Skip to content

Commit 6e2a4c7

Browse files
ChongWei905IASZHTChongWei905
authored
merge commits renewing forms from branch v0.4.0 to main (#800)
* zht: renew README forms. (#792) * chore: renew README forms (#791) Co-authored-by: ChongWei905 <weichong4@huawei.com> --------- Co-authored-by: JingRiGuMing <48117115+JingRiGuMing@users.noreply.github.com> Co-authored-by: ChongWei905 <weichong4@huawei.com>
1 parent 90c7887 commit 6e2a4c7

File tree

50 files changed

+1261
-498
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

50 files changed

+1261
-498
lines changed

‎configs/bit/README.md

Lines changed: 11 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -17,13 +17,19 @@ too low. 5) With BiT fine-tuning, good performance can be achieved even if there
1717

1818
Our reproduced model performance on ImageNet-1K is reported as follows.
1919

20+
performance tested on ascend 910*(8p) with graph mode
21+
22+
*coming soon*
23+
24+
performance tested on ascend 910(8p) with graph mode
25+
26+
2027
<div align="center">
2128

22-
| Model | Context | Top-1 (%) | Top-5 (%) | Params(M) | Recipe | Download |
23-
|----------------| -------- |-----------|-----------|-----------|--------------------------------------------------------------------------------------------------| -------------------------------------------------------------------------- |
24-
| bit_resnet50 | D910x8-G | 76.81 | 93.17 | 25.55 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/bit/bit_resnet50_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/bit/BiT_resnet50-1e4795a4.ckpt) |
25-
| bit_resnet50x3 | D910x8-G | 80.63 | 95.12 | 217.31 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/bit/bit_resnet50x3_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/bit/BiT_resnet50x3-a960f91f.ckpt) |
26-
| bit_resnet101 | D910x8-G | 77.93 | 93.75 | 44.54 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/bit/bit_resnet101_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/bit/BiT_resnet101-2efa9106.ckpt) |
29+
| Model | Top-1 (%) | Top-5 (%) | Params(M) | Batch Size | Recipe | Download |
30+
| ------------ | --------- | --------- | --------- | ---------- | ---------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------- |
31+
| bit_resnet50 | 76.81 | 93.17 | 25.55 | 32 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/bit/bit_resnet50_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/bit/BiT_resnet50-1e4795a4.ckpt) |
32+
2733

2834
</div>
2935

‎configs/cmt/README.md

Lines changed: 9 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -14,11 +14,17 @@ on ImageNet-1K dataset.
1414

1515
Our reproduced model performance on ImageNet-1K is reported as follows.
1616

17+
performance tested on ascend 910*(8p) with graph mode
18+
19+
*coming soon*
20+
21+
performance tested on ascend 910(8p) with graph mode
22+
1723
<div align="center">
1824

19-
| Model | Context | Top-1 (%) | Top-5 (%) | Params(M) | Recipe | Download |
20-
|-----------| -------- |-----------|-----------|-----------|---------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------|
21-
| cmt_small | D910x8-G | 83.24 | 96.41 | 26.09 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/cmt/cmt_small_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/cmt/cmt_small-6858ee22.ckpt) |
25+
| Model | Top-1 (%) | Top-5 (%) | Params(M) | Batch Size | Recipe | Download |
26+
| --------- | --------- | --------- | --------- | ---------- | ------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------ |
27+
| cmt_small | 83.24 | 96.41 | 26.09 | 128 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/cmt/cmt_small_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/cmt/cmt_small-6858ee22.ckpt) |
2228

2329

2430
</div>

‎configs/coat/README.md

Lines changed: 10 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -10,14 +10,18 @@ Co-Scale Conv-Attentional Image Transformer (CoaT) is a Transformer-based image
1010

1111
Our reproduced model performance on ImageNet-1K is reported as follows.
1212

13+
performance tested on ascend 910*(8p) with graph mode
14+
15+
*coming soon*
16+
17+
18+
performance tested on ascend 910(8p) with graph mode
19+
1320
<div align="center">
1421

