Mmcv

What's Changed
* [Feature] Add the support of BallQuery op for Ascend device by Ginray in https://github.com/open-mmlab/mmcv/pull/2963
* [Feature] Add the support of three_interpolate op for Ascend device by lihao7212148 in https://github.com/open-mmlab/mmcv/pull/2962
* Update version info by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2966
* [Docs] Add api reference table by fanqiNO1 in https://github.com/open-mmlab/mmcv/pull/2971
* [Fix] Fix deform_conv.py for torch_npu v2.1 by shun001 in https://github.com/open-mmlab/mmcv/pull/2967
* DIOPI adapt codes for npu device by CokeDong in https://github.com/open-mmlab/mmcv/pull/2973
* [Fix] Fix modulated_deform_conv.py for torch_npu v2.1 by shun001 in https://github.com/open-mmlab/mmcv/pull/2983
* [Feature] Add NPU operator RotatedFeatureAlign by huaweiZJX in https://github.com/open-mmlab/mmcv/pull/2994
* torch_npu support aclnn and add op by momo609 in https://github.com/open-mmlab/mmcv/pull/2998
* [Feature] Add the support for stack_group_points ops with Ascend by 673958639 in https://github.com/open-mmlab/mmcv/pull/3023
* Fix ut of stack_grouping_points by 673958639 in https://github.com/open-mmlab/mmcv/pull/3026
* Add multi_scale_deform_attn op adapter for NPU. by DaGaiBa in https://github.com/open-mmlab/mmcv/pull/3032
* add multi_scale_deform_attn_grad op adapter for NPU by RRaoyzee in https://github.com/open-mmlab/mmcv/pull/3042
* Bump version to 2.2.0 by HAOCHENYE in https://github.com/open-mmlab/mmcv/pull/3087
* add multi npu op by momo609 in https://github.com/open-mmlab/mmcv/pull/3085

New Contributors
* lihao7212148 made their first contribution in https://github.com/open-mmlab/mmcv/pull/2962
* fanqiNO1 made their first contribution in https://github.com/open-mmlab/mmcv/pull/2971
* shun001 made their first contribution in https://github.com/open-mmlab/mmcv/pull/2967
* huaweiZJX made their first contribution in https://github.com/open-mmlab/mmcv/pull/2994

**Full Changelog**: https://github.com/open-mmlab/mmcv/compare/v2.1.0...v2.2.0

&x1F600; **We are glad to announce that we are providing the mmcv pre-built packages for PyTorch 2.1.0 starting from MMCV v2.1.0.**

bash
pip install -U openmim
mim install mmcv


Features

* Add the implementation of dynamic_scatter with mlu-ops by Danielmic in https://github.com/open-mmlab/mmcv/pull/2847
* Add the implementation of diff_iou_rotated with mlu-ops by Danielmic in https://github.com/open-mmlab/mmcv/pull/2840
* Add the support of points_in_polyogns for Ascend device by long11111111 in https://github.com/open-mmlab/mmcv/pull/2848
* Add the support of BoxIouRotated op for ascend device by MrShadowY in https://github.com/open-mmlab/mmcv/pull/2842
* Support modulated_deform_conv and deform_conv with cambricon MLU backend by qipengh in https://github.com/open-mmlab/mmcv/pull/2823
* Add the support for group_points ops with Ascend by 673958639 in https://github.com/open-mmlab/mmcv/pull/2935
* Support torch_npu 2.1 by momo609 in https://github.com/open-mmlab/mmcv/pull/2909
* Adapt new dipu device by CokeDong in https://github.com/open-mmlab/mmcv/pull/2905
* Use PyTorch WARP_SHFL_DOWN macro for half support by zstreet87 in https://github.com/open-mmlab/mmcv/pull/2843

