# configure args
from opts import *
from opts import dataset_mean, dataset_std # set them in opts
modelFns = {
'unet': Models.VanillaUnet.VanillaUnet,
'segnet': Models.Segnet.Segnet,
'vgg_unet': Models.VGGUnet.VGGUnet,
'vgg_unet2': Models.VGGUnet.VGGUnet2,
'fcn8': Models.FCN8.FCN8,
'fcn32': Models.FCN32.FCN32,
'crfunet': Models.CRFunet.CRFunet
}
# save and compute metrics
vis = VIS(save_path=opt.checkpoint_path)
# configuration session
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
''' Users define data loader (with train and test) '''
img_shape = [opt.imSize, opt.imSize]
label_classes = vis.palette_info()
train_generator, train_samples = dataLoader(opt.data_path + '/train/',
opt.batch_size,
img_shape,
label_classes,
mean=dataset_mean,
std=dataset_std)
test_generator, test_samples = dataLoader(opt.data_path + '/val/',
np.random.seed(SEED)
import tensorflow as tf
tf.set_random_seed(SEED)
import os, shutil, glob
from skimage import transform, io
from model import UNet
from utils import VIS, mean_IU
# configure args
from opts import *
import cv2
# save and compute metrics
vis = VIS(save_path=opt.checkpoint_path)
print('num_class = %d' % opt.num_class)
# configuration session
config = tf.ConfigProto(
device_count={'GPU': 0}
)
# config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
img_shape = [280, 500] # Resized to be the same.
# define input holders
train_list = glob.glob(opt.data_path + '/train/img/0/*.png')
train_num = len(train_list)
test_list = glob.glob(opt.data_path + '/val/img/0/*.png')
import tensorflow as tf
import os, sys
import numpy as np
import scipy.misc as misc
from model import UNet
from utils import dice_coef, dice_coef_loss
from loader import dataLoader, deprocess, dataLoaderNp
from PIL import Image
from utils import VIS, mean_IU
# configure args
from opts import *
from opts import dataset_mean, dataset_std # set them in opts
vis = VIS(save_path=opt.load_from_checkpoint, is_train=False)
# configuration session
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
# define data loader
img_shape = [233, 369] #img_shape = [opt.imSize, opt.imSize]
test_generator, test_samples = dataLoaderNp(opt.data_path, 1, train_mode=False)
# define model, the last dimension is the channel
label = tf.placeholder(tf.int32, shape=[None] + img_shape)
with tf.name_scope('unet'):
model = UNet().create_model(img_shape=img_shape + [3],
num_class=opt.num_class)
img = model.input
# configure args
from opts import *
from opts import dataset_mean, dataset_std # set them in opts
modelFns = {
'unet': Models.VanillaUnet.VanillaUnet,
'segnet': Models.Segnet.Segnet,
'vgg_unet': Models.VGGUnet.VGGUnet,
'vgg_unet2': Models.VGGUnet.VGGUnet2,
'fcn8': Models.FCN8.FCN8,
'fcn32': Models.FCN32.FCN32,
'crfunet': Models.CRFunet.CRFunet
}
vis = VIS(save_path=opt.load_from_checkpoint)
# configuration session
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
# define data loader
img_shape = [opt.imSize, opt.imSize]
label_classes = vis.palette_info()
print('++++++++++++++++++++++++class')
print(label_classes)
test_generator, test_samples = dataLoader(opt.data_path + '/val/',
1,
img_shape,
label_classes,
from keras import backend as K import tensorflow as tf import os, sys import numpy as np import scipy.misc as misc from model import UNet from utils import dice_coef, dice_coef_loss from loader import dataLoader, folderLoader from PIL import Image from utils import VIS, mean_IU # configure args from opts import * # assert(opt.load_from_checkpoint != '') # assert(opt.batch_size == 1) vis = VIS(save_path=opt.load_from_checkpoint) # configuration session config = tf.ConfigProto() config.gpu_options.allow_growth = True sess = tf.Session(config=config) K.set_session(sess) K.set_learning_phase(0) # define data loader test_generator, test_samples = folderLoader(opt.data_path) # define model, the last dimension is the channel img_shape = (opt.imSize, opt.imSize, 3) # img = tf.placeholder(tf.float32, shape=img_shape) label = tf.placeholder(tf.int32, shape=(None, opt.imSize, opt.imSize))
import os, sys
import numpy as np
import scipy.misc as misc
from model import UNet
from utils import dice_coef, dice_coef_loss
from loader import dataLoader, folderLoader
from PIL import Image
from utils import VIS, mean_IU
import cv2
# configure args
from opts import *
# assert(opt.load_from_checkpoint != '')
# assert(opt.batch_size == 1)
from matplotlib import pyplot as plt
#vis = VIS(save_path=opt.load_from_checkpoint)
vis = VIS(save_path='./trainlog')
# configuration session
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
sess = tf.Session(config=config)
# define data loader
img_shape = [opt.imHeight, opt.imWidth]
test_generator, test_samples = folderLoader(opt.data_path, imSize=img_shape)
# define model, the last dimension is the channel
label = tf.placeholder(tf.int32, shape=[None] + img_shape)
with tf.name_scope('unet'):
model = UNet().create_model(img_shape=img_shape + [3],
num_class=opt.num_class)
img = model.input