Mmselfsup

We are excited to announce the release of MMSelfSup v1.0.0rc1.
MMSelfSup v1.0.0rc1 is the first version of MMSelfSup 1.x, a part of the OpenMMLab 2.0 projects.
The `master` branch is still 0.x version and we will checkout a new `1.x` branch to release 1.x version. The two versions will be maintained simultaneously in the future.

We briefly list the major breaking changes here. Please refer to the [migration guide](https://github.com/open-mmlab/mmselfsup/blob/1.x/docs/en/migration.md) for details and migration instructions.

Highlight

- Based on [MMEngine](https://github.com/open-mmlab/mmengine) and [MMCV](https://github.com/open-mmlab/mmcv).
- Released with refactor.
- Datasets
- Models
- Config
- ...
- Refine all documents.

New Features

- Add `SelfSupDataSample` to unify the components' interface.
- Add `SelfSupVisualizer` for visualization.
- Add `SelfSupDataPreprocessor` for data preprocess in model.

Improvements

- Most algorithms now support non-distributed training.
- Change the interface of different data augmentation transforms to `dict`.
- Run classification downstream task with MMClassification.

Docs

- Refine all documents and reorganize the directory.
- Add concepts for different components.

We are excited to announce the release of MMSelfSup v1.0.0rc0.
MMSelfSup v1.0.0rc0 is the first version of MMSelfSup 1.x, a part of the OpenMMLab 2.x projects.
The `master` branch is still 0.x version and we will checkout a new `1.x` branch to release 1.x version. The two versions will be maintained simultaneously in the future.

We briefly list the major breaking changes here. Please refer to the [migration guide](https://github.com/open-mmlab/mmselfsup/blob/1.x/docs/en/migration.md) for details and migration instructions.

Highlight

- Based on [MMEngine](https://github.com/open-mmlab/mmengine) and [MMCV](https://github.com/open-mmlab/mmcv).
- Released with refactor.
- Datasets
- Models
- Config
- ...
- Refine all documents.

New Features

- Add `SelfSupDataSample` to unify the components' interface.
- Add `SelfSupVisualizer` for visualization.
- Add `SelfSupDataPreprocessor` for data preprocess in model.

Improvements

- Most algorithms now support non-distributed training.
- Change the interface of different data augmentation transforms to `dict`.
- Run classification downstream task with MMClassification.

Docs

- Refine all documents and reorganize the directory.
- Add concepts for different components.

New Features

- Support InterCLR ([609](https://github.com/open-mmlab/mmselfsup/pull/609))

Bug Fixes

- Fix potential bug of hook registration ([647](https://github.com/open-mmlab/mmselfsup/pull/647))
- Fix sampling_replace config kwargs bug ([646](https://github.com/open-mmlab/mmselfsup/pull/646))
- Change sklearn to scikit-learn in requirements ([583](https://github.com/open-mmlab/mmselfsup/pull/583))

Improvements

- Update CI check rules ([581](https://github.com/open-mmlab/mmselfsup/pull/581))
- Update assignee schedule ([606](https://github.com/open-mmlab/mmselfsup/pull/606))

Docs

- Add global notes and the version switcher menu ([573](https://github.com/open-mmlab/mmselfsup/pull/573))


Contributors

A total of 3 developers contributed to this release.
Thanks fangyixiao18 , Jiahao000 , bencwallace

Improvements

- Update issue template and management file ([550](https://github.com/open-mmlab/mmselfsup/pull/550), [#512](https://github.com/open-mmlab/mmselfsup/pull/512))
- Move res-layer to models.utils ([537](https://github.com/open-mmlab/mmselfsup/pull/537))

Docs

- Add MMYOLO description for README ([541](https://github.com/open-mmlab/mmselfsup/pull/541))
- Fix typo for README ([545](https://github.com/open-mmlab/mmselfsup/pull/545))
- Fix lint of MaskFeat ([520](https://github.com/open-mmlab/mmselfsup/pull/520))

Contributor

A total of 3 developers contributed to this release.
Thanks fangyixiao18 , isLinXu , nijkah

Highlight

- Support MaskFeat ([485](https://github.com/open-mmlab/mmselfsup/pull/485))
- Update README to announce 1.0.0rc version ([474](https://github.com/open-mmlab/mmselfsup/pull/474))

New Features

- Support MaskFeat ([485](https://github.com/open-mmlab/mmselfsup/pull/485))

Bug Fixes

- Fix DenseCL init weights bug ([411](https://github.com/open-mmlab/mmselfsup/pull/411))
- Fix norm typo in config ([418](https://github.com/open-mmlab/mmselfsup/pull/418))
- Fix read image bug ([386](https://github.com/open-mmlab/mmselfsup/pull/386))

Improvements

- Change hook_cfg type access ([409](https://github.com/open-mmlab/mmselfsup/pull/409))
- Support to dump training config ([410](https://github.com/open-mmlab/mmselfsup/pull/410))
- Support to save MAE visualization results ([388](https://github.com/open-mmlab/mmselfsup/pull/388))
- Remove default value of deprecated option ([490](https://github.com/open-mmlab/mmselfsup/pull/490))

Docs

- Update the link of MAE ([497](https://github.com/open-mmlab/mmselfsup/pull/497))
- Update README to announce 1.0.0rc version ([474](https://github.com/open-mmlab/mmselfsup/pull/474))
- Update get_started.md ([402](https://github.com/open-mmlab/mmselfsup/pull/402))
- Update model zoo ([499](https://github.com/open-mmlab/mmselfsup/pull/499))

Contributors

A total of 8 developers contributed to this release.
Thanks fangyixiao18 , YuanLiuuuuuu , soonera , Jason-Study , daidaiershidi , lorinczszabolcs , jingt2ch , Happylkx

New Features

- Support MAE Reconstructed Image Visualization ([376](https://github.com/open-mmlab/mmselfsup/pull/376))

Bug Fixes

- Fix args/cfg bug in extract.py, use cfg.work_dir to save files ([357](https://github.com/open-mmlab/mmselfsup/pull/357))
- Fix SimMIM mask generator config bug ([360](https://github.com/open-mmlab/mmselfsup/pull/360))

Improvements

- Update mdformat settings ([323](https://github.com/open-mmlab/mmselfsup/pull/323))
- Add circleci ([374](https://github.com/open-mmlab/mmselfsup/pull/374))

Docs

- Fix the link of switch language ([327](https://github.com/open-mmlab/mmselfsup/pull/327))
- Update lr_updater.py links in tutorials/4_schedule.md ([354](https://github.com/open-mmlab/mmselfsup/pull/354))

Contributors

A total of 4 developers contributed to this release.
Thanks fangyixiao18 , YuanLiuuuuuu , soonera , doem97