-
Notifications
You must be signed in to change notification settings - Fork 35
/
bsds.py
57 lines (57 loc) · 1.78 KB
/
bsds.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
# dataset settings
dataset_type = 'BSDSDataset'
data_root = 'data/BSDS'
img_norm_cfg = dict(
mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225], to_rgb=True)
crop_size = (320, 320)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='LoadAnnotations'),
#dict(type='Resize', img_scale=(2048, 320), ratio_range=(0.5, 2.0)),
dict(type='RandomCropTrain', crop_size=crop_size, cat_max_ratio=0.75),
#dict(type='RandomFlip', flip_ratio=0.5),
#dict(type='PhotoMetricDistortion'),
dict(type='PadBSDS', size=crop_size, pad_val=0, seg_pad_val=255),
#dict(type='DefaultFormatBundle'),
dict(type='NormalizeBSDS', **img_norm_cfg),
dict(type='Collect', keys=['img', 'gt_semantic_seg']),
]
test_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='MultiScaleFlipAug',
img_scale=(2048, 320),
# img_ratios=[0.5, 0.75, 1.0, 1.25, 1.5, 1.75],
flip=False,
transforms=[
#dict(type='Resize', keep_ratio=True),
#dict(type='RandomFlip'),
dict(type='NormalizeBSDSTest', **img_norm_cfg),
#dict(type='ImageToTensor', keys=['img']),
dict(type='Collect', keys=['img']),
])
]
data = dict(
samples_per_gpu=4,
workers_per_gpu=4,
train=dict(
type=dataset_type,
data_root=data_root,
img_dir='',
ann_dir='',
split='ImageSets/train_pair.txt',
pipeline=train_pipeline),
val=dict(
type=dataset_type,
data_root=data_root,
img_dir='',
ann_dir='',
split='ImageSets/test.txt',
pipeline=test_pipeline),
test=dict(
type=dataset_type,
data_root=data_root,
img_dir='',
ann_dir='',
split='ImageSets/test.txt',
pipeline=test_pipeline))