYokBf2s//4UTPpi7+MV0uZEds2doL0POxd+uM/+UbP+mOn9QUpddMEKW6bvqNK0ghSEMglLMmlyeMCgT21Wmu5TJL3eufDDmSUJFJv6w5AxSCfIEj1G3+dyFClUSzSSbEQgv423T8Pxc5f2CDOLFKNRYxSe/mf92mjJi8rXz457VezazIPMQt0kl7VyLdjPxtunYePt03ped7wwRlmOOxd+OFWOQwcdiG2Pk6ZOmNU3G2hKQTYD7x4lX61pVKPBrARiVG0PkybGgCG3HJapT7rQPVGciL5CRGuJ6GUiepiIBhPRRCJaTkRtRPQTIjo4b3+dTrHbb2w8iXjcPyfPoRv+wNLMqbbNkeeQx5b3meJz9BP1mjRuK5O++JxyNVuFNAkGKXcRvOhxUKFr5MGLDPWl2EQ0FsCXAJzMzO8R0SMALgdwEYA7mXkBEX0fwDUAYhZQTcdJQaaJMXyeLknaLMeAINUOT+8JSKtgm6T73NMaHh/w7OqSemIGVUnqjjhPWjUQ687szPSeoYMOVBY9AkbmQQ4EcAgRfQBgCIBtAP4cwGf84w8AuBnNIkhVOYbPT5NkQFSWLkjRVtKkGD23DpIE1OdGFuWkVQP7/FQRZdVyBDKPLY4iopWh/fnMPL/3WtxBRP8G4HUA7wF4EsAqAG8zc/AX0g5gbN7+OidIk9RFiJ1btsZGkabJIsW497ouyUaoRI8qY4xArxTj2k5CG37ffmLs+1TkaLaKnTnF3sXMk5MOEtFwALMATATwNoCfAphZqIMRUkfyieg+ItpJRC+H2kYQ0VNEtMH/OdxvJyK6yx8cXUNEZ+jsbNboUSiPInKsE3EiNFXMSeKoWX/sN8cxbc4jLXnRvByhXqBRFOknAGxi5jeZ+QMAPwdwDoAjiCj4LTIOQEfePqtEkD8E8F0AD4ba5gFYzMy3EdE8f38ugAsBtPrbVHh5/1RUSKP0WhBMULYQk8gzEdw0mmfvvA7gbCIaAi/FngFgJYCnAVwKYAGA2QAW5v2A1AiSmZ8BEF2kbha8wU/4Py8JtT/IHsvgmXxM3s4VReToYbpAI9GjoATrnebDzMsBPArgeQAvwfPZfHjB2t8TURuAkQDuzdvlvGOQo5l5m/96O4DR/uuxAML/G4MB0m2IQERzAMwBgMEYkrMb8YgY3aPO449CLxzzfPdC12P+JoBvRpo3AtBSLStcpGFmJqLMkbNfjZoPAMNohLbIW+ToUeW0HhsY+YfhffZ3n7Onop4IYWy5Q0aVvILcQURjmHmbn0Lv9Ns7AITLp4UGSLMicnQTndFjVIzRdtdFefzcpUBMFdsFXFysIu/9aIvgDX4CfQdBFwH4rF/NPhvAO6FUvDDj/nlJPwkGbSLH6sgruAHPrtYmx5F/GJ4ox+h5tlL0Dhmg8XxI01XqVBgAk/pmAam/iojoYQAfhzdpsx1evn8bgEeI6BoAWwBc5p/+OLzbfNoAvAvgagN9FhlayIBnVysXa3SPN9osvawcP3dp4nzINIGqTBanJS8aX5CiEbVLsZn5ioRDM2LOZQDXFu2U4CZh8cXJ0kQhJo8cR/5huHOpdliO687s7DdZPOsth2GKRpYDBg/us9/9/vvJJ9dNkIKQhzKq0nWKHAFgwKneSjwnPhQjb/9Y95ricxt1CzHueLwk7XmcqyqyJlbFbPhupfPonaWoHG2TayBH1fOKRIx5GTB4cKocU+EMmwWIICtE5CgA6nKMnh9IsgpZ5kLzRPEyEEFWRFiOIsps2Bb9VUGVkWQhHIsgZQxSEIRUCqfWPdgRGaoigiwAnXlKbDuvWpv5Whu+OxWt18mzZcrCtSp2HANO/bCWok3q52iTI6yJDFURQWYkSYrRc/JIUlBj9zl7CqXZdZCjzcg0nyZFRY7hc0WS5sgjSZ1i3PZ/T+rXNuaSdbmu1b3mj5kLNbaSKkdLii+qiCArQIoyeig7EoyTYtzxPKKskyQbUbs7aQSPLNGjUB/SpNjoPVlFGR1PTBNmGeOPKjSMGqOIIAWhHuSRo05sEWAjMskRcC7FlnmQhpDxR6FqwZqk+/33s8sRALH6ZgMiSEVcE96WW6Zhyy1qT8kThEYEMgxvucgySVwEWV+qlmlYjCLJ6shb1a4vGdaCtCQVlzFIzdgkR0GwDksiQ1VEkBngVWsbVrNtleOWW6bhuG8UX6262YhGgKpjihI5NkAE6QbtN00HAIz7VrbVyQMJBqLUJcWitxlK5GieQHyNRDnmknVYf/+ZmHT1Kqy//0wAwKSrV5XSPycQQdpPIMfw67yiFJqPRhFiIMXgZ/h104vSwTtppEjjE5ama0j0WD3r7z+zjxSTzml2ZJqPw7TfNL0SUcoqPs1D00vSsWk+TZli1wmJHgWXsCUyVEUiyBjKjCLLiB6lgi1Yg2PzIEWQDiPRo5s0bZrt4J00kmIn0H7T9MyVbVVar1te2griLkWP27/SP3I/+k4z34FuwtN6hAZYIj5VmjKCVBWfyVRb5j32sv0r02Pl6BqTrl6lNJWnmaf76K5iE9ERRPQoEf2RiNYR0TQiGkFETxHRBv9n7uXnJYIUKkNFitu/Mt1IFNn2o9N7Xp941Qtarx0VYDiybGY5AjARQX4HwBPMfCkRHQxgCICbACxm5tuIaB6AeQDm5rl40wpy3LeWKEWIJlNt06im169/o/fv4dhbyvmzVhkxhuVYBk0vxTAaBUlEhwM4D8DnAICZDwA4QESzAHzcP+0BAL9DTkE2ZYot9BKWY9y+bqpMp9t+dHqsHMsWZrOSJb1WTLEnAngTwP1E9AIR/ScRDQUwmpm3+edsBzA6b5+dFGTL8OE9WxFsGIvMg67xxzgZmoogbRRj1XTMnY6OuXb92zJOtmk+o4hoZWibE7naQABnALibmU8HsA9eOt37ccyFauJNm2IHqKbarpGWXpuOFG3BRjEC6CPG8Ouxt7s5nKNMNlXtYubJDY63A2hn5qDi+Sg8Qe4gojHMvI2IxgDYmauvcDCCjEaNRaNIIPtCFS7z+jemly7HKiLHLFGj7iJNGo2ixrpHlDpTbGbeDmArEX3Ib5oB4BUAiwDM9ttmA1iYt79ORZA6ZJhEnSLJuOhRRYom0uuicsxTwbY1agTUBNgxd3p9I0n9Vey/A/CQX8HeCOBqeIHfI0R0DYAtAC7Le3GnBJlEy/Dh6NpT/BnJQSRZF1EC6qm0jXLMQ1Y5lh09qlJLSRpYpYeZVwOIS8Nn6Lh+LQSpG1dT7mjkWFXUWBW2ptR5qeXYpGN30