Skip to content

Commit a95f6d8

Browse files
author
谢昕辰
authored
[Feature] support mim (open-mmlab#549)
* dice loss * format code, add docstring and calculate denominator without valid_mask * minor change * restore * add metafile * add manifest.in and add config at setup.py * add requirements * modify manifest * modify manifest * Update MANIFEST.in * add metafile * add metadata * fix typo * Update metafile.yml * Update metafile.yml * minor change * Update metafile.yml * add subfix * fix mmshow * add more metafile * add config to model_zoo * fix bug * Update mminstall.txt * [fix] Add models * [Fix] Add collections * [fix] Modify collection name * [Fix] Set datasets to unet metafile * [Fix] Modify collection names * complement inference time
1 parent dc5d53b commit a95f6d8

File tree

31 files changed

+5566
-0
lines changed

31 files changed

+5566
-0
lines changed

‎MANIFEST.in‎

Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,5 @@
1+
2+
include requirements/*.txt
3+
include mmseg/model_zoo.yml
4+
recursive-include mmseg/configs *.py *.yml
5+
recursive-include mmseg/tools *.sh *.py

‎configs/ann/metafile.yml‎

Lines changed: 231 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,231 @@
1+
Collections:
2+
- Name: ANN
3+
Metadata:
4+
Training Data:
5+
- Cityscapes
6+
- Pascal VOC 2012 + Aug
7+
- ADE20K
8+
9+
Models:
10+
11+
- Name: ann_r50-d8_512x1024_40k_cityscapes
12+
In Collection: ANN
13+
Metadata:
14+
inference time (fps): 3.71
15+
Results:
16+
- Task: Semantic Segmentation
17+
Dataset: Cityscapes
18+
Metrics:
19+
mIoU: 77.40
20+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r50-d8_512x1024_40k_cityscapes/ann_r50-d8_512x1024_40k_cityscapes_20200605_095211-049fc292.pth
21+
Config: configs/ann/ann_r50-d8_512x1024_40k_cityscapes.py
22+
23+
24+
25+
- Name: ann_r101-d8_512x1024_40k_cityscapes
26+
In Collection: ANN
27+
Metadata:
28+
inference time (fps): 2.55
29+
Results:
30+
- Task: Semantic Segmentation
31+
Dataset: Cityscapes
32+
Metrics:
33+
mIoU: 76.55
34+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r101-d8_512x1024_40k_cityscapes/ann_r101-d8_512x1024_40k_cityscapes_20200605_095243-adf6eece.pth
35+
Config: configs/ann/ann_r101-d8_512x1024_40k_cityscapes.py
36+
37+
38+
39+
- Name: ann_r50-d8_769x769_40k_cityscapes
40+
In Collection: ANN
41+
Metadata:
42+
inference time (fps): 1.70
43+
Results:
44+
- Task: Semantic Segmentation
45+
Dataset: Cityscapes
46+
Metrics:
47+
mIoU: 78.89
48+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r50-d8_769x769_40k_cityscapes/ann_r50-d8_769x769_40k_cityscapes_20200530_025712-2b46b04d.pth
49+
Config: configs/ann/ann_r50-d8_769x769_40k_cityscapes.py
50+
51+
52+
53+
- Name: ann_r101-d8_769x769_40k_cityscapes
54+
In Collection: ANN
55+
Metadata:
56+
inference time (fps): 1.15
57+
Results:
58+
- Task: Semantic Segmentation
59+
Dataset: Cityscapes
60+
Metrics:
61+
mIoU: 79.32
62+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r101-d8_769x769_40k_cityscapes/ann_r101-d8_769x769_40k_cityscapes_20200530_025720-059bff28.pth
63+
Config: configs/ann/ann_r101-d8_769x769_40k_cityscapes.py
64+
65+
66+
67+
- Name: ann_r50-d8_512x1024_80k_cityscapes
68+
In Collection: ANN
69+
Metadata:
70+
inference time (fps): 3.71
71+
Results:
72+
- Task: Semantic Segmentation
73+
Dataset: Cityscapes
74+
Metrics:
75+
mIoU: 77.34
76+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r50-d8_512x1024_80k_cityscapes/ann_r50-d8_512x1024_80k_cityscapes_20200607_101911-5a9ad545.pth
77+
Config: configs/ann/ann_r50-d8_512x1024_80k_cityscapes.py
78+
79+
80+
81+
- Name: ann_r101-d8_512x1024_80k_cityscapes
82+
In Collection: ANN
83+
Metadata:
84+
inference time (fps): 2.55
85+
Results:
86+
- Task: Semantic Segmentation
87+
Dataset: Cityscapes
88+
Metrics:
89+
mIoU: 77.14
90+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r101-d8_512x1024_80k_cityscapes/ann_r101-d8_512x1024_80k_cityscapes_20200607_013728-aceccc6e.pth
91+
Config: configs/ann/ann_r101-d8_512x1024_80k_cityscapes.py
92+
93+
94+
95+
- Name: ann_r50-d8_769x769_80k_cityscapes
96+
In Collection: ANN
97+
Metadata:
98+
inference time (fps): 1.70
99+
Results:
100+
- Task: Semantic Segmentation
101+
Dataset: Cityscapes
102+
Metrics:
103+
mIoU: 78.88
