9f43b8c310
add npu roipoint_pool3d_forward ( #3147 )
...
* add npu roipoint_pool3d_forward
* nms_rotated npu add new attr is_angle
2024-07-19 01:33:34 +08:00
b91cfded58
[Feature] add group_points support npu ( #3120 )
2024-06-04 14:35:39 +08:00
210090075e
[Fix] Fix chamfer_distance cuda bug len should div to 4 ( #3113 )
2024-05-23 21:17:31 +08:00
6a1646f1a3
Support fps and points_in_box ops for Ascend device ( #3085 )
2024-04-23 17:31:51 +08:00
780ffed9f3
Add multi_scale_deform_attn_grad op adapter for NPU ( #3042 )
2024-03-13 15:15:46 +08:00
cd05d71254
Add multi_scale_deform_attn op adapter for NPU ( #3032 )
2024-03-01 15:02:23 +08:00
d9e10e1184
Fix ut of stack_grouping_points ( #3026 )
2024-01-29 19:18:01 +08:00
815f5d75a7
[Feature] Add the support for stack_group_points ops with Ascend ( #3023 )
2024-01-29 17:05:23 +08:00
c7c02a7c5b
[Feature] torch_npu support aclnn and add op ( #2998 )
2024-01-07 17:11:36 +08:00
2e44eaeba3
[Feature] Add NPU operator RotatedFeatureAlign ( #2994 )
2023-11-29 17:24:28 +08:00
94dff26554
[Feature] Add the support of three_interpolate op for Ascend device ( #2962 )
2023-10-17 10:40:09 +08:00
c0774b52a8
[Feature] Add the support of BallQuery op for Ascend device ( #2963 )
2023-10-13 14:56:56 +08:00
cac0c44a75
temporarily disable mps ops for torch2.1.0 ( #2958 )
2023-10-11 13:08:35 +08:00
8523eeefac
[Feature] Add the support for group_points ops with Ascend ( #2935 )
2023-09-20 11:00:58 +08:00
9036241eac
[Enhancement] Change the order of condition to make fx wok ( #2883 )
2023-08-03 17:29:07 +08:00
f64d485843
rename fast_conv_bn_eval to efficient_conv_bn_eval ( #2884 )
2023-08-03 16:48:32 +08:00
86a38aa3f0
[Feature] Support modulated_deform_conv and deform_conv with cambricon MLU backend ( #2823 )
2023-07-27 16:28:42 +08:00
987d34b0cf
[Feature] Add the support of BoxIouRotated op for ascend device ( #2842 )
...
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
2023-07-19 10:36:26 +08:00
43c5c76f9b
[Feature] Add support of points_in_polyogns for Ascend device ( #2848 )
2023-07-06 15:34:51 +08:00
d28aa8a9cc
[Feature] Add the implementation of diff_iou_rotated with mlu-ops ( #2840 )
2023-06-30 15:18:39 +08:00
10c8b9e78b
[Feature] Add the implementation of dynamic_scatter with mlu-ops ( #2847 )
2023-06-29 17:13:57 +08:00
3df6414efa
[Enhance] Add ConvModule.turn_on_fast_conv_bn_eval to reduce repetitive code and dynamically bind conv during forward ( #2835 )
2023-06-16 12:43:53 +08:00
36003b7617
[Feature] Add fast_conv_bn_eval option in ConvModule for fast validation and training in Eval mode ( #2807 )
2023-06-13 23:13:01 +08:00
9b1209fadb
[Feature] Support mmcv ext with DIOPI impl ( #2790 )
2023-06-13 17:14:49 +08:00
ba8aa76458
[Refactor] Repalce the implementation of roi_align_rotated with mlu-ops ( #2808 )
2023-06-01 00:56:49 +08:00
d2aecbe494
[Feature] Add the support for rotated_feature_align with MLU ( #2809 )
2023-06-01 00:55:20 +08:00
7ff7095ca7
[Fix] Fix the support for ROCm ( #2811 )
2023-05-31 12:47:46 +08:00
0c23eb0215
Sync main with mmcv1.x branch ( #2800 )
2023-05-19 15:40:12 +08:00
59c1418e45
[Enhancement] Make build_xxx_layer allow accepting a class type ( #2782 )
2023-05-11 17:51:13 +08:00
9a5685a995
[Feature] Add the support of arf op for ascend device ( #2792 )
2023-05-10 17:07:17 +08:00
d493a5d523
Pick changes from 1.x branch ( #2738 )
2023-04-20 19:27:11 +08:00
733e6ff84e
Pick MLU modifications from master (1.x) to main (2.x) ( #2704 )
...
