def get_data_training(img_hdf5, msk_hdf5, patch_h, patch_w, N_subimgs,
inside_FOV):
train_img = load_hdf5(img_hdf5)
train_msk = load_hdf5(msk_hdf5)
consistency_check(train_img, train_msk)
patches_train_img, patches_train_msk = extract_random(train_img,
train_msk,
patch_h,
patch_w,
N_subimgs,
inside=inside_FOV)
consistency_check(patches_train_img, patches_train_msk)
#patches_train_msk = mask_transform(patches_train_msk)
return patches_train_img, patches_train_msk
def visualize_predicts(predicts, images_path, labels, masks, height, width,
name):
"""
right: green
error: red
missing: blue
"""
if not path.exists('./logs/' + name + '/predicts/'):
system('mkdir ./logs/' + name + '/predicts/')
images = load_hdf5(images_path + 'test_images.hdf5')
pad_height = labels.shape[1]
pad_width = labels.shape[2]
for i in range(len(predicts)):
vis = np.zeros((pad_height, pad_width, 3))
right = predicts[i] * np.squeeze(labels[i]) * np.squeeze(masks[i])
error = predicts[i] - right
missing = np.squeeze(labels[i]) - right
vis[:, :, 0] = error
vis[:, :, 1] = right
vis[:, :, 2] = missing
vis = np.concatenate((images[i], vis[:height, :width] * 255), axis=0)
visualize(vis, './logs/' + name + '/predicts/' + str(i + 1) + '.png')
def __call__(self):
images = load_hdf5(self.images_path)
labels = load_hdf5(self.labels_path)
masks = load_hdf5(self.mask_path)
images = pre_processing(images)
if np.max(labels) > 1:
labels = labels / 255.
masks = masks / 255.
if self.train_test == 'train':
return self.extract_ordered(images, labels)
if self.train_test == 'test':
sub_images, sub_labels = self.extract_ordered(images, labels)
return (sub_images, images, labels, masks)
def __call__(self):
images = load_hdf5(self.images_path)
labels = load_hdf5(self.labels_path)
masks = load_hdf5(self.mask_path)
assert(images.shape[1]==self.height and images.shape[2]==self.width)
assert(labels.shape[1]==self.height and labels.shape[2]==self.width)
assert(masks.shape[1]==self.height and masks.shape[2]==self.width)
images = pre_processing(images)
if np.max(labels) > 1:
labels = labels / 255.
masks = masks / 255.
images, labels, masks = self.padding(images, labels, masks)
print('images:', images.shape, images.dtype, np.min(images), np.max(images))
print('labels:', labels.shape, labels.dtype, np.min(labels), np.max(labels))
print('masks:', masks.shape, masks.dtype, np.min(masks), np.max(masks))
return images, labels, masks
def __call__(self):
images = load_hdf5(self.images_path)
labels = load_hdf5(self.labels_path)
masks = load_hdf5(self.mask_path)
images = pre_processing(images)
if np.max(labels) > 1:
labels = labels / 255.
masks = masks / 255.
#visualize(group_images(images, 4)).show()
#visualize(group_images(labels, 4)).show()
#visualize(group_images(masks, 4)).show()
#print(images.shape, images.dtype, np.min(images), np.max(images))
#print(labels.shape, labels.dtype, np.min(labels), np.max(labels))
#print(masks.shape, masks.dtype, np.min(masks), np.max(masks))
if self.train_test == 'train':
return self.extract_ordered(images, labels)
if self.train_test == 'test':
sub_images, sub_labels = self.extract_ordered(images, labels)
return (sub_images, images, labels, masks)
images_std = np.std(images)
images_mean = np.mean(images)
images_normalized = (images - images_mean) / images_std
for i in range(images.shape[0]):
minv = np.min(images_normalized[i])
images_normalized[i] = ((images_normalized[i] - minv) /
(np.max(images_normalized[i]) - minv)) * 255
return images_normalized
def adjust_gamma(images, gamma=1.0):
assert (len(images.shape) == 4)
assert (images.shape[3] == 1)
invGamma = 1.0 / gamma
table = np.array([((i / 255.0)**invGamma) * 255
for i in np.arange(0, 256)]).astype("uint8")
new_images = np.empty(images.shape)
for i in range(images.shape[0]):
new_images[i, :, :,
0] = cv2.LUT(np.array(images[i, :, :, 0], dtype=np.uint8),
table)
return new_images
if __name__ == '__main__':
from utils import load_hdf5
images = load_hdf5('./datasets/CHASEDB/h5py/train_images.hdf5')
images = pre_processing(images)
from dataset import Dataset
import os
from PIL import Image
import numpy as np
from utils import load_hdf5
'''
pth = '/home/tianshu/unet/data/comb/label/'
#imgs = sorted(os.listdir(pth))
img_pth = pth+'21_manual1.gif'
label = Image.open(img_pth).convert('L')
label = np.array(label)[9:574, :]
print(label.shape)
label = label / 255
classes = np.unique(label)
print(classes)
'''
msks = load_hdf5('./hdf5/DRIVE_dataset_imgs_train.hdf5')
print(np.max(msks), np.min(msks))
def load_dataset(sketch_data_dir,
photo_data_dir,
model_params,
inference_mode=False):
"""Loads the .npz file, and splits the set into train/test."""