oggBSvJmlq7Isexty+p/ThjQ0SQ5ujasyd2HFJHeu06dYoeTdw5k1eMhz87sl/bO+fuLtqdzNQl5ZblzgwTlWEzyjEqw6wLUhx7y5JS5Hj0nUsyyS7r+WGSBHjiVS9olWOjdlXyiq6pI8+KcCqCDGhGKUbJu0qPjVGjrojxxKte6Em1i6bSRSWYRtOm2o5FkE4KUshHVXIsc8kyHWOMKnI8/NmRhVPtppOkRc+aUSU1xSai+4hoJxG9HGq7mYg6iGi1v10UOnYjEbUR0atEdIGpjgvZsDFytBHTkWOUrOm280J1bMFclTHIHwKYGdN+JzOf5m+PAwARnQzgcgCn+O/5DyJq0dVZQZ2wEEWO6Rz+7MjS5Rgw9vYltSjAKOGYIFNTbGZ+hogmKF5vFoAFzLwfwCYiagMwBYA7y1rXCBGjGlWJsTQGKMYo3V1Gu0GoYYrdgOuIaI2fggdzb8YC2Bo6p91v6wcRzQlW6fgA+wt0QxDqgUokmSnSHNCiLsc85+fBsQgyryDvBnACgNMAbANwR9YLMPN8Zp7MzJMPwqCc3RCE+qEl3S4iOlOSZIC61TcbyFXFZuYdwWsiugfAL/3dDgDjQ6eO89sEoVaYniweSDIoypQ+RjmgxUzKbUlkqEouQQZrrfm7nwYQVLgXAfgxEX0bwDEAWgE8V7iXgtCk5BKj6TS5AK6NQaYKkogehvd8h1FE1A7gmwA+TkSnwft9sBnAFwCAmdcS0SPw1mTrBHAtM5sd+RWEkqniVsNKMBFF1k2QzHxFTPO9Dc6/FcCtRTolCLbSNHI0gUXFF1XkThqh6Xnn3N2pU31EjHqoXYotCM1AkiRFjJoRQQq28qf/eXa/tsN+skzpvTuv9aqpR32vvpPPRYbmcS2CdG65MyEfcXJUJZBj9LVgKboKK6am+Tg0UVwiyAK8dXXy86lH3G/P3ZW65CgIhbBIfKqIIHPQSIzhc2yQpMixSenuKjYf0kD0SP7mEpJiG0RFpC4i4qw5JhetkBRbsAWJHpucQHRZIknTK/pYIj5VRJCGsSXV1kmdK9lptJzyoX5tXWtfraAnGVBJtw2LsQcRZL1xJW2W6FE/cXIMt1styrIEmIZjgpQxyIzkiQbLlqqqHOPmQKbJsZmjxzSSBCr4+M+kUd1sQAQp9CCRY3FEkilIkUZwAdU7aMJI9Ni8fPCpyT2vD3pyZe7r2BIZqiIRZA5sLrrkHXuU1FpIIizHuP1MOBZBiiAFQTN1SrMLyTAG18YgJcXOyYj7lyoXX2yLOKPpteno8fWb+17/2JvdjEa71r5aK/k1QrcYARiJDP3HSq8E0MHMFxPRRAALAIwEsArAVcx8IO/1JYJsMrKOPRaR4+s3T+8nR9dRncrjskiNyDFAf4p9PYB1of3bAdzJzCcC2APgmiLdFUEWQCUytC16zEJROQpCmOC52LpSbCIaB+AvAPynv08A/hzAo/4pDwC4pEifJcUuSCDAaLptoxizRI955agixtdvnq6cZm/9et/rjf+n6tPzOqfaRqNHIGuKPYqIwiXz+cw8P7T/7wD+AcBh/v5IAG8zc6e/3w5gbM6eAhBBasNGIYZpJMejvrekzzikSTlmISrHoE2HJNv+vbfaf+KXs0950inJbTf0/3OOuaP8XwSqcsw9zYcB6s5kyF3MHNspIroYwE5mXkVEH8/XoXREkAIA/YWYIsSJUSdhOQb7VUsyyrYbplciSdNorE6fA+C/EdFFAAYDGAbgOwCOIKKBfhQ5DkBHkQ+RMciaERcp5pkUnoWscmyUXpuWo26SijY67suOiyxNYTx6DNBUpGHmG5l5HDNPAHA5gP/HzFcCeBrApf5pswEsLNJdEWTNsU2OrtJy4kS0nDgx9lhUhlYvWlGAwnJEKfMg5wL4eyJqgzcmmfiIahUkxa4hpqVoCpXosej4YzS9DrcnpdlhMbacOBFdbZv6nWNKirVLtQ1MAGfm3wH4nf96I4Apuq4tEaRQKq5OEg+TFEmaoAw5Gq9cB8hqPoKQTB3kGFCmJG1AR3oNQO7FFoQ46iRH3dieQuuSo+6J4mUgghSMU1c56owio5Icc8cSK8SpLXIMYFbfLECKNIJRdMrRhrtoTFKVEA96cmXsOKR2OcKeyFAVEWTF/Nf/mNrz+tCfLq+wJ3rRHTXaKsekqrZrhCVpQowArBpbVEUEWSFhOQb7rkrSZBqdBxLPAAANhUlEQVRdlhzz3lFTF4yJMQR1G/8IrYggLcM1SepOoW1cnMIUQ585ss/+vvPerKgnJeJYBClFmoqIRo+Ch4tC1FWsiQqzjrhWxZYI0kJciyJ1Y5sku9o2aZ/3mCTDoc8cWd9IkmFNdVoViSCFpqKZxxhtQCJIQXAUkWcJWCI+VSSCFAQFyprKU9v0Gm7eSSMRZAM23NVbSGn9Urljgs0+DmkjJsYiA6oW42t39F3l6IQbDETTFt0ho4pEkIqEZSm4TZFUWlckue+8N/tsthEVpi5ciyBTBUlE44noaSJ6hYjWEtH1fvsIInqKiDb4P4f77UREdxFRGxGtIaIzTP8hTCBCrDdRSWaRZlSSrt9JY0qGsTi2mo9Kit0J4AZmfp6IDgOwioieAvA5AIuZ+TYimgdgHrzVfC8E0OpvUwHc7f90ng13TS091RbMYUMkWTWlyhH2RIaqpEaQzLyNmZ/3X/8J3kO6xwKYBe+5s0Df58/OAvAgeyyD9xCdMdp7LghCIRrJscgY5KYFp2LTglP7H2AA3ay+WUCmMUgimgDgdADLAYxm5m3+oe0ARvuvxwLYGnpb4WfT2oSk3kIdKDty7KGGKTYAgIgOBfAzAF9m5r1E1HOMmZkoW/BMRHMAzAGAwRiS5a2CI+z6wrR+baN+YPfzw5uByuSIzM/FrhylCJKIDoInx4eY+ed+844gdfZ/7vTbOwCMD7099tm0zDyfmScz8+SDMChv/2uNy/drx8lRcAMjU3x8XKtip0aQ5IWK9wJYx8zfDh1aBO+5s7eh7/NnFwG4jogWwCvOvBNKxQWfQ3+6vDQBvnNlfMRw+ENm/iM0kuOuL0yTKLJCqowebUqdVVFJsc8BcBWAl4hotd92EzwxPkJE1wDYAuAy/9jjAC4C0AbgXQBXa+2xkIkkOYaP6RSlRI7NTWxxxse7k8YtQ6YKkpmfhfdni2NGzPkM4NqC/RI00EiO0fNMRZOCW5hMrwEAji2YK3fSCADUZdoIiR6FNIhZebMBEWSFpN1rXfa92DokKdhNUoR4wg3LzEePWab42OFHWaxCJ3TWR/vs84qXUt9TZrFGyMf+i85qeHzQ4ytK6okejIswEVmsQtBEXVfyqWMFO02gRWldMahnc53aTfMR1IhGj1kIZBhEknWVo4uoym//RWdpjyR1CLFl1MjY9q5duwtfOxcSQTYfReQY5tCfLtcqR9sq065Fj6Yjw0YkyVFXFJkkTqOw99hX1c0GRJA1xzZJ1hldQi1LgtVIktW3FLIuxZgHEWRGosud6YoeTaIqSZMydS16dJ1K5KeC3ip2sBTjyQDOBnAtEZ0Mb+nFxczcCmCxv58LEaRgHJGjOq4XYiZevqbhcZ3zIHMsxZgZEWQGXF4sNy061BE9NrsIixZpVOS44az9hT6jcrKl2KOIaGVom5N0WcWlGDMjVewEWr+0vHZrPx7+0LJ+k8F1p9WjfrC0546aMoT52kOnAwBOuPIF7dce9PiKSgs1eVFNr01VsoMo8rXoAUbWWw13MfPktJN0L8UYRgTZgKIRI531UaXJ4mVSRtHGpBgDIca1VyFJHVN7VNNqndFjFdN8CPpvIWy0FCMzb4ssxZgZEaTgDElyDB83JUmgt0pt+50zXbt25yvSnB2zEs+yxmOKmdEoyBxLMWZGBCk4QZocy8CEGKuIHmOJk6MJ9EaQWZdizIwIsoZs/idvDHDC1+tRNLFBjnWiX3pdmhyhdbmzrEsx5kGq2IYpe55kIMfgdXjfReosRyuix7Lk6CPLnQnW4bokBX30iR5LliMArXfSlIEIskkQSbpJ3ugxrkpduRyRQY6WCFLGIAugkj7bNs2nzpioYFeJ69N6+sGwRnyqSATZRLgWRdZ5/DENHXLs2rW7Z+uDavSoe4oPAOpi5c0GRJBNhmuSVKVO0WOdijL9kBRbEATnMRA9etN87BCfKhJBFiBtfNHW8UdXokjVqNDV6HHDWfv7RIvRfe1UmFp7SJFGEEqjSjG+fZX3S+aIHxWfjO/8Cj1ZsER8qjgVQXb8/BR0/PyUqrvREF7xUs9WBap3z7gSRcZxwpUvVCbHt6+a1iPHYN8JKo8efSSC1I/NUuQVL1m5ak9dsCl9TpLh21dN0xJJ1h4Zg9RPnBxtE6ZtcqzLPdg24UykWATT0SMY4G71zQKsFqRtInQJkWS5WC1QlfTauBx9HEuxrRWkyNE8Lo9DlonV8tNBaXKEl2KrbhZgrSDTEIEKggbKkmOARJDFEfnpQdJsoSFlyxEQQRZFVY5j/2qt4Z7UA5FkMWqRXseJsAo5ykTxYkjkaIYJX18q442GsX6aTyVCjMAAuu2oTqtiTQSZRY4SPWYnLpJstuhy1y8mYdcvJlXdjeZGIkjBVppNiAFRKe76xSSM+sv1Su894kdLU9Ns66NHm7BEfKqIIAXjdH/MW9dxwO/LvSvG5Whx/Q96n8U96Qt2P2ZWHXum76gighSMEEgxrs20KHWLMSmKNBU5huUY7JcpyZbRRwEAunbs1HthBtiSO2RUcU6QMv5oP3FyjB43JUlTUaPJNDoswKgc484xSSDH4LV2SUoEaQ6Ro/2kydEkrqXUYRkmibEswmKMtmuVpGNjkKlVbCIaT0RPE9ErRLSWiK73228mog4iWu1vF4XecyMRtRHRq0R0gUpHRH7uU6UcXaNqIYZJkqPqcWWYvWk+qpsFqESQnQBuYObniegwAKuI6Cn/2J3M/G/hk4noZACXAzgFwDEAfktEk5i5K+2Dxv7V2sTpPrYKdNyyQwEA7Wf/V8U9aW6yRI+qFew8dP722J7XAz/xurHPcRbHIshUQTLzNgDb/Nd/IqJ1AMY2eMssAAuYeT+ATUTUBmAKAKVBHFtFGCUQY3S/WUUp0WNfOQb7cZJ0KXrUDVsSGaqSaaI4EU0AcDqA5X7TdUS0hojuI6LhfttYAFtDb2tHY6E6xbhlh/aTY7NTtRxtiB6jckxrt4Gy5VjrWw2J6FAAPwPwZWbeS0R3A/hHeDcQ/SOAOwD8dYbrzQEwBwAGY0iWPleCSNFtdIpxx99N77M/+v8sUX6vTdFj6TCArtSRNqtQEiQRHQRPjg8x888BgJl3hI7fA+CX/m4HgPGht4/z2/rAzPMBzAeAYTTCjl8XCYgck8kTPZY9YVwXUTGG20eiveTeFKP86NHzIzs2zUelik0A7gWwjpm/HWofEzrt0wBe9l8vAnA5EQ0iookAWgE8p6/L5SJytBuV9FpH9Jgkx4AnTnqs8Gc0QuccyKxy1DbNh/U/coGIZvqzZdqIaJ6ejvaiEkGeA+AqAC8R0Wq/7SYAVxDRafB+MWwG8AUAYOa1RPQIgFfgVcCvValgC81B2dGjyYq1kB2dESQRtQD4HoBPwqt1rCCiRcz8iq7PUKliPwuAYg493uA9twK4tUC/nKRZK9iqmJDjqL9c79wE8aZG762GUwC0MfNGACCiBfBm0WgTJLEF1SIiehPAPgC7qu5LAUZB+l8l0v9qiev/ccx8ZLBDRE/456kyGMD7of35fu0iuN6lAGYy8//y968CMJWZr8va+SSsuNWQmY8kopXMPLnqvuRF+l8t0v9qUek/M88sqz+6sGbBXEEQhIwozZgpgghSEARXWQGglYgmEtHB8G5xXqTzA6xIsX3mp59iNdL/apH+V0vp/WfmTiK6DsBvALQAuI+Ztd6rbEWRRhAEwUYkxRYEQUhABCkIgpBA5YI0fauQCYhoMxG95C8UvNJvG0FETxHRBv/n8LTrlIW/2tJOIno51BbbX/K4y/8+1hDRGdX1vKevcf3XumCzSRosOu3Ed1DWotlWwsyVbfAGVl8DcDyAgwG8CODkKvuk2O/NAEZF2v4FwDz/9TwAt1fdz1DfzgNwBoCX0/oL4CIAv4Z399TZAJZb2v+bAfzvmHNP9v8dDQIw0f/31VJx/8cAOMN/fRiA9X4/nfgOGvTfme8g71Z1BNlzqxAzHwAQ3CrkIrMAPOC/fgDAJRX2pQ/M/AyAtyLNSf2dBeBB9lgG4IjIwiSlk9D/JHoWbGbmTQCCBZsrg5m3MfPz/us/AQgWnXbiO2jQ/ySs+w7yUrUgXV1clwE8SUSr/HUtAWA0e6uvA8B2AKOr6ZoySf116TtxbsHmyKLTzn0HzbZodtWCdJVzmfkMABcCuJaIzgsfZC/PcGb+lGv99bkbwAkAToP3SJA7qu1OOtFFp8PHXPgOYvrv3HeQlaoFafxWIRMwc4f/cyeAx+ClDzuCNMj/qfmBwtpJ6q8T3wkz72DmLvaeRH8PelM4K/sft+g0HPoOkhbNduk7yEPVgjR+q5BuiGgoeU93BBENBfApeIsFLwIw2z9tNoCF1fRQmaT+LgLwWb+SejaAd0JpoDWQQws2E8UvOg1HvoOk/rv0HeSm6ioRvIrdeniVrq9V3R+F/h4Pr0L3IoC1QZ8BjASwGMAGAL8FMKLqvob6/DC8FOgDeONB1yT1F17l9Hv+9/ESgMmW9v9Hfv/WwPsPOSZ0/tf8/r8K4EIL+n8uvPR5DYDV/naRK99Bg/478x3k3eRWQ0EQhASqTrEFQRCsRQQpCIKQgAhSEAQhARGkIAhCAiJIQRCEBESQgiAICYggBUEQEvj//9thJWo8K6QAAAAASUVORK5CYII=\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "name": "stderr", | |
