forked from open-mmlab/mmsegmentation
-
Notifications
You must be signed in to change notification settings - Fork 0
/
metafile.yaml
109 lines (109 loc) · 4.57 KB
/
metafile.yaml
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
Collections:
- Name: SegNeXt
License: Apache License 2.0
Metadata:
Training Data:
- ADE20K
Paper:
Title: 'SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation'
URL: https://arxiv.org/abs/2209.08575
README: configs/segnext/README.md
Frameworks:
- PyTorch
Models:
- Name: segnext_mscan-t_1xb16-adamw-160k_ade20k-512x512
In Collection: SegNeXt
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 41.5
mIoU(ms+flip): 42.59
Config: configs/segnext/segnext_mscan-t_1xb16-adamw-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- MSCAN-T
- SegNeXt
Training Resources: 1x A100 GPUS
Memory (GB): 17.88
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/segnext/segnext_mscan-t_1x16_512x512_adamw_160k_ade20k/segnext_mscan-t_1x16_512x512_adamw_160k_ade20k_20230210_140244-05bd8466.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/segnext/segnext_mscan-t_1x16_512x512_adamw_160k_ade20k/segnext_mscan-t_1x16_512x512_adamw_160k_ade20k_20230210_140244.log.json
Paper:
Title: 'SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation'
URL: https://arxiv.org/abs/2209.08575
Code: https://github.com/open-mmlab/mmsegmentation/blob/main/mmseg/models/backbones/mscan.py#L328
Framework: PyTorch
- Name: segnext_mscan-s_1xb16-adamw-160k_ade20k-512x512
In Collection: SegNeXt
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 44.16
mIoU(ms+flip): 45.81
Config: configs/segnext/segnext_mscan-s_1xb16-adamw-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- MSCAN-S
- SegNeXt
Training Resources: 1x A100 GPUS
Memory (GB): 21.47
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/segnext/segnext_mscan-s_1x16_512x512_adamw_160k_ade20k/segnext_mscan-s_1x16_512x512_adamw_160k_ade20k_20230214_113014-43013668.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/segnext/segnext_mscan-s_1x16_512x512_adamw_160k_ade20k/segnext_mscan-s_1x16_512x512_adamw_160k_ade20k_20230214_113014.log.json
Paper:
Title: 'SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation'
URL: https://arxiv.org/abs/2209.08575
Code: https://github.com/open-mmlab/mmsegmentation/blob/main/mmseg/models/backbones/mscan.py#L328
Framework: PyTorch
- Name: segnext_mscan-b_1xb16-adamw-160k_ade20k-512x512
In Collection: SegNeXt
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 48.03
mIoU(ms+flip): 49.68
Config: configs/segnext/segnext_mscan-b_1xb16-adamw-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- MSCAN-B
- SegNeXt
Training Resources: 1x A100 GPUS
Memory (GB): 31.03
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/segnext/segnext_mscan-b_1x16_512x512_adamw_160k_ade20k/segnext_mscan-b_1x16_512x512_adamw_160k_ade20k_20230209_172053-b6f6c70c.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/segnext/segnext_mscan-b_1x16_512x512_adamw_160k_ade20k/segnext_mscan-b_1x16_512x512_adamw_160k_ade20k_20230209_172053.log.json
Paper:
Title: 'SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation'
URL: https://arxiv.org/abs/2209.08575
Code: https://github.com/open-mmlab/mmsegmentation/blob/main/mmseg/models/backbones/mscan.py#L328
Framework: PyTorch
- Name: segnext_mscan-l_1xb16-adamw-160k_ade20k-512x512
In Collection: SegNeXt
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 50.99
mIoU(ms+flip): 52.1
Config: configs/segnext/segnext_mscan-l_1xb16-adamw-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- MSCAN-L
- SegNeXt
Training Resources: 1x A100 GPUS
Memory (GB): 43.32
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/segnext/segnext_mscan-l_1x16_512x512_adamw_160k_ade20k/segnext_mscan-l_1x16_512x512_adamw_160k_ade20k_20230209_172055-19b14b63.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/segnext/segnext_mscan-l_1x16_512x512_adamw_160k_ade20k/segnext_mscan-l_1x16_512x512_adamw_160k_ade20k_20230209_172055.log.json
Paper:
Title: 'SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation'
URL: https://arxiv.org/abs/2209.08575
Code: https://github.com/open-mmlab/mmsegmentation/blob/main/mmseg/models/backbones/mscan.py#L328
Framework: PyTorch