104+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r50-d8_769x769_80k_cityscapes/ann_r50-d8_769x769_80k_cityscapes_20200607_044426-cc7ff323.pth
105+
Config: configs/ann/ann_r50-d8_769x769_80k_cityscapes.py
106+
107+
108+
109+
- Name: ann_r101-d8_769x769_80k_cityscapes
110+
In Collection: ANN
111+
Metadata:
112+
inference time (fps):
113+
Results:
114+
- Task: Semantic Segmentation
115+
Dataset: Cityscapes
116+
Metrics:
117+
mIoU: 78.80
118+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r101-d8_769x769_80k_cityscapes/ann_r101-d8_769x769_80k_cityscapes_20200607_013713-a9d4be8d.pth
119+
Config: configs/ann/ann_r101-d8_769x769_80k_cityscapes.py
120+
121+
122+
123+
- Name: ann_r50-d8_512x512_80k_ade20k
124+
In Collection: ANN
125+
Metadata:
126+
inference time (fps): 21.01
127+
Results:
128+
- Task: Semantic Segmentation
129+
Dataset: ADE20K
130+
Metrics:
131+
mIoU: 41.01
132+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r50-d8_512x512_80k_ade20k/ann_r50-d8_512x512_80k_ade20k_20200615_014818-26f75e11.pth
133+
Config: configs/ann/ann_r50-d8_512x512_80k_ade20k.py
134+
135+
136+
137+
- Name: ann_r101-d8_512x512_80k_ade20k
138+
In Collection: ANN
139+
Metadata:
140+
inference time (fps): 14.12
141+
Results:
142+
- Task: Semantic Segmentation
143+
Dataset: ADE20K
144+
Metrics:
145+
mIoU: 42.94
146+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r101-d8_512x512_80k_ade20k/ann_r101-d8_512x512_80k_ade20k_20200615_014818-c0153543.pth
147+
Config: configs/ann/ann_r101-d8_512x512_80k_ade20k.py
148+
149+
150+
151+
- Name: ann_r50-d8_512x512_160k_ade20k
152+
In Collection: ANN
153+
Metadata:
154+
inference time (fps): 21.01
155+
Results:
156+
- Task: Semantic Segmentation
157+
Dataset: ADE20K
158+
Metrics:
159+
mIoU: 41.74
160+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r50-d8_512x512_160k_ade20k/ann_r50-d8_512x512_160k_ade20k_20200615_231733-892247bc.pth
161+
Config: configs/ann/ann_r50-d8_512x512_160k_ade20k.py
162+
163+
164+
165+
- Name: ann_r101-d8_512x512_160k_ade20k
166+
In Collection: ANN
167+
Metadata:
168+
inference time (fps): 14.12
169+
Results:
170+
- Task: Semantic Segmentation
171+
Dataset: ADE20K
172+
Metrics:
173+
mIoU: 42.94
174+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r101-d8_512x512_160k_ade20k/ann_r101-d8_512x512_160k_ade20k_20200615_231733-955eb1ec.pth
175+
Config: configs/ann/ann_r101-d8_512x512_160k_ade20k.py
176+
177+
178+
179+
- Name: ann_r50-d8_512x512_20k_voc12aug
180+
In Collection: ANN
181+
Metadata:
182+
inference time (fps): 20.92
183+
Results:
184+
- Task: Semantic Segmentation
185+
Dataset: Pascal VOC 2012 + Aug
186+
Metrics:
187+
mIoU: 74.86
188+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r50-d8_512x512_20k_voc12aug/ann_r50-d8_512x512_20k_voc12aug_20200617_222246-dfcb1c62.pth
189+
Config: configs/ann/ann_r50-d8_512x512_20k_voc12aug.py
190+
191+
192+
193+
- Name: ann_r101-d8_512x512_20k_voc12aug
194+
In Collection: ANN
195+
Metadata:
196+
inference time (fps): 13.94
197+
Results:
198+
- Task: Semantic Segmentation
199+
Dataset: Pascal VOC 2012 + Aug
200+
Metrics:
201+
mIoU: 77.47
202+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r101-d8_512x512_20k_voc12aug/ann_r101-d8_512x512_20k_voc12aug_20200617_222246-2fad0042.pth
203+
Config: configs/ann/ann_r101-d8_512x512_20k_voc12aug.py
204+
205+
206+
207+
- Name: ann_r50-d8_512x512_40k_voc12aug
208+
In Collection: ANN
209+
Metadata:
210+
inference time (fps): 20.92
211+
Results:
212+
- Task: Semantic Segmentation
213+
Dataset: Pascal VOC 2012 + Aug
214+
Metrics:
215+
mIoU: 76.56
216+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r50-d8_512x512_40k_voc12aug/ann_r50-d8_512x512_40k_voc12aug_20200613_231314-b5dac322.pth
217+
Config: configs/ann/ann_r50-d8_512x512_40k_voc12aug.py
218+
219+
220+
221+
- Name: ann_r101-d8_512x512_40k_voc12aug
222+
In Collection: ANN
223+
Metadata:
224+
inference time (fps): 13.94
225+
Results:
226+
- Task: Semantic Segmentation
227+
Dataset: Pascal VOC 2012 + Aug
228+
Metrics:
229+
mIoU: 76.70
230+
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/ann/ann_r101-d8_512x512_40k_voc12aug/ann_r101-d8_512x512_40k_voc12aug_20200613_231314-bd205bbe.pth
231+
Config: configs/ann/ann_r101-d8_512x512_40k_voc12aug.py

0 commit comments

Comments
 (0)