| "output_type": "stream", | |
| "text": [ | |
| " 10%|█ | 1/10 [00:00<00:04, 1.84it/s]" | |
| ] | |
| }, | |
| { | |
| "ename": "ValueError", | |
| "evalue": "Cannot feed value of shape (603, 256, 256) for Tensor 'conv2d_19_target:0', which has shape '(?, ?, ?, ?)'", | |
| "output_type": "error", | |
| "traceback": [ | |
| "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", | |
| "\u001b[0;31mValueError\u001b[0m Traceback (most recent call last)", | |
| "\u001b[0;32m<ipython-input-8-f76d8214d268>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 21\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 22\u001b[0m \u001b[0;31m# results = model.fit_generator(generator=training_generator, use_multiprocessing=True, workers=8, callbacks=[earlystopper, checkpointer])\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 23\u001b[0;31m \u001b[0mresults\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit_generator\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mgenerator\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtraining_generator\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcallbacks\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mearlystopper\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcheckpointer\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m", | |
| "\u001b[0;32m~/.conda/envs/yee/lib/python3.6/site-packages/keras/legacy/interfaces.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m 89\u001b[0m warnings.warn('Update your `' + object_name +\n\u001b[1;32m 90\u001b[0m '` call to the Keras 2 API: ' + signature, stacklevel=2)\n\u001b[0;32m---> 91\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 92\u001b[0m \u001b[0mwrapper\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_original_function\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 93\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mwrapper\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", | |
| "\u001b[0;32m~/.conda/envs/yee/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mfit_generator\u001b[0;34m(self, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)\u001b[0m\n\u001b[1;32m 2222\u001b[0m outs = self.train_on_batch(x, y,\n\u001b[1;32m 2223\u001b[0m \u001b[0msample_weight\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msample_weight\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2224\u001b[0;31m class_weight=class_weight)\n\u001b[0m\u001b[1;32m 2225\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2226\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", | |