* [Feature] Support Voxelization with cambricon MLU device (#2500 )
* [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
* [Enhance] Optimize the performace of ms_deform_attn for MLU device (#2510 )
* 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
* [Feature] Support ball_query with cambricon MLU backend and mlu-ops library. (#2520 )
* [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.
* [Docs] Steps to compile mmcv-full on MLU machine (#2571 )
* [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>
* [Fix] Fix tensor descriptor setting in MLU ball_query. (#2579 )
* [Feature] Add MLU support for Sparse Convolution op (#2589 )
* [Feature] Add sparse convolution MLU API
* [Feature] update cpp code style
* end-of-file
* delete libext.a
* code style
* update ops.md
---------
Co-authored-by: budefei <budefei@cambricon.com>
* [Enhancement] Replace the implementation of deform_roi_pool with mlu-ops (#2598 )
* [Feature] Replace the implementation of deform_roi_pool with mlu-ops
* [Feature] Modify code
---------
Co-authored-by: budefei <budefei@cambricon.com>
* [Enhancement] ms_deform_attn performance optimization (#2616 )
* 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>
* [Feature] Support NmsRotated with cambricon MLU backend (#2643 )
* [Feature] Support NmsRotated with cambricon MLU backend
* [Feature] remove foolproofs in nms_rotated_mlu.cpp
* [Feature] fix lint in test_nms_rotated.py
* [Feature] fix kMLU not found in nms_rotated.cpp
* [Feature] modify mlu support in nms.py
* [Feature] modify nms_rotated support in ops.md
* [Feature] modify ops/nms.py
* [Enhance] Add a default value for MMCV_MLU_ARGS (#2688 )
* add mlu_args
* add mlu_args
* Modify the code
---------
Co-authored-by: budefei <budefei@cambricon.com>
* [Enhance] Ignore mlu-ops files (#2691 )
Co-authored-by: budefei <budefei@cambricon.com>
---------
Co-authored-by: ZShaopeng <108382403+ZShaopeng@users.noreply.github.com>
Co-authored-by: BinZheng <38182684+Wickyzheng@users.noreply.github.com>
Co-authored-by: liuduanhui <103939338+DanieeelLiu@users.noreply.github.com>
Co-authored-by: budefei <budefei@cambricon.com>
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
Co-authored-by: duzekun <108381389+duzekunKTH@users.noreply.github.com>
Co-authored-by: dongchengwei <dongchengwei@cambricon.com>
Co-authored-by: liuyuan1-v <125547457+liuyuan1-v@users.noreply.github.com>
2023-04-19 10:42:07 +08:00
1db3967e86
Fix lint ( #2702 )
2023-03-26 02:06:28 +08:00
5b0cdccc5c
[Fix] Enhance the compatibility of training stylegan 2 ( #2694 )
...
* [Fix] Enhance the compatibility of training stylegan 2
* Fix unit test in PyTorch 2.0
* Apply suggestions from code review
---------
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
2023-03-26 01:58:38 +08:00
a6d7dde77f
[Fix] Fix torch2.0 dcn/mdcn symbolic ( #2695 )
...
* fix
* fix lint
2023-03-24 13:01:56 +08:00
c458829763
Refine rfsearch and fix a typo
2023-03-20 18:26:52 +08:00
4ad79591f9
[Enhancement] Support MultiScaleDeformableAttention with AMP ( #2541 )
...
* [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>
2023-03-20 18:26:52 +08:00
0e14ce2c1d
[Feature] Add C++ implementation for bbox_overlaps ( #2477 )
...