Enhancements

* Support `dipu_mock_cuda=False` in `dipu` for mmcv ext ops with cpu fallback by CokeDong in https://github.com/open-mmlab/mmcv/pull/2839
* Add torch mluops check before calling mluOpsxxx interface by ClowDragon in https://github.com/open-mmlab/mmcv/pull/2871
* Change the order of condition to make fx wok by youkaichao in https://github.com/open-mmlab/mmcv/pull/2883
* Rename fast_conv_bn_eval to efficient_conv_bn_eval by youkaichao in https://github.com/open-mmlab/mmcv/pull/2884
* Replace focal_loss_sigmoid op of MLU backend with mlu-ops by qirun-uiuc in https://github.com/open-mmlab/mmcv/pull/2855
* Replace bbox_overlaps op of MLU backend with mlu-ops by qirun-uiuc in https://github.com/open-mmlab/mmcv/pull/2879
* Replace roipoint_pool3d op of MLU backend with mlu-ops by ClowDragon in https://github.com/open-mmlab/mmcv/pull/2875
* Replace tin_shift op of MLU backend with mlu-ops by ClowDragon in https://github.com/open-mmlab/mmcv/pull/2910
* Use unified interface in Ascend kernel by momo609 in https://github.com/open-mmlab/mmcv/pull/2945

Bug Fixes

* Fix roi_align npu bug by momo609 in https://github.com/open-mmlab/mmcv/pull/2862
* Fix format error in roi_align_rotated by ClowDragon in https://github.com/open-mmlab/mmcv/pull/2887
* Fix data race risk of `cache_randomness` by HarborYuan in https://github.com/open-mmlab/mmcv/pull/2927
* Update mlu-ops version by ClowDragon in https://github.com/open-mmlab/mmcv/pull/2921
* Fix setup.py for torch_npu v2.1 by momo609 in https://github.com/open-mmlab/mmcv/pull/2928
* Fix the support for nms_rotated in Ascend by momo609 in https://github.com/open-mmlab/mmcv/pull/2931
* Fix modulated_deform_conv.py for torch_npu v2.1 by 6Vvv in https://github.com/open-mmlab/mmcv/pull/2941
* Temporarily disable mps ops for torch2.1.0 by grimoire in https://github.com/open-mmlab/mmcv/pull/2958
* Update c++17 in setup.py by grimoire in https://github.com/open-mmlab/mmcv/pull/2865
* Adapt to rocm by xiabo123 in https://github.com/open-mmlab/mmcv/pull/2827

Others

* Install wheel package in CI to support bdist_wheel command by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2851
* Add torch2.1.0 checking in CI by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2955
* Bump version to 2.1.0 by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2959

New Contributors

* 673958639 made their first contribution in https://github.com/open-mmlab/mmcv/pull/2935
* 6Vvv made their first contribution in https://github.com/open-mmlab/mmcv/pull/2941

**Full Changelog**: https://github.com/open-mmlab/mmcv/compare/v2.0.1...v2.1.0

Features

* Add fast_conv_bn_eval option in ConvModule for fast validation and training in Eval mode by youkaichao in https://github.com/open-mmlab/mmcv/pull/2807
* Support mmcv ext with DIOPI impl by CokeDong in https://github.com/open-mmlab/mmcv/pull/2790
* Add the support of arf op for ascend device by dflhw in https://github.com/open-mmlab/mmcv/pull/2792
* Add the support for rotated_feature_align with MLU by tudejiang79 in https://github.com/open-mmlab/mmcv/pull/2809
* Replace the implementation of roi_align_rotated with mlu-ops by tudejiang79 in https://github.com/open-mmlab/mmcv/pull/2808