| "\u001b[0;32m~/.conda/envs/yee/lib/python3.6/site-packages/keras/engine/training.py\u001b[0m in \u001b[0;36mtrain_on_batch\u001b[0;34m(self, x, y, sample_weight, class_weight)\u001b[0m\n\u001b[1;32m 1881\u001b[0m \u001b[0mins\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mx\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0my\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0msample_weights\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1882\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_make_train_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1883\u001b[0;31m \u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mins\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1884\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1885\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", | |
| "\u001b[0;32m~/.conda/envs/yee/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m 2476\u001b[0m \u001b[0msession\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mget_session\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2477\u001b[0m updated = session.run(fetches=fetches, feed_dict=feed_dict,\n\u001b[0;32m-> 2478\u001b[0;31m **self.session_kwargs)\n\u001b[0m\u001b[1;32m 2479\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mupdated\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2480\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n", | |
| "\u001b[0;32m~/.conda/envs/yee/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36mrun\u001b[0;34m(self, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m 903\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 904\u001b[0m result = self._run(None, fetches, feed_dict, options_ptr,\n\u001b[0;32m--> 905\u001b[0;31m run_metadata_ptr)\n\u001b[0m\u001b[1;32m 906\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 907\u001b[0m \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", | |
| "\u001b[0;32m~/.conda/envs/yee/lib/python3.6/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m_run\u001b[0;34m(self, handle, fetches, feed_dict, options, run_metadata)\u001b[0m\n\u001b[1;32m 1114\u001b[0m \u001b[0;34m'which has shape %r'\u001b[0m \u001b[0;34m%\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1115\u001b[0m (np_val.shape, subfeed_t.name,\n\u001b[0;32m-> 1116\u001b[0;31m str(subfeed_t.get_shape())))\n\u001b[0m\u001b[1;32m 1117\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgraph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mis_feedable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msubfeed_t\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1118\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'Tensor %s may not be fed.'\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0msubfeed_t\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", | |
| "\u001b[0;31mValueError\u001b[0m: Cannot feed value of shape (603, 256, 256) for Tensor 'conv2d_19_target:0', which has shape '(?, ?, ?, ?)'" | |
| ] | |
| }, | |
| { | |
| "name": "stderr", | |
| "output_type": "stream", | |
| "text": [ | |
| "603it [03:16, 3.06it/s] \n" | |
| ] | |
| }, | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "[[[ 0. 0. 57.]\n", | |
| " [ 0. 0. 22.]\n", | |
| " [ 0. 0. 14.]\n", | |
| " ...\n", | |
| " [ 0. 0. 12.]\n", | |
| " [ 0. 0. 11.]\n", | |
| " [ 0. 0. 13.]]\n", | |
| "\n", | |
| " [[ 0. 0. 77.]\n", | |
| " [ 0. 0. 29.]\n", | |
| " [ 0. 0. 16.]\n", | |
| " ...\n", | |
| " [ 0. 0. 13.]\n", | |
| " [ 0. 0. 12.]\n", | |
| " [ 0. 0. 13.]]\n", | |
| "\n", | |
| " [[ 0. 0. 89.]\n", | |
