-
Notifications
You must be signed in to change notification settings - Fork 119
/
loader.py
176 lines (146 loc) · 5.72 KB
/
loader.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
# Copyright 2019 Image Analysis Lab, German Center for Neurodegenerative Diseases (DZNE), Bonn
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# IMPORTS
from torchvision import transforms
from torch.utils.data import DataLoader
import yacs.config
from FastSurferCNN.data_loader import dataset as dset
from FastSurferCNN.data_loader.augmentation import ToTensor, ZeroPad2D, AddGaussianNoise
from FastSurferCNN.utils import logging
logger = logging.getLogger(__name__)
def get_dataloader(cfg: yacs.config.CfgNode, mode: str):
"""
Create the dataset and pytorch data loader.
Parameters
----------
cfg : yacs.config.CfgNode
Configuration node.
mode : str
Loading data for train, val and test mode.
Returns
-------
torch.utils.data.DataLoader
Dataloader with given configs and mode.
"""
assert mode in ["train", "val"], f"dataloader mode is incorrect {mode}"
padding_size = cfg.DATA.PADDED_SIZE
if mode == "train":
if "None" in cfg.DATA.AUG:
tfs = [ZeroPad2D((padding_size, padding_size)), ToTensor()]
# old transform
if "Gaussian" in cfg.DATA.AUG:
tfs.append(AddGaussianNoise(mean=0, std=0.1))
data_path = cfg.DATA.PATH_HDF5_TRAIN
shuffle = True
logger.info(
f"Loading {mode.capitalize()} data ... from {data_path}. Using standard Aug"
)
dataset = dset.MultiScaleDatasetVal(data_path, cfg, transforms.Compose(tfs))
else:
import torchio as tio
# Elastic
elastic = tio.RandomElasticDeformation(
num_control_points=7,
max_displacement=(20, 20, 0),
locked_borders=2,
image_interpolation="linear",
include=["img", "label", "weight"],
)
# Scales
scaling = tio.RandomAffine(
scales=(0.8, 1.15),
degrees=0,
translation=(0, 0, 0),
isotropic=True, # If True, scaling factor along all dimensions is the same
center="image",
default_pad_value="minimum",
image_interpolation="linear",
include=["img", "label", "weight"],
)
# Rotation
rot = tio.RandomAffine(
scales=(1.0, 1.0),
degrees=10,
translation=(0, 0, 0),
isotropic=True, # If True, scaling factor along all dimensions is the same
center="image",
default_pad_value="minimum",
image_interpolation="linear",
include=["img", "label", "weight"],
)
# Translation
tl = tio.RandomAffine(
scales=(1.0, 1.0),
degrees=0,
translation=(15.0, 15.0, 0),
isotropic=True, # If True, scaling factor along all dimensions is the same
center="image",
default_pad_value="minimum",
image_interpolation="linear",
include=["img", "label", "weight"],
)
# Random Anisotropy (Downsample image along an axis, then upsample back to initial space
ra = tio.transforms.RandomAnisotropy(
axes=(0, 1),
downsampling=(1.1, 1.5),
image_interpolation="linear",
include=["img"],
)
# Bias Field
bias_field = tio.transforms.RandomBiasField(
coefficients=0.5, order=3, include=["img"]
)
# Gamma
random_gamma = tio.transforms.RandomGamma(
log_gamma=(-0.1, 0.1), include=["img"]
)
#
all_augs = {
"Elastic": elastic,
"Scaling": scaling,
"Rotation": rot,
"Translation": tl,
"RAnisotropy": ra,
"BiasField": bias_field,
"RGamma": random_gamma,
}
all_tfs = {all_augs[aug]: 0.8 for aug in cfg.DATA.AUG if aug != "Gaussian"}
gaussian_noise = True if "Gaussian" in cfg.DATA.AUG else False
transform = tio.Compose(
[tio.Compose(all_tfs, p=0.8)], include=["img", "label", "weight"]
)
data_path = cfg.DATA.PATH_HDF5_TRAIN
shuffle = True
logger.info(
f"Loading {mode.capitalize()} data ... from {data_path}. Using torchio Aug"
)
dataset = dset.MultiScaleDataset(data_path, cfg, gaussian_noise, transform)
elif mode == "val":
data_path = cfg.DATA.PATH_HDF5_VAL
shuffle = False
transform = transforms.Compose(
[
ZeroPad2D((padding_size, padding_size)),
ToTensor(),
]
)
logger.info(f"Loading {mode.capitalize()} data ... from {data_path}")
dataset = dset.MultiScaleDatasetVal(data_path, cfg, transform)
dataloader = DataLoader(
dataset,
batch_size=cfg.TRAIN.BATCH_SIZE,
num_workers=cfg.TRAIN.NUM_WORKERS,
shuffle=shuffle,
pin_memory=True,
)
return dataloader