Enhancements

* Refactor the pytorch and cuda combination for testing in CI by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2693
* Remove two outdated files by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2740
* Update the supported torch and cuda versions in README by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2687
* Remove .owners.yml by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2743
* Enhance issue templates by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2776
* Raise an error if fails to read image by Ezra-Yu in https://github.com/open-mmlab/mmcv/pull/2791
* Make `build_xxx_layer` allow accepting a class type by HAOCHENYE in https://github.com/open-mmlab/mmcv/pull/2782
* Replace carafe op of MLU backend with mlu-ops by qipengh in https://github.com/open-mmlab/mmcv/pull/2817
* Replace the implementation of psa_mask with mlu-ops. by DanieeelLiu in https://github.com/open-mmlab/mmcv/pull/2810
* Simplify the logic of sparse_conv by duzekunKTH in https://github.com/open-mmlab/mmcv/pull/2802
* Drop the outdated torch versions in CI by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2830
* Add ConvModule.turn_on_fast_conv_bn_eval to reduce repetitive code and dynamically bind conv during forward by youkaichao in https://github.com/open-mmlab/mmcv/pull/2835

Bug Fixes

* ops can not be built in github action by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2783
* Build CUDA ops in CI by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2794
* Fix type hint in `_scale_size` and `rescale_size` by MambaWong in https://github.com/open-mmlab/mmcv/pull/2799
* Fix the support for ROCm by xiabo123 in https://github.com/open-mmlab/mmcv/pull/2811
* Keep the shape value of ascend iou op's parameter 2 smaller than parameter 1 by dflhw in https://github.com/open-mmlab/mmcv/pull/2821
* Fix arf op's write conflict when num_orientations is not 1 by dflhw in https://github.com/open-mmlab/mmcv/pull/2824
* Correct mmengine version requirement by wateryzephyr in https://github.com/open-mmlab/mmcv/pull/2816
* Fix deform_conv ops on Ascend NPU by Ginray in https://github.com/open-mmlab/mmcv/pull/2832

Documentations

* Describe branch changes by zhouzaida in https://github.com/open-mmlab/mmcv/pull/2741
* Add npu build.md and update roi_pool_npu adpater by momo609 in https://github.com/open-mmlab/mmcv/pull/2812
* small fix for RandomResize doc by i-aki-y in https://github.com/open-mmlab/mmcv/pull/2829

New Contributors

* MambaWong made their first contribution in https://github.com/open-mmlab/mmcv/pull/2799
* i-aki-y made their first contribution in https://github.com/open-mmlab/mmcv/pull/2829
* wateryzephyr made their first contribution in https://github.com/open-mmlab/mmcv/pull/2816

**Full Changelog**: https://github.com/open-mmlab/mmcv/compare/v2.0.0...v2.0.1

</td>
<td valign="top">

bash
pip install openmim

Contains ops
mim install mmcv

Ops are not included, because the starting version of
mmcv-lite is 2.0.0, so there is no need to add version
restrictions
mim install mmcv-lite


</td>
</tr>
</thead>
</table>

Bug Fixes

- Improved compatibility of training stylegan2 (2694)
- Replace torch/extension.h with torch/types.h to fix compilation error on Windows (2698)
- Fixed torch2.0 dcn/mdcn symbolic (2695)
- Skipped fused_lrelu op when gcc is less than 6.0 or CUDA is less than 10.2 (2671)
- Skipped filtered_lrelu ut when CUDA is less than 10.2 (2677)
- Called operator __half explicitly (2608)
- Ensured fileio related interfaces maintain backward compatibility (2539)
- Fixed CI failure due to protobuf version (2529)
- Corrected _flip_keypoint result of RandomFlip (2527)
- Fixed Windows OS version to 2019 (2503)
- Resolved optimizer 'ValueError' when using PReLU activation (2444)
- Fixed improper assertion in RandomChoiceResize (2450)
- Fixed Correlation op (2274)
- Fixed roi align symbolic for torch>=1.13 (2443)
- Added swap_label_pairs in RandomFlip (2332)
- Fixed MPS ops for torch1.12+ (2302)
- Corrected calculation error of out_w in MaskedConv2dFunction (2264)
- Cast mask type to enable training with amp (2220)