| " [ 0. 0. 34.]\n", | |
| " [ 0. 0. 17.]\n", | |
| " ...\n", | |
| " [ 0. 0. 12.]\n", | |
| " [ 0. 0. 12.]\n", | |
| " [ 0. 0. 11.]]\n", | |
| "\n", | |
| " ...\n", | |
| "\n", | |
| " [[ 0. 0. 11.]\n", | |
| " [ 0. 0. 12.]\n", | |
| " [ 0. 0. 11.]\n", | |
| " ...\n", | |
| " [ 0. 0. 12.]\n", | |
| " [ 0. 0. 13.]\n", | |
| " [ 0. 0. 13.]]\n", | |
| "\n", | |
| " [[ 0. 0. 13.]\n", | |
| " [ 0. 0. 12.]\n", | |
| " [ 0. 0. 11.]\n", | |
| " ...\n", | |
| " [ 0. 0. 14.]\n", | |
| " [ 0. 0. 13.]\n", | |
| " [ 0. 0. 14.]]\n", | |
| "\n", | |
| " [[ 0. 0. 12.]\n", | |
| " [ 0. 0. 12.]\n", | |
| " [ 0. 0. 10.]\n", | |
| " ...\n", | |
| " [ 0. 0. 14.]\n", | |
| " [ 0. 0. 14.]\n", | |
| " [ 0. 0. 13.]]]\n" | |
| ] | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<Figure size 432x288 with 2 Axes>" | |
| ] | |
| }, | |
| "metadata": {}, | |
| "output_type": "display_data" | |
| } | |
| ], | |
| "source": [ | |
| "# Fit model\n", | |
| "earlystopper = EarlyStopping(patience=20, verbose=1)\n", | |
| "checkpointer = ModelCheckpoint('model-dsbowl2018-unet.h5', verbose=1, save_best_only=True)\n", | |
| "\n", | |
| "# partition\n", | |
| "random.shuffle(train_ids)\n", | |
| "train_ids_ = train_ids[:int(len(train_ids) * 0.9)]\n", | |
| "validation_ids_ = train_ids[int(len(train_ids) * 0.9):]\n", | |
| "\n", | |
| "# Parameters\n", | |
| "params = {'IMG_WIDTH': IMG_WIDTH,\n", | |
| " 'IMG_HEIGHT': IMG_HEIGHT,\n", | |
| " 'IMG_CHANNELS': IMG_CHANNELS,\n", | |
| " 'batch_size': 10,\n", | |
| " 'training': True}\n", | |
| "\n", | |
| "# Generators\n", | |
| "training_generator = data_generator(train_ids_, **params)\n", | |
| "params['training'] = False\n", | |
| "validation_generator = data_generator(validation_ids_, **params)\n", | |
| "\n", | |
| "# results = model.fit_generator(generator=training_generator, use_multiprocessing=True, workers=8, callbacks=[earlystopper, checkpointer])\n", | |
| "results = model.fit_generator(generator=training_generator, callbacks=[earlystopper, checkpointer])" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "1f381f5b-1b71-4daa-a417-e02f4894540b", | |
| "_uuid": "bb15226ea617cf91ed8f43179fccb5a15809e5a0" | |
| }, | |
| "source": [ | |
| "All right, looks good! Loss seems to be a bit erratic, though. I'll leave it to you to improve the model architecture and parameters! \n", | |
| "\n", | |
| "# Make predictions\n", | |
| "\n", | |
| "Let's make predictions both on the test set, the val set and the train set (as a sanity check). Remember to load the best saved model if you've used early stopping and checkpointing." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": null, | |
| "metadata": { | |
| "_cell_guid": "2daa48d5-ac98-4e18-af3f-a582baaa44f0", | |
| "_uuid": "f841760b4abca1a25cb750822f88268bd79bf2ce" | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "# Predict on train, val and test\n", | |
| "model = load_model('model-dsbowl2018-unet.h5', custom_objects={'mean_iou': mean_iou_unet})\n", | |
| "# preds_train = model.predict(X[:int(X.shape[0]*0.9)], verbose=1)\n", | |
| "# preds_val = model.predict(X[int(X.shape[0]*0.9):], verbose=1)\n", | |
| "preds_test = model.predict(X_test, verbose=1)\n", | |
| "\n", | |
| "# Threshold predictions\n", | |