15-
| Model | Context | Top-1 (%) | Top-5 (%) | Params (M) | Recipe | Weight |
16-
|-----------------|-----------|-------|------------|------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------|----------------------------------------------------------------------------------|
17-
| coat_lite_tiny | D910x8-G | 77.35 | 93.43 | 5.72 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/coat/coat_lite_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/coat/coat_lite_tiny-fa7bf894.ckpt) |
18-
| coat_lite_mini | D910x8-G | 78.51 | 93.84 | 11.01 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/coat/coat_lite_mini_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/coat/coat_lite_mini-55a52f05.ckpt) |
19-
| coat_tiny | D910x8-G | 79.67 | 94.88 | 5.50 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/coat/coat_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/coat/coat_tiny-071cb792.ckpt) |
20-
| coat_mini | D910x8-G | 81.08 | 95.34 | 10.34 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/coat/coat_mini_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/coat/coat_mini-57c5bce7.ckpt) |
22+
| Model | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe | Weight |
23+
| --------- | --------- | --------- | ---------- | ---------- | -------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------- |
24+
| coat_tiny | 79.67 | 94.88 | 5.50 | 32 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/coat/coat_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/coat/coat_tiny-071cb792.ckpt) |
2125

2226
</div>
2327

‎configs/convit/README.md

Lines changed: 16 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -24,16 +24,24 @@ while offering a much improved sample efficiency.[[1](#references)]
2424

2525
Our reproduced model performance on ImageNet-1K is reported as follows.
2626

27+
performance tested on ascend 910*(8p) with graph mode
28+
29+
30+
<div align="center">
31+
32+
| Model | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe | Download |
33+
| ----------- | --------- | --------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------ |-------------|
34+
| convit_tiny | 73.79 | 91.70 | 5.71 | 256 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_tiny_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/convit/convit_tiny-1961717e-910v2.ckpt) |
35+
36+
</div>
37+
38+
performance tested on ascend 910(8p) with graph mode
39+
2740
<div align="center">
2841

29-
| Model | Context | Top-1 (%) | Top-5 (%) | Params (M) | Recipe | Download |
30-
|-------------------|----------|-----------|-----------|------------|--------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------|
31-
| convit_tiny | D910x8-G | 73.66 | 91.72 | 5.71 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convit/convit_tiny-e31023f2.ckpt) |
32-
| convit_tiny_plus | D910x8-G | 77.00 | 93.60 | 9.97 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_tiny_plus_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convit/convit_tiny_plus-e9d7fb92.ckpt) |
33-
| convit_small | D910x8-G | 81.63 | 95.59 | 27.78 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_small_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convit/convit_small-ba858604.ckpt) |
34-
| convit_small_plus | D910x8-G | 81.80 | 95.42 | 48.98 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_small_plus_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convit/convit_small_plus-2352b9f7.ckpt) |
35-
| convit_base | D910x8-G | 82.10 | 95.52 | 86.54 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_base_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convit/convit_base-c61b808c.ckpt) |
36-
| convit_base_plus | D910x8-G | 81.96 | 95.04 | 153.13 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_base_plus_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convit/convit_base_plus-5c61c9ce.ckpt) |
42+
| Model | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe | Download |
43+
| ----------- | --------- | --------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------ | ----------------------------------------------------------------------------------------- |
44+
| convit_tiny | 73.66 | 91.72 | 5.71 | 256 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convit/convit_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convit/convit_tiny-e31023f2.ckpt) |
3745

3846
</div>
3947

‎configs/convnext/README.md

Lines changed: 17 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -21,13 +21,25 @@ simplicity and efficiency of standard ConvNets.[[1](#references)]
2121

2222
Our reproduced model performance on ImageNet-1K is reported as follows.
2323

24+
performance tested on ascend 910*(8p) with graph mode
25+
2426
<div align="center">
2527

26-
| Model | Context | Top-1 (%) | Top-5 (%) | Params (M) | Recipe | Download |
27-
|----------------|-----------|-----------|-----------|------------|-------------------------------------------------------------------------------------------------------|------------------------------------------------------------------------------------------------|
28-
| convnext_tiny | D910x64-G | 81.91 | 95.79 | 28.59 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnext/convnext_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convnext/convnext_tiny-ae5ff8d7.ckpt) |
29-
| convnext_small | D910x64-G | 83.40 | 96.36 | 50.22 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnext/convnext_small_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convnext/convnext_small-e23008f3.ckpt) |
30-
| convnext_base | D910x64-G | 83.32 | 96.24 | 88.59 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnext/convnext_base_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convnext/convnext_base-ee3544b8.ckpt) |
28+
| Model | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe | Download |
29+
| ------------- | --------- | --------- | ---------- | ---------- | ---------------------------------------------------------------------------------------------------- |-------------|
30+
| convnext_tiny | 81.28 | 95.61 | 28.59 | 16 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnext/convnext_tiny_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/convnext/convnext_tiny-db11dc82-910v2.ckpt) |
31+
32+
33+
</div>
34+
35+
performance tested on ascend 910(8p) with graph mode
36+
37+
<div align="center">
38+
39+
| Model | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe | Download |
40+
| ------------- | --------- | --------- | ---------- | ---------- | ---------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------- |
41+
| convnext_tiny | 81.91 | 95.79 | 28.59 | 16 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnext/convnext_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convnext/convnext_tiny-ae5ff8d7.ckpt) |
42+
3143

