* [Fix] Fix the bug in binary cross entropy loss
Fix the bug in binary cross entropy loss when using multi-label datasets e.g.VOC2007
* update ci
---------
Co-authored-by: Ezra-Yu <18586273+Ezra-Yu@users.noreply.github.com>
* update ci
* update window ci
* update window ci
* update window ci
* update window ci
* update window ci
* update window ci
* update window ci
* update window ci
* update window ci
* update window ci
* update window ci
* ignore tests/test_tools.py
* [Feature] Add TIMM and HuggingFace wrappers to build classifiers from them directly.
* Support `with_cp` and add docstring.
* Add unit tests.
* Update CI.
* Update docs.
* [CI] Add Circle CI for mmcls 1.x
* Fix circle-ci config
* Use `torch.cat` instead of `torch.hstack` to compat with PyTorch 1.6
* Compat the behavior of LongTensor in PyTorch 1.6
* Avoid random behavior in the unit test of Lighting
* Fix cuda ci
* Remove github workflow temporarily.
* Update workflow to add Windows CI
* Add some optional requirements to requirements/optional.txt
* Update setup.py
* Update PyTorch version in Windows CI
* Update CI and requirement
* Replace `tempfile.NamedTemporaryFile` to avoid unit test error on
Windows
* Update tests
* Update CI
* Add OpenCV installation
* Update CI
* Add wrapper to use backbones from timm
* Add tests
* Remove timm from optional deps and modify GitHub workflow.
Co-authored-by: mzr1996 <mzr1996@163.com>
* Delete the mmcv requirement to support mmcv-full installed environment
* install mmcv in CI
* fix mmcv version error
Co-authored-by: Xinran <lixinran@sensetime.com>
* workflow: modify build, add deploy
* fix type: mmseg -> mmcls
* simplify build by delete cuda related settings
* re-sort requirements by pre-commit hook
* build: support more pytorch versions
* build: fix torchvision version bug
* del useless test imports
* merge upstream
* build: del docstring checking to pass coverage check