| "# preds_train_t = (preds_train > 0.5).astype(np.uint8)\n", | |
| "# preds_val_t = (preds_val > 0.5).astype(np.uint8)\n", | |
| "preds_test_t = (preds_test > 0.5).astype(np.uint8)\n", | |
| "\n", | |
| "# Create list of upsampled test masks\n", | |
| "preds_test_upsampled = []\n", | |
| "for i in range(len(preds_test)):\n", | |
| " preds_test_upsampled.append(resize(np.squeeze(preds_test[i]), \n", | |
| " (sizes_test[i][0], sizes_test[i][1]), \n", | |
| " mode='constant', preserve_range=True))" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": null, | |
| "metadata": { | |
| "_cell_guid": "649248cd-a1fb-4da6-ade2-4bebad44bcab", | |
| "_uuid": "7e06242a50870e07a080064a4912b761775990fa" | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "# # Perform a sanity check on some random training samples\n", | |
| "# ix = random.randint(0, len(preds_train_t))\n", | |
| "# imshow(X_train[ix])\n", | |
| "# plt.show()\n", | |
| "# imshow(np.squeeze(Y_train[ix]))\n", | |
| "# plt.show()\n", | |
| "# imshow(np.squeeze(preds_train_t[ix]))\n", | |
| "# plt.show()" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "af602aea-5e56-42a8-9331-54b4b2650593", | |
| "_uuid": "5fcee2b9aee2fba5c60d43ad48a14139e9c1318c" | |
| }, | |
| "source": [ | |
| "The model is at least able to fit to the training data! Certainly a lot of room for improvement even here, but a decent start. How about the validation data?" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": null, | |
| "metadata": { | |
| "_cell_guid": "4f66b75c-c694-41a1-8c91-34bb6595837b", | |
| "_uuid": "d4ccbb559375bc2777ffb692a20adc313159f2cc" | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "# # Perform a sanity check on some random validation samples\n", | |
| "# ix = random.randint(0, len(preds_val_t))\n", | |
| "# imshow(X_train[int(X_train.shape[0]*0.9):][ix])\n", | |
| "# plt.show()\n", | |
| "# imshow(np.squeeze(Y_train[int(Y_train.shape[0]*0.9):][ix]))\n", | |
| "# plt.show()\n", | |
| "# imshow(np.squeeze(preds_val_t[ix]))\n", | |
| "# plt.show()" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "a6690535-b2e4-49ac-98d9-7191bfabfb6f", | |
| "_uuid": "6a34c98de7c6ae473f676a34fe7e099b46764eca" | |
| }, | |
| "source": [ | |
| "Not too shabby! Definitely needs some more training and tweaking.\n", | |
| "\n", | |
| "# Encode and submit our results\n", | |
| "\n", | |
| "Now it's time to submit our results. I've stolen [this](https://www.kaggle.com/rakhlin/fast-run-length-encoding-python) excellent implementation of run-length encoding." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": null, | |
| "metadata": { | |
| "_cell_guid": "59a0af60-a7d7-41ef-a6fe-9e3c72defa07", | |
| "_uuid": "4f99c1bf852e82b60bd4f982ca0df293f712cdf0" | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "# Run-length encoding stolen from https://www.kaggle.com/rakhlin/fast-run-length-encoding-python\n", | |
| "def rle_encoding(x):\n", | |
| " dots = np.where(x.T.flatten() == 1)[0]\n", | |
| " run_lengths = []\n", | |
| " prev = -2\n", | |
| " for b in dots:\n", | |
| " if (b>prev+1): run_lengths.extend((b + 1, 0))\n", | |
| " run_lengths[-1] += 1\n", | |
| " prev = b\n", | |
| " return run_lengths\n", | |
| "\n", | |
| "def prob_to_rles(x, cutoff=0.5):\n", | |
