Skip to content

Commit 8c63bb5

Browse files
mzr1996mzr1996
committed
Merge branch 'dev'
2 parents a9489f6 + 29c54dd commit 8c63bb5

File tree

21 files changed

+156
-25
lines changed

21 files changed

+156
-25
lines changed

‎README.md

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -33,6 +33,8 @@
3333
[🆕 Update News](https://mmclassification.readthedocs.io/en/latest/changelog.html) |
3434
[🤔 Reporting Issues](https://github.com/open-mmlab/mmclassification/issues/new/choose)
3535

36+
:point_right: **MMClassification 1.0 branch is in trial, welcome every to [try it](https://github.com/open-mmlab/mmclassification/tree/1.x) and [discuss with us](https://github.com/open-mmlab/mmclassification/discussions)!** :point_left:
37+
3638
</div>
3739

3840
## Introduction
@@ -62,6 +64,11 @@ The MMClassification 1.0 has released! It's still unstable and in release candid
6264
to [the 1.x branch](https://github.com/open-mmlab/mmclassification/tree/1.x) and discuss it with us in
6365
[the discussion](https://github.com/open-mmlab/mmclassification/discussions).
6466

67+
v0.24.1 was released in 31/10/2022.
68+
Highlights of the new version:
69+
70+
- Support HUAWEI Ascend device.
71+
6572
v0.24.0 was released in 30/9/2022.
6673
Highlights of the new version:
6774

‎README_zh-CN.md

Lines changed: 8 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -33,6 +33,10 @@
3333
[🆕 更新日志](https://mmclassification.readthedocs.io/en/latest/changelog.html) |
3434
[🤔 报告问题](https://github.com/open-mmlab/mmclassification/issues/new/choose)
3535

36+
:point_right: **MMClassification 1.0 版本即将正式发布，欢迎大家 [试用](https://github.com/open-mmlab/mmclassification/tree/1.x) 并 [参与讨论](https://github.com/open-mmlab/mmclassification/discussions)！** :point_left:
37+
38+
</div>
39+
3640
</div>
3741

3842
## Introduction
@@ -59,15 +63,17 @@ MMClassification 是一款基于 PyTorch 的开源图像分类工具箱，是 [O
5963

6064
MMClassification 1.0 已经发布！目前仍在公测中，如果希望试用，请切换到 [1.x 分支](https://github.com/open-mmlab/mmclassification/tree/1.x)，并在[讨论版](https://github.com/open-mmlab/mmclassification/discussions) 参加开发讨论！
6165

66+
2022/10/31 发布了 v0.24.1 版本
67+
68+
- 支持了华为昇腾 NPU 设备。
69+
6270
2022/9/30 发布了 v0.24.0 版本
6371

6472
- 支持了 **HorNet**，**EfficientFormerm**，**SwinTransformer V2**，**MViT** 等主干网络。
6573
- 支持了 Support Standford Cars 数据集。
6674

6775
2022/5/1 发布了 v0.23.0 版本
6876

69-
新版本亮点：
70-
7177
- 支持了 **DenseNet**，**VAN** 和 **PoolFormer** 三个网络，并提供了预训练模型。
7278
- 支持在 IPU 上进行训练。
7379
- 更新了 API 文档的样式，更方便查阅，[欢迎查阅](https://mmclassification.readthedocs.io/en/master/api/models.html)。

‎docker/serve/Dockerfile

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -3,8 +3,8 @@ ARG CUDA="10.2"
33
ARG CUDNN="7"
44
FROM pytorch/pytorch:${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel
55

6-
ARG MMCV="1.6.2"
7-
ARG MMCLS="0.24.0"
6+
ARG MMCV="1.7.0"
7+
ARG MMCLS="0.24.1"
88

99
ENV PYTHONUNBUFFERED TRUE
1010

‎docs/en/_static/css/readthedocs.css

Lines changed: 9 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -9,12 +9,19 @@ pre {
99
white-space: pre;
1010
}
1111

12-
article.pytorch-article .section :not(dt) > code {
12+
article.pytorch-article section code {
1313
padding: .2em .4em;
1414
background-color: #f3f4f7;
1515
border-radius: 5px;
1616
}
1717

18-
table.colwidths-auto td {
18+
/* Disable the change in tables */
19+
article.pytorch-article section table code {
20+
padding: unset;
21+
background-color: unset;
22+
border-radius: unset;
23+
}
24+
25+
table.autosummary td {
1926
width: 50%
2027
}

‎docs/en/changelog.md

Lines changed: 10 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,15 @@
11
# Changelog
22

3+
## v0.24.1(31/10/2022)
4+
5+
### New Features
6+
7+
- Support mmcls with NPU backend. ([#1072](https://github.com/open-mmlab/mmclassification/pull/1072))
8+
9+
### Bug Fixes
10+
11+
- Fix performance issue in convnext DDP train. ([#1098](https://github.com/open-mmlab/mmclassification/pull/1098))
12+
313
## v0.24.0(30/9/2022)
414

515
### Highlights

‎docs/en/conf.py

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -48,7 +48,6 @@ def get_version():
4848
'sphinx.ext.intersphinx',
4949
'sphinx.ext.napoleon',
5050
'sphinx.ext.viewcode',
51-
'sphinx_markdown_tables',
5251
'myst_parser',
5352
'sphinx_copybutton',
5453
]

‎docs/en/device/npu.md

Lines changed: 34 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,34 @@
1+
# NPU (HUAWEI Ascend)
2+
3+
## Usage
4+
5+
Please install MMCV with NPU device support according to {external+mmcv:doc}`the tutorial <get_started/build>`.
6+
7+
Here we use 8 NPUs on your computer to train the model with the following command:
8+
9+
```shell
10+
bash tools/dist_train.sh configs/cspnet/resnet50_8xb32_in1k.py 8 --device npu
11+
```
12+
13+
Also, you can use only one NPU to trian the model with the following command:
14+
15+
```shell
16+
python tools/train.py configs/cspnet/resnet50_8xb32_in1k.py --device npu
17+
```
18+
19+
## Verified Models
20+
21+
| Model | Top-1 (%) | Top-5 (%) | Config | Download |
22+
| :--------------------------------------------------------: | :-------: | :-------: | :-----------------------------------------------------------: | :-------------------------------------------------------------: |
23+
| [CSPResNeXt50](../papers/cspnet.md) | 77.10 | 93.55 | [config](https://github.com/open-mmlab/mmclassification/blob/master/configs/cspnet/cspresnext50_8xb32_in1k.py) | [model](<>) \| [log](https://download.openmmlab.com/mmclassification/v0/device/npu/cspresnext50_8xb32_in1k.log.json) |
24+
| [DenseNet121](../papers/densenet.md) | 72.62 | 91.04 | [config](https://github.com/open-mmlab/mmclassification/blob/master/configs/densenet/densenet121_4xb256_in1k.py) | [model](<>) \| [log](https://download.openmmlab.com/mmclassification/v0/device/npu/densenet121_4xb256_in1k.log.json) |
25+
| [EfficientNet-B4(AA + AdvProp)](../papers/efficientnet.md) | 75.55 | 92.86 | [config](https://github.com/open-mmlab/mmclassification/blob/master/configs/efficientnet/efficientnet-b4_8xb32-01norm_in1k.py) | [model](<>) \| [log](https://download.openmmlab.com/mmclassification/v0/device/npu/efficientnet-b4_8xb32-01norm_in1k.log.json) |
26+
| [HRNet-W18](../papers/hrnet.md) | 77.01 | 93.46 | [config](https://github.com/open-mmlab/mmclassification/blob/master/configs/hrnet/hrnet-w18_4xb32_in1k.py) | [model](<>) \| [log](https://download.openmmlab.com/mmclassification/v0/device/npu/hrnet-w18_4xb32_in1k.log.json) |
27+
| [ResNetV1D-152](../papers/resnet.md) | 77.11 | 94.54 | [config](https://github.com/open-mmlab/mmclassification/blob/master/configs/resnet/resnetv1d152_8xb32_in1k.py) | [model](<>) \| [log](<>) |
28+
| [ResNet-50](../papers/resnet.md) | 76.40 | - | [config](https://github.com/open-mmlab/mmclassification/blob/master/configs/resnet/resnet50_8xb32_in1k.py) | [model](<>) \| [log](<>) |
29+
| [ResNetXt-32x4d-50](../papers/resnext.md) | 77.55 | 93.75 | [config](https://github.com/open-mmlab/mmclassification/blob/master/configs/resnext/resnext50-32x4d_8xb32_in1k.py) | [model](<>) \| [log](https://download.openmmlab.com/mmclassification/v0/device/npu/resnext50-32x4d_8xb32_in1k.log.json) |
30+
| [SE-ResNet-50](../papers/seresnet.md) | 77.64 | 93.76 | [config](https://github.com/open-mmlab/mmclassification/blob/master/configs/seresnet/seresnet50_8xb32_in1k.py) | [model](<>) \| [log](https://download.openmmlab.com/mmclassification/v0/device/npu/seresnet50_8xb32_in1k.log.json) |
31+
| [VGG-11](../papers/vgg.md) | 68.92 | 88.83 | [config](https://github.com/open-mmlab/mmclassification/blob/master/configs/vgg/vgg11_8xb32_in1k.py) | [model](<>) \| [log](https://download.openmmlab.com/mmclassification/v0/device/npu/vgg11_8xb32_in1k.log.json) |
32+
| [ShuffleNetV2 1.0x](../papers/shufflenet_v2.md) | 69.53 | 88.82 | [config](https://github.com/open-mmlab/mmclassification/blob/master/configs/shufflenet_v2/shufflenet-v2-1x_16xb64_in1k.py) | [model](<>) \| [log](https://download.openmmlab.com/mmclassification/v0/device/npu/shufflenet-v2-1x_16xb64_in1k.json) |
33+
34+
**All above models are provided by Huawei Ascend group.**