# normalizes the x and y columns using the training set.
# applies same scaling factor to test set.
if isinstance(model_params.data_set, list):
datasets = model_params.data_set
else:
datasets = [model_params.data_set]
train_strokes = None
test_strokes = None
train_image_paths = []
test_image_paths = []
for dataset in datasets:
if model_params.data_type == 'QMUL':
train_data_filepath = os.path.join(sketch_data_dir, dataset,
'train_svg_sim_spa_png.h5')
test_data_filepath = os.path.join(sketch_data_dir, dataset,
'test_svg_sim_spa_png.h5')
train_data_dict = utils.load_hdf5(train_data_filepath)
test_data_dict = utils.load_hdf5(test_data_filepath)
train_sketch_data = utils.reassemble_data(
train_data_dict['image_data'], train_data_dict['data_offset']
) # list of [N_sketches], each [N_points, 4]
train_photo_names = train_data_dict[
'image_base_name'] # [N_sketches, 1], byte
train_photo_paths = [
os.path.join(photo_data_dir,
train_photo_names[i, 0].decode() + '.png')
for i in range(train_photo_names.shape[0])
] # [N_sketches], str
test_sketch_data = utils.reassemble_data(
test_data_dict['image_data'], test_data_dict['data_offset']
) # list of [N_sketches], each [N_points, 4]
test_photo_names = test_data_dict[
'image_base_name'] # [N_sketches, 1], byte
test_photo_paths = [
os.path.join(photo_data_dir,
test_photo_names[i, 0].decode() + '.png')
for i in range(test_photo_names.shape[0])
] # [N_sketches], str
# transfer stroke-4 to stroke-3
train_sketch_data = utils.to_normal_strokes_4to3(train_sketch_data)
test_sketch_data = utils.to_normal_strokes_4to3(
test_sketch_data) # [N_sketches,], each with [N_points, 3]
if train_strokes is None:
train_strokes = train_sketch_data
test_strokes = test_sketch_data
else:
train_strokes = np.concatenate(
(train_strokes, train_sketch_data))
test_strokes = np.concatenate((test_strokes, test_sketch_data))
elif model_params.data_type == 'QuickDraw':
data_filepath = os.path.join(sketch_data_dir, dataset, 'npz',
'sketchrnn_' + dataset + '.npz')
if six.PY3:
data = np.load(data_filepath, encoding='latin1')
else:
data = np.load(data_filepath)
if train_strokes is None:
train_strokes = data[
'train'] # [N_sketches,], each with [N_points, 3]
test_strokes = data['test']
else:
train_strokes = np.concatenate((train_strokes, data['train']))
test_strokes = np.concatenate((test_strokes, data['test']))
train_photo_paths = [
os.path.join(
sketch_data_dir, dataset, 'png', 'train',
str(model_params.image_size) + 'x' +
str(model_params.image_size),
str(im_idx) + '.png')
for im_idx in range(len(data['train']))
]
test_photo_paths = [
os.path.join(
sketch_data_dir, dataset, 'png', 'test',
str(model_params.image_size) + 'x' +
str(model_params.image_size),
str(im_idx) + '.png')
for im_idx in range(len(data['test']))
]
else:
raise Exception('Unknown data type:', model_params.data_type)
print('Loaded {}/{} from {} {}'.format(len(train_photo_paths),
len(test_photo_paths),
model_params.data_type,
dataset))
train_image_paths += train_photo_paths
test_image_paths += test_photo_paths
all_strokes = np.concatenate((train_strokes, test_strokes))
num_points = 0
for stroke in all_strokes:
num_points += len(stroke)
avg_len = num_points / len(all_strokes)
print('Dataset combined: {} ({}/{}), avg len {}'.format(
len(all_strokes), len(train_strokes), len(test_strokes), int(avg_len)))
assert len(train_image_paths) == len(train_strokes)
assert len(test_image_paths) == len(test_strokes)
# calculate the max strokes we need.
max_seq_len = utils.get_max_len(all_strokes)
# overwrite the hps with this calculation.
model_params.max_seq_len = max_seq_len
print('model_params.max_seq_len %i.' % model_params.max_seq_len)
eval_model_params = sketch_p2s_model.copy_hparams(model_params)
eval_model_params.use_input_dropout = 0
eval_model_params.use_recurrent_dropout = 0
eval_model_params.use_output_dropout = 0
eval_model_params.is_training = 1
if inference_mode:
eval_model_params.batch_size = 1
eval_model_params.is_training = 0
sample_model_params = sketch_p2s_model.copy_hparams(eval_model_params)
sample_model_params.batch_size = 1 # only sample one at a time
sample_model_params.max_seq_len = 1 # sample one point at a time
train_set = utils.DataLoader(
train_strokes,
train_image_paths,
model_params.image_size,
model_params.image_size,
model_params.batch_size,
max_seq_length=model_params.max_seq_len,
random_scale_factor=model_params.random_scale_factor,
augment_stroke_prob=model_params.augment_stroke_prob)
normalizing_scale_factor = train_set.calculate_normalizing_scale_factor()
train_set.normalize(normalizing_scale_factor)
# valid_set = utils.DataLoader(
# valid_strokes,
# eval_model_params.batch_size,
# max_seq_length=eval_model_params.max_seq_len,
# random_scale_factor=0.0,
# augment_stroke_prob=0.0)
# valid_set.normalize(normalizing_scale_factor)
test_set = utils.DataLoader(test_strokes,
test_image_paths,
model_params.image_size,
model_params.image_size,
eval_model_params.batch_size,
max_seq_length=eval_model_params.max_seq_len,
random_scale_factor=0.0,
augment_stroke_prob=0.0)
test_set.normalize(normalizing_scale_factor)
print('normalizing_scale_factor %4.4f.' % normalizing_scale_factor)
result = [
train_set, None, test_set, model_params, eval_model_params,
sample_model_params
]
return result