| " lab_img = label(x > cutoff)\n", | |
| " for i in range(1, lab_img.max() + 1):\n", | |
| " yield rle_encoding(lab_img == i)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "31133f8c-3f40-4dff-8e1d-898d56672332", | |
| "_uuid": "2e07f6afc4787b068ba714428145dcb3951d718f" | |
| }, | |
| "source": [ | |
| "Let's iterate over the test IDs and generate run-length encodings for each seperate mask identified by skimage ..." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": null, | |
| "metadata": { | |
| "_cell_guid": "22fe24a1-7659-4cc9-9d23-211f38e5b99f", | |
| "_uuid": "089587843ed6a3955fdcb9b23a6ec3bf5d703688" | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "new_test_ids = []\n", | |
| "rles = []\n", | |
| "for n, id_ in enumerate(test_ids):\n", | |
| " rle = list(prob_to_rles(preds_test_upsampled[n]))\n", | |
| " rles.extend(rle)\n", | |
| " new_test_ids.extend([id_] * len(rle))" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "20b6b627-0fd6-425d-888f-da7f39efb124", | |
| "_uuid": "849184a40a2c9c21506d8b8eb10ad9155fa229e8" | |
| }, | |
| "source": [ | |
| "... and then finally create our submission!" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": null, | |
| "metadata": { | |
| "_cell_guid": "1ba0ee3a-cca0-4349-83f6-09a1ac6fcb44", | |
| "_uuid": "ba589f56f5be1e6886bc88f5bf9e7d0a408e4048" | |
| }, | |
| "outputs": [], | |
| "source": [ | |
| "# Create submission DataFrame\n", | |
| "sub = pd.DataFrame()\n", | |
| "sub['ImageId'] = new_test_ids\n", | |
| "sub['EncodedPixels'] = pd.Series(rles).apply(lambda x: ' '.join(str(y) for y in x))\n", | |
| "sub.to_csv('sub-dsbowl2018-1.csv', index=False)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": { | |
| "_cell_guid": "222475b9-3171-461a-90f0-a820a6bd2634", | |
| "_uuid": "fb5e6f8cca872f1bd7036f6d9ac2ed2cab615536", | |
| "collapsed": true | |
| }, | |
| "source": [ | |
| "This scored 0.233 on the LB for me. That was with version 2 of this notebook; be aware that the results from the neural network are extremely erratic and vary greatly from run to run (version 3 is significantly worse, for example). Version 7 scores 0.277!\n", | |
| "\n", | |
| "You should easily be able to stabilize and improve the results just by changing a few parameters, tweaking the architecture a little bit and training longer with early stopping.\n", | |
| "\n", | |
| "**Have fun!**\n", | |
| "\n", | |
| "LB score history:\n", | |
| "- Version 7: 0.277 LB" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": null, | |
| "metadata": { | |
| "_cell_guid": "3f5e5a47-6133-4870-976a-a8e4fa7bf46c", | |
| "_uuid": "2a83eab66bf55194f300953bea5534b6a043130f" | |
| }, | |
| "outputs": [], | |
| "source": [] | |
| } | |
| ], | |
| "metadata": { | |
| "kernelspec": { | |
| "display_name": "Python 3", | |
| "language": "python", | |
| "name": "python3" | |
| }, | |
| "language_info": { | |
| "codemirror_mode": { | |
| "name": "ipython", | |
| "version": 3 | |
| }, | |
| "file_extension": ".py", | |
| "mimetype": "text/x-python", | |
| "name": "python", | |
| "nbconvert_exporter": "python", | |
| "pygments_lexer": "ipython3", | |
| "version": "3.6.4" | |
| } | |
| }, | |
| "nbformat": 4, | |
| "nbformat_minor": 1 | |
| } |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment