GitLab

* ms_opt_v2

* ms_opt_v2_1

* optimize MultiScaleDeformableAttention ops for MLU

* ms_opt_v2_1

* [Feature] ms_deform_attn performance optimization V2

* [Feature] ms_deform_attn performance optimization V2

* [Feature] ms_deform_attn performance optimization V2

* [Feature] ms_deform_attn performance optimization V2

* [Feature] ms_deform_attn performance optimization V2

* [Feature] ms_deform_attn performance optimization V2

* [Feature] ms_deform_attn performance optimization V2

---------
Co-authored-by: dongchengwei <dongchengwei@cambricon.com>

* Add verification for scale, when scale less than 0

* Update mmcv/ops/fused_bias_leakyrelu.py
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* Update fused_bias_leakyrelu.py

* Update fused_bias_leakyrelu.py

---------
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* [Enhance] Support FP16 for MSDeformAttn

* [Fix] Data type mismatch

* Update mmcv/ops/multi_scale_deform_attn.py

* Add UT

Author:    nijkah <nijkah@gmail.com>

* Add cuda available condition

---------
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* Add support for Ascend devices with bbox_overlaps

* ...

* ...

* add roipoolgrad adapter

* amend

* Refactor the issue template

* fix links

* fix

Addresses issue https://github.com/open-mmlab/mmcv/issues/2542

For nms_rotated_cuda only adds contiguous call to dets_sorted which is accessed directly.
For nms_rotated_cpu the implementation now matches what's in the detectron2 repo.

* [Docs] Steps to compile mmcv-full on MLU machine

* [Docs] Adjust paragraph order

* Update docs/zh_cn/get_started/build.md
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* Update docs/zh_cn/get_started/build.md
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* Update docs/en/get_started/build.md
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* Update docs/en/get_started/build.md
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* [Docs] Modify the format

---------
Co-authored-by: budefei <budefei@cambricon.com>
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* 📃

 docs(hooks): EvalHook comments

Refine the EvalHook comments

* Refine comments EvalHook

Update mmcv/runner/hooks/evaluation.py
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* 📃

 docs(Hooks): update Eval Hook

Slove line too long

---------
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* add ops bbox_overlaps

* format code

* Return the pytorch version

* Intermediate modification

* Solve problems in parameter passing

* revise bug

* "add test case"

* [Feature]: add nms_rotated npu adaptater code

* [BugFix]: modify param in nms_rotated_npu.cpp

* [clean code]: nms_rotated_npu.cpp

* [clean code]: nms_rotated_npu.cpp

* [clean code]: nms_rotated_npu.cpp

* [clean code]: nms_rotated.cpp

* [Doc]: add nms_rotated op in supported op list at ops.md

* [Test]: add nms_rotated unit_test

* [Bug]: remove device parameter in test_batched_nms function

* [Feature] Support ball_query with cambricon MLU backend and mlu-ops library.

* [Fix] update operator data layout setting.

* [Fix] add cxx compile option to avoid symbol conflict.

* [Fix] fix lint errors.

* [Fix] update ops.md with info of ball_query support by MLU backend.

* [Feature] Fix typo.

* [Fix] Remove print.

* [Fix] get mlu-ops from MMCV_MLU_OPS_PATH env.

* [Fix] update MMCV_MLU_OPS_PATH check logic.

* [Fix] update error info when failed to download mlu-ops.

* [Fix] check mlu-ops version matching info in mmcv.

* [Fix] revise wrong filename.

* [Fix] remove f.close and re.

