/
imagenet_bs32.py
51 lines (46 loc) · 1.31 KB
/
imagenet_bs32.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
# dataset settings
dataset_type = 'ImageNet'
data_preprocessor = dict(
num_classes=1000,
# RGB format normalization parameters
mean=[123.675, 116.28, 103.53],
std=[58.395, 57.12, 57.375],
# convert image from BGR to RGB
to_rgb=True,
)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='RandomResizedCrop', scale=224),
dict(type='RandomFlip', prob=0.5, direction='horizontal'),
dict(type='PackInputs'),
]
test_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='ResizeEdge', scale=256, edge='short'),
dict(type='CenterCrop', crop_size=224),
dict(type='PackInputs'),
]
train_dataloader = dict(
batch_size=32,
num_workers=5,
dataset=dict(
type=dataset_type,
data_root='data/imagenet',
split='train',
pipeline=train_pipeline),
sampler=dict(type='DefaultSampler', shuffle=True),
)
val_dataloader = dict(
batch_size=32,
num_workers=5,
dataset=dict(
type=dataset_type,
data_root='data/imagenet',
split='val',
pipeline=test_pipeline),
sampler=dict(type='DefaultSampler', shuffle=False),
)
val_evaluator = dict(type='Accuracy', topk=(1, 5))
# If you want standard test, please manually configure the test dataset
test_dataloader = val_dataloader
test_evaluator = val_evaluator