* add ops bbox_overlaps
* format code
* Return the pytorch version
* Intermediate modification
* Solve problems in parameter passing
* revise bug
* "add test case"
2023-03-20 18:26:52 +08:00
bc10b5f0c0
[Feature] Support receptive field search of CNN models ( #2056 )
...
* support rfsearch
* add labs for rfsearch
* format
* format
* add docstring and type hints
* clean code
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* rm unused func
* update code
* update code
* update code
* update details
* fix details
* support asymmetric kernel
* support asymmetric kernel
* Apply suggestions from code review
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* Apply suggestions from code review
* Apply suggestions from code review
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* Apply suggestions from code review
* Apply suggestions from code review
* Apply suggestions from code review
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* Apply suggestions from code review
* add unit tests for rfsearch
* set device for Conv2dRFSearchOp
* Apply suggestions from code review
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* remove unused function search_estimate_only
* move unit tests
* Update tests/test_cnn/test_rfsearch/test_operator.py
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* Apply suggestions from code review
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* Update mmcv/cnn/rfsearch/operator.py
Co-authored-by: Yue Zhou <592267829@qq.com>
* change logger
* Update mmcv/cnn/rfsearch/operator.py
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
Co-authored-by: lzyhha <819814373@qq.com>
Co-authored-by: Zhongyu Li <44114862+lzyhha@users.noreply.github.com>
Co-authored-by: Yue Zhou <592267829@qq.com>
[Fix] Fix skip_layer for RF-Next (#2489 )
* judge skip_layer by fullname
* lint
* skip_layer first
* update unit test
2023-03-20 18:26:52 +08:00
e7adffb9e0
[Fix] Skip filtered_lrelu ut when cuda is less than 10.2 ( #2677 )
2023-03-19 16:17:45 +08:00
869dbf1bf2
[Feature] Add Ops of StyleGAN3 ( #2290 )
...
* add bias_act
* support bias_act
* support filtered_lrelu
* support filtered_lrelu and upfirdn2d
* support conv2d_gradfix and fix filtered_lrelu
* fix lint
* fix lint
* fix c++ lint
* fix part comments
* fix lint
* rm redundant header
* fix upgrade pip
* fix as comment
* fix c++ lint
* fix ci
* fix-ut
* fix as comments
* add grad check
* remove redundant template
* Update mmcv/ops/bias_act.py
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* add typehint
* fix as comment:
* complete type hints
* fix lint
* add test for conv_gradfix
* add test for conv_gradfix
* fix lint
* modify licenses and ops.md
* add zh op md
* add torch version policy for conv2d_gradfix
* fix lint
* fix as comments
* rename impl
* rm redudant function and add ut
* fix as comment
* fix lint
* fix lint
* fix as comments
* fix lint
* fix ut
* fix as comment
* fix as comment
* fix as comment
---------
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
2023-03-13 16:05:11 +08:00
0fb07d0e69
[Feature] Add the support of voxelization op for ascend device ( #2614 )
...
* Add voxelization op npu adaptor
* update
* update
2023-03-03 20:07:11 +08:00
834f94db23
[Feature] Add support of some ops for Ascend device ( #2594 )
...
* [Feature]: add supports of gather_points, nms_rotated, bbox_overlaps for Ascend device
* Apply suggestions from code review
---------
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
2023-02-13 16:43:42 +08:00
009cfe602d
fileio related interfaces should keep bc ( #2539 )
...
* fileio related interfaces should keep bc
* fix format
* update unit tests of transforms
* update format
* lazily infer client until calling get function of fileio
2023-01-17 13:10:12 +08:00
34bdf448ac
[Feature] Pick npu ops from master to 2.x ( #2501 )
...
* merge npu ops from master to 2.x
* BugFix: fix merge bugs
* {[Feature]: add psamask, roipool to 2.x, and fix the SigmoidFocalLoss assert condition
* merge conflicts in ops.md
* [fix]: fix merge bug
2023-01-13 13:39:35 +08:00
7156604eba
[Feature] Add BezierAlign CUDA op ( #2393 )
...
* bezier align
* add ut
