Merge commit 'refs/pull/622/head' of https://github.com/open-mmlab/mm…

…engine

.github/workflows/merge_stage_test.yml

@@ -159,7 +159,8 @@ jobs:
         with:
           python-version: ${{ matrix.python-version }}
       - name: Upgrade pip
-        run: pip install pip --upgrade
+        # Windows CI could fail If we call `pip install pip --upgrade` directly.
+        run: python -m pip install pip --upgrade
       - name: Install PyTorch
         run: pip install torch==1.8.1+${{matrix.platform}} torchvision==0.9.1+${{matrix.platform}} -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html
       - name: Build MMEngine from source

.github/workflows/pr_stage_test.yml

@@ -108,7 +108,8 @@ jobs:
         with:
           python-version: ${{ matrix.python-version }}
       - name: Upgrade pip
-        run: pip install pip --upgrade
+        # Windows CI could fail If we call `pip install pip --upgrade` directly.
+        run: python -m pip install pip --upgrade
       - name: Install PyTorch
         run: pip install torch==1.8.1+${{matrix.platform}} torchvision==0.9.1+${{matrix.platform}} -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html
       - name: Build MMEngine from source

README.md

@@ -56,6 +56,17 @@ Major features:
    - Provides rich components and strategies.
    - Complete controls on the training process with different levels of APIs.
 
+## What's New
+
+v0.2.0 was released in 2022-10-11.
+
+1. Refactor FileIO to provide a more unified interface without BC breaking.
+2. Add the base model for test time augmentation.
+3. Support converting `BN` to `SyncBN` during distributed data-parallel training.
+4. Add SMDDP backend and support running on AWS.
+
+Read [Changelog](./docs/en/notes/changelog.md#v0.2.0 "11/10/2022") for more details.
+
 ## Installation
 
 Before installing MMEngine, please ensure that PyTorch has been successfully installed following the [official guide](https://pytorch.org/get-started/locally/).
@@ -145,7 +156,7 @@ val_dataloader = DataLoader(batch_size=32,
 <details>
 <summary>Build Metrics</summary>
 
-To validate and test the model, we need to define a **Metric** called accuracy to evaluate the model. This metric needs inherit from `BaseMetric` and implements the `process` and `compute_metrics` methods.
+To validate and test the model, we need to define a **Metric** called accuracy to evaluate the model. This metric needs to inherit from `BaseMetric` and implements the `process` and `compute_metrics` methods.
 
 ```python
 from mmengine.evaluator import BaseMetric

README_zh-CN.md

@@ -58,6 +58,17 @@ MMEngine 是一个基于 PyTorch 用于深度学习模型训练的基础库，
    - 对神经网络模型中各个层的优化配置进行细粒度调整
    - 混合精度训练的灵活控制
 
+## 最近进展
+
+最新版本 v0.2.0 在 2022.10.11 发布。
+
+1. 重构 FileIO 以提供更加易用的接口并保持向下兼容。
+2. 新增 Test time augmentation 模型基类。
+3. 分布式训练时，支持将 BN 自动转化为 SyncBN。
+4. 新增了 SMDDP 后端并支持在 AWS 进行分布式训练。
+
+如果想了解更多版本更新细节和历史信息，请阅读[更新日志](./docs/en/notes/changelog.md#v0.2.0 "11/10/2022")
+
 ## 安装
 
 在安装 MMengine 之前，请确保 PyTorch 已成功安装在环境中，可以参考 [PyTorch 官方安装文档](https://pytorch.org/get-started/locally/)。

docs/en/index.rst

@@ -89,6 +89,12 @@ You can switch between Chinese and English documents in the lower-left corner of
    mmengine.utils <api/utils>
    mmengine.utils.dl_utils <api/utils.dl_utils>
 
+.. toctree::
+   :maxdepth: 2
+   :caption: Notes
+
+   notes/changelog.md
+
 .. toctree::
    :caption: Switch Language
 

docs/en/notes/changelog.md

@@ -0,0 +1,57 @@
+# Changelog of v0.x
+
+## v0.2.0 (11/10/2022)
+
+### New Features & Enhancements
+
+- Add SMDDP backend and support running on AWS by @austinmw in https://github.com/open-mmlab/mmengine/pull/579
+- Refactor `FileIO` but without breaking bc by @zhouzaida in https://github.com/open-mmlab/mmengine/pull/533
+- Add test time augmentation base model by @HAOCHENYE in https://github.com/open-mmlab/mmengine/pull/538
+- Use `torch.lerp\_()` to speed up EMA by @RangiLyu in https://github.com/open-mmlab/mmengine/pull/519
+- Support converting `BN` to `SyncBN` by config by @HAOCHENYE in https://github.com/open-mmlab/mmengine/pull/506
+- Support defining metric name in wandb backend by @okotaku in https://github.com/open-mmlab/mmengine/pull/509
+- Add dockerfile by @zhouzaida in https://github.com/open-mmlab/mmengine/pull/347
+
+### Docs
+
+- Fix API files of English documentation by @zhouzaida in https://github.com/open-mmlab/mmengine/pull/525
+- Fix typo in `instance_data.py` by @Dai-Wenxun in https://github.com/open-mmlab/mmengine/pull/530
+- Fix the docstring of the model sub-package by @zhouzaida in https://github.com/open-mmlab/mmengine/pull/573
+- Fix a spelling error in docs/zh_cn by @cxiang26 in https://github.com/open-mmlab/mmengine/pull/548
+- Fix typo in docstring by @MengzhangLI in https://github.com/open-mmlab/mmengine/pull/527
+- Update `config.md` by @Zhengfei-0311 in https://github.com/open-mmlab/mmengine/pull/562
+
+### Bug Fixes
+
+- Fix `LogProcessor` does not smooth loss if the name of loss doesn't start with `loss` by @liuyanyi in
+  https://github.com/open-mmlab/mmengine/pull/539
+- Fix failed to enable `detect_anomalous_params` in `MMSeparateDistributedDataParallel` by @HAOCHENYE in https://github.com/open-mmlab/mmengine/pull/588
+- Fix CheckpointHook behavior unexpected if given `filename_tmpl` argument by @C1rN09 in https://github.com/open-mmlab/mmengine/pull/518
+- Fix error argument sequence in `FSDP` by @HAOCHENYE in https://github.com/open-mmlab/mmengine/pull/520
+- Fix uploading image in wandb backend @okotaku in https://github.com/open-mmlab/mmengine/pull/510
+- Fix loading state dictionary in `EMAHook` by @okotaku in https://github.com/open-mmlab/mmengine/pull/507
+- Fix circle import in `EMAHook` by @HAOCHENYE in https://github.com/open-mmlab/mmengine/pull/523
+- Fix unit test could fail caused by `MultiProcessTestCase`  by @HAOCHENYE in https://github.com/open-mmlab/mmengine/pull/535
+- Remove unnecessary "if statement" in `Registry` by @MambaWong in https://github.com/open-mmlab/mmengine/pull/536
+- Fix `_save_to_state_dict` by @HAOCHENYE in https://github.com/open-mmlab/mmengine/pull/542
+- Support comparing NumPy array dataset meta in `Runner.resume` by @HAOCHENYE in https://github.com/open-mmlab/mmengine/pull/511
+- Use `get` instead of `pop` to dump `runner_type` in `build_runner_from_cfg` by @nijkah in https://github.com/open-mmlab/mmengine/pull/549
+- Upgrade pre-commit hooks by @zhouzaida in https://github.com/open-mmlab/mmengine/pull/576
+- Delete the error comment in `registry.md` by @vansin in https://github.com/open-mmlab/mmengine/pull/514
+- Fix Some out-of-date unit tests by @C1rN09 in https://github.com/open-mmlab/mmengine/pull/586
+- Fix typo in `MMFullyShardedDataParallel` by @yhna940 in https://github.com/open-mmlab/mmengine/pull/569
+- Update Github Action CI and CircleCI by @zhouzaida in https://github.com/open-mmlab/mmengine/pull/512
+- Fix unit test in windows by @HAOCHENYE in https://github.com/open-mmlab/mmengine/pull/515
+- Fix merge ci & multiprocessing unit test by @HAOCHENYE in https://github.com/open-mmlab/mmengine/pull/529
+
+### New Contributors
+
+- @okotaku made their first contribution in https://github.com/open-mmlab/mmengine/pull/510
+- @MengzhangLI made their first contribution in https://github.com/open-mmlab/mmengine/pull/527
+- @MambaWong made their first contribution in https://github.com/open-mmlab/mmengine/pull/536
+- @cxiang26 made their first contribution in https://github.com/open-mmlab/mmengine/pull/548
+- @nijkah made their first contribution in https://github.com/open-mmlab/mmengine/pull/549
+- @Zhengfei-0311 made their first contribution in https://github.com/open-mmlab/mmengine/pull/562
+- @austinmw made their first contribution in https://github.com/open-mmlab/mmengine/pull/579
+- @yhna940 made their first contribution in https://github.com/open-mmlab/mmengine/pull/569
+- @liuyanyi made their first contribution in https://github.com/open-mmlab/mmengine/pull/539

docs/zh_cn/index.rst

@@ -89,6 +89,12 @@
    mmengine.utils <api/utils>
    mmengine.utils.dl_utils <api/utils.dl_utils>
 
+.. toctree::
+   :maxdepth: 2
+   :caption: Notes
+
+   notes/changelog.md
+
 .. toctree::
    :caption: 语言切换
 

docs/zh_cn/tutorials/config.md

@@ -308,15 +308,15 @@ print(cfg.a)
 {'type': 'MobileNet', 'depth': 50}
 ```
 
-解析后发现，`a` 的 type 变成了 `MobileNet。`
+解析后发现，`a` 的 type 变成了 `MobileNet`。
 
 ## 配置文件的导出
 
 在启动训练脚本时，用户可能通过传参的方式来修改配置文件的部分字段，为此我们提供了 `dump`
 接口来导出更改后的配置文件。与读取配置文件类似，用户可以通过 `cfg.dump('config.xxx')` 来选择导出文件的格式。`dump`
 同样可以导出有继承关系的配置文件，导出的文件可以被独立使用，不再依赖于 `_base_` 中定义的文件。
 
-基于继承一节定义的 `resnet50.py`,我们将其加载后导出:
+基于继承一节定义的 `resnet50.py`，我们将其加载后导出：
 
 ```python
 cfg = Config.fromfile('resnet50.py')
@@ -394,14 +394,14 @@ print(cfg.work_dir)
 
 - `{{fileDirname}}` - 当前文件的目录名，例如 `/home/your-username/your-project/folder`
 - `{{fileBasename}}` - 当前文件的文件名，例如 `file.py`
-- `{{fileBasenameNoExtension}}` - 当前文件不包含扩展名的文件名，例如 file
+- `{{fileBasenameNoExtension}}` - 当前文件不包含扩展名的文件名，例如 `file`
 - `{{fileExtname}}` - 当前文件的扩展名，例如 `.py`
 
 ### 命令行修改配置
 
 有时候我们只希望修改部分配置，而不想修改配置文件本身，例如实验过程中想更换学习率，但是又不想重新写一个配置文件，常用的做法是在命令行传入参数来覆盖相关配置。考虑到我们想修改的配置通常是一些内层参数，如优化器的学习率、模型卷积层的通道数等，因此 MMEngine 提供了一套标准的流程，让我们能够在命令行里轻松修改配置文件中任意层级的参数。
 
-1. 使用 `argparser` 解析脚本运行的参数
+1. 使用 `argparse` 解析脚本运行的参数
 2. 使用 `argparse.ArgumentParser.add_argument` 方法时，让 `action` 参数的值为 [DictAction](mmengine.config.DictAction)，用它来进一步解析命令行参数中用于修改配置文件的参数
 3. 使用配置类的 `merge_from_dict` 方法来更新配置
 

docs/zh_cn/tutorials/model.md

@@ -29,7 +29,7 @@ from torchvision import datasets
 from torchvision.transforms import ToTensor
 from mmengine.model import BaseModel
 from mmengine.evaluator import BaseMetric
-from mmengine import Runner
+from mmengine.runner import Runner
 
 
 training_data = datasets.FashionMNIST(

mmengine/version.py

@@ -1,6 +1,6 @@
 # Copyright (c) OpenMMLab. All rights reserved.
 
-__version__ = '0.1.0'
+__version__ = '0.2.0'
 
 
 def parse_version_info(version_str):