Documentations

- Fixed torch2.0 dcn/mdcn symbolic (2695)
- Skipped fused_lrelu op when gcc is less than 6.0 or CUDA is less than 10.2 (2671)
- Skipped filtered_lrelu ut when CUDA is less than 10.2 (2677)
- Called operator __half explicitly (2608)
- Used official isort (2575)
- Ensured fileio related interfaces maintain backward compatibility (2539)
- Fixed CI failure due to protobuf version (2529)
- Corrected _flip_keypoint result of RandomFlip (2527)
- Fixed Windows OS version to 2019 (2503)
- Resolved optimizer 'ValueError' when using PReLU activation (2444)
- Fixed improper assertion in RandomChoiceResize (2450)
- Fixed Correlation op (2274)
- Fixed roi align symbolic for torch>=1.13 (2443)
- Added swap_label_pairs in RandomFlip (2332)
- Fixed MPS ops for torch1.12+ (2302)
- Corrected calculation error of out_w in MaskedConv2dFunction (2264)
- Cast mask type to enable training with amp (2220)

Contributors

HAOCHENYE grimoire zhouzaida nijkah BainOuO enemy1205 C1rN09 triple-Mu okotaku vansinhu Harold-lkk KeiChiTse xiexinch xin-li-67 VVsssssk LYMDLUT MeowZheng yxzhao2022 mzr1996 ZwwWayne YuanLiuuuuuu ly015 plyfager GT9505 zytx121 hhaAndroid fangyixiao18 JingweiZhang12 xcnick zengyh1900

Features

- Support LayerScale in FFN (2451)
- Add BezierAlign CUDA op (2393)
- Pick Ascend ops from the master branch (2501)

Improvements

- Support value_proj_ratio in MultiScaleDeformableAttention (2452)
- Use the new interface of fileio from mmengine (2468, 2539)

Bug fixes

- Fix optimizer 'ValueError' when using PReLU activation (2444)
- Fix _flip_keypoint of RandomFlip returns a wrong result (2527)

Documentations

- Update the supported PyTorch and CUDA versions (2456)
- Add convention for the order of data transforms (2442)
- Update qq group link (2454)

Others

- Fix the version of Windows OS to 2019 in CI (2503)
- Deprecate the support for python3.6 (2525)
- Fix CI failed due to the version of protobuf (2529)

Contributors
A total of 9 developers contributed to this release.
zhouzaida KeiChiTse Harold-lkk vansin okotaku grimoire HAOCHENYE CokeDong ckirchhoff2021

New Contributors
* KeiChiTse made their first contribution in https://github.com/open-mmlab/mmcv/pull/2444

**Full Changelog**: https://github.com/open-mmlab/mmcv/compare/v2.0.0rc3...v2.0.0rc4

Features
- Add stack ball query and stack group points ops (2292)
- Support RoIAwarePool3d, Iou3d and MultiScaleDeformableAttn with MLU backend (2359, 2339)
- Support NMS with MLU590 backend (2401)

Improvements
- Adapt mlu code to cntoolkit3.0.1

Bug fixes
- Fix roi align symbolic for torch>=1.13 (2443)
- Create Tensor with new_* method to support AMP (2389)
- Fix three nn op can not accept half tensor (2348)
- Fix the potential NaN bug in calc_square_dist (2356)
- Fix Correlation op (2274)
- Fix a not proper assertion in RandomChoiceResize (2450)

Documentations
- Make installation commands compatible with windows CMD (2353)
- Add macOS building notice (2364)
- Refine contributing.md (1941)
- Add code style guide (2406)

Others
- Add torch1.13 checking in CI (2441)

Contributors
A total of 13 developers contributed to this release.
LYMDLUT LYM-fire zhouzaida VVsssssk xin-li-67 grimoire ZCMax ZhangLearning cathyzhang222 HAOCHENYE defei-coder ZShaopeng xiexinch

New Contributors
* LYMDLUT made their first contribution in https://github.com/open-mmlab/mmcv/pull/2353
* xiexinch made their first contribution in https://github.com/open-mmlab/mmcv/pull/2450

**Full Changelog**: https://github.com/open-mmlab/mmcv/compare/v2.0.0rc2...v2.0.0rc3