Co-authored-by: wangxiaoxin_sherie <wangxiaoxin7@huawei.com>

* [Feature] Support hard_voxelize with cambricon MLU backend

* [Feature](bangc-ops): add voxelization op

* [Feature](bangc-ops): add voxelization op

* [Feature](bangc-ops): add voxelization op

* [Feature](bangc-ops): add voxelization op

* [Feature](bangc-ops): add voxelization op

* [Feature](bangc-ops): add voxelization op

* [Feature](bangc-ops): add voxelization op

* [Feature](bangc-ops): add voxelization op

* Update image

* Bugfix for parrots ext ops

* Move parrots ci task to build_pat.yml

* Fix on:push

* Remove commets and fix

* Fix beautiful job name

* debug build_cu102 ci jobs

* debug build_cu102 ci jobs

* add tmate ssh

* add tmate ssh

* install specified version for setuptools

* revert comments of ci jobs

* revert comments of ci jobs

* specify setuptools version for build_cu116

* Fix c/cuda lint failed

* minor refine

* minor refine

* ms_opt

* ms_opt

* ms_opt

* ms_opt

* ms_opt

* [Feature] ms_deform_attn performance optimization

* [Feature] ms_deform_attn performance optimization

* [Feature] ms_deform_attn performance optimization

* update lately npu modification--DCNv1

update lately npu modification--DCNv1

* update lately npu modification--DCNv1

* update lately npu modification--DCNv1

* update lately npu modification--DCNv1

* update lately npu modification--DCNv1

* update lately npu modification--DCNv1

* check code

* Add ops to EN/ZH documents

* Fix parrots compile error(only support aten1.3)

* Fix setup.py for parrots

* judge skip_layer by fullname

* lint

* skip_layer first

* update unit test

* The default block size of rocm is 256.The designed kenel has a block size of 512.Change the block size to 1024 during compilation.

* The default block size of rocm is 256.The designed kenel has a block size of 512.Change the block size to 1024 during compilation.

* The default block size of rocm is 256.The designed kenel has a block size of 512.Change the block size to 1024 during compilation.

* The default block size of rocm is 256.The designed kenel has a block size of 512.Change the block size to 1024 during compilation.

* add npu adapter for deformroipool

* cleancode

* Update deform_roi_pool.cpp

* Update deform_roi_pool.cpp

* Update deform_roi_pool.cpp

* Update deform_roi_pool.cpp

* Update deform_roi_pool.cpp

* Update deform_roi_pool.cpp

* Update deform_roi_pool.cpp

* init npu

* add npu extension and focal loss adapter

* clean code

* clean code

* clean code

* clean code

* fix autocast bugs on npu (#2273)

fix autocast bugs on npu (#2273)

* code format

* code format

* code format

* bug fix

* pytorch_npu_helper.hpp clean code

* Npu dev (#2306)

* fix autocast bugs on npu
* using scatter_kwargs in mmcv.device.scatter_gather

* raise ImportError when compile with npu

* add npu test case (#2307)

* add npu test case

* Update focal_loss.py

* add comment

* clean lint

* update dtype assert

* update DDP forward and comment

* fix bug
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* sigmoidfocalloss npu adapter bug fix

* BugFix: modify softmaxFocalLoss adapter

* BugFix: remove equal sign in the code

* add npu install information in README

* add modulatedDeformConv npu adapter

* init npu

* add npu extension and focal loss adapter

* clean code

* clean code

* clean code

* add modulatedDeformConv npu adapter

* merge master branch 20221103

* Add masked_ Conv2d operator in NPU

* add nms_npu

* fix bug

* fix code check

* fix code check

* fix code check

* Masked_conv2d NPU

* Masked_conv2d NPU

* Masked_conv2d NPU

* remove npu-install-info in README.md

* annotate the clang-format in pre-commit-config-zh-ch.yaml

* Clean code: fix the clean code problem in masked_conv2d and modulated_deform_conv

* Create fused_bias_leakyrelu_npu.cpp

Add NPU adapter for fused_bias_leaky_relu operator

* Update fused_bias_leakyrelu_npu.cpp

* Update fused_bias_leakyrelu_npu.cpp

* Update ops.md

* Update ops.md

* Update fused_bias_leakyrelu_npu.cpp

* Update fused_bias_leakyrelu_npu.cpp

* Update test_fused_bias_leakyrelu.py

* Update fused_bias_leakyrelu.py

* Update test_fused_bias_leakyrelu.py

* Update fused_bias_leakyrelu.py

* Update test_fused_bias_leakyrelu.py

* Update ops.md

* amend for CI

* bugfix

* amend ops.md

* Update test_fused_bias_leakyrelu.py

* clean code

* bugfix

* clean code

* Update fused_bias_leakyrelu_npu.cpp

* Update fused_bias_leakyrelu_npu.cpp
Co-authored-by: wangjiangben <wangjiangben_hw@126.com>
Co-authored-by: ckirchhoff <515629648@qq.com>
Co-authored-by: wangjiangben-hw <111729245+wangjiangben-hw@users.noreply.github.com>
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
Co-authored-by: zcc-zjut <zcczxy2019@163.com>
Co-authored-by: wangxiaoxin_sherie <wangxiaoxin7@huawei.com>
Co-authored-by: momo609 <963372609@qq.com>