3244
</div>
3345

‎configs/convnextv2/README.md

Lines changed: 15 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -20,11 +20,23 @@ benchmarks, including ImageNet classification, COCO detection, and ADE20K segmen
2020

2121
Our reproduced model performance on ImageNet-1K is reported as follows.
2222

23+
performance tested on ascend 910*(8p) with graph mode
24+
25+
<div align="center">
26+
27+
| Model | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe | Download |
28+
| --------------- | --------- | --------- | ---------- | ---------- | -------------------------------------------------------------------------------------------------------- |-------------|
29+
| convnextv2_tiny | 82.39 | 95.95 | 28.64 | 128 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnextv2/convnextv2_tiny_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/convnextv2/convnextv2_tiny-a35b79ce-910v2.ckpt) |
30+
31+
</div>
32+
33+
performance tested on ascend 910(8p) with graph mode
34+
2335
<div align="center">
2436

25-
| Model | Context | Top-1 (%) | Top-5 (%) | Params (M) | Recipe | Download |
26-
|------------------|----------|-----------|-----------|------------|----------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------|
27-
| convnextv2_tiny | D910x8-G | 82.43 | 95.98 | 28.64 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnextv2/convnextv2_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convnextv2/convnextv2_tiny-d441ba2c.ckpt) |
37+
| Model | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe | Download |
38+
| --------------- | --------- | --------- | ---------- | ---------- | -------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------- |
39+
| convnextv2_tiny | 82.43 | 95.98 | 28.64 | 128 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/convnextv2/convnextv2_tiny_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/convnextv2/convnextv2_tiny-d441ba2c.ckpt) |
2840

2941
</div>
3042

‎configs/crossvit/README.md

Lines changed: 14 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -19,14 +19,23 @@ Fusion is achieved by an efficient cross-attention module, in which each transfo
1919

2020
Our reproduced model performance on ImageNet-1K is reported as follows.
2121

22+
performance tested on ascend 910*(8p) with graph mode
23+
2224
<div align="center">
2325

24-
| Model | Context | Top-1 (%) | Top-5 (%) | Params (M) | Recipe | Download |
25-
|-------------|----------|-----------|-----------|------------|-----------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------|
26-
| crossvit_9 | D910x8-G | 73.56 | 91.79 | 8.55 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/crossvit/crossvit_9_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/crossvit/crossvit_9-e74c8e18.ckpt) |
27-
| crossvit_15 | D910x8-G | 81.08 | 95.33 | 27.27 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/crossvit/crossvit_15_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/crossvit/crossvit_15-eaa43c02.ckpt) |
28-
| crossvit_18 | D910x8-G | 81.93 | 95.75 | 43.27 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/crossvit/crossvit_18_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/crossvit/crossvit_18-ca0a2e43.ckpt) |
26+
| Model | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe | Download |
27+
| ---------- | --------- | --------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------- |-------------|
28+
| crossvit_9 | 73.38 | 91.51 | 8.55 | 256 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/crossvit/crossvit_9_ascend.yaml) | [weights](https://download-mindspore.osinfra.cn/toolkits/mindcv/crossvit/crossvit_9-32c69c96-910v2.ckpt) |
29+
30+
</div>
31+
32+
performance tested on ascend 910(8p) with graph mode
33+
34+
<div align="center">
2935

36+
| Model | Top-1 (%) | Top-5 (%) | Params (M) | Batch Size | Recipe | Download |
37+
| ---------- | --------- | --------- | ---------- | ---------- | ------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------ |
38+
| crossvit_9 | 73.56 | 91.79 | 8.55 | 256 | [yaml](https://github.com/mindspore-lab/mindcv/blob/main/configs/crossvit/crossvit_9_ascend.yaml) | [weights](https://download.mindspore.cn/toolkits/mindcv/crossvit/crossvit_9-e74c8e18.ckpt) |
3039

3140
</div>
3241

0 commit comments

Comments
 (0)