‎docs/en/faq.md

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -17,8 +17,8 @@ and make sure you fill in all required information in the template.
1717

1818
| MMClassification version | MMCV version |
1919
| :----------------------: | :--------------------: |
20-
| dev | mmcv>=1.6.0, \<1.7.0 |
21-
| 0.24.0 (master) | mmcv>=1.4.2, \<1.7.0 |
20+
| dev | mmcv>=1.7.0, \<1.9.0 |
21+
| 0.24.1 (master) | mmcv>=1.4.2, \<1.9.0 |
2222
| 0.23.2 | mmcv>=1.4.2, \<1.7.0 |
2323
| 0.22.1 | mmcv>=1.4.2, \<1.6.0 |
2424
| 0.21.0 | mmcv>=1.4.2, \<=1.5.0 |

‎docs/en/index.rst

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -78,6 +78,13 @@ You can switch between Chinese and English documentation in the lower-left corne
7878
compatibility.md
7979
faq.md
8080

81+
82+
.. toctree::
83+
:maxdepth: 1
84+
:caption: Device Support
85+
86+
device/npu.md
87+
8188
.. toctree::
8289
:caption: Language Switch
8390

‎docs/zh_CN/_static/css/readthedocs.css

Lines changed: 9 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -9,12 +9,19 @@ pre {
99
white-space: pre;
1010
}
1111

12-
article.pytorch-article .section :not(dt) > code {
12+
article.pytorch-article section code {
1313
padding: .2em .4em;
1414
background-color: #f3f4f7;
1515
border-radius: 5px;
1616
}
1717

18-
table.colwidths-auto td {
18+
/* Disable the change in tables */
19+
article.pytorch-article section table code {
20+
padding: unset;
21+
background-color: unset;
22+
border-radius: unset;
23+
}
24+
25+
table.autosummary td {
1926
width: 50%
2027
}

0 commit comments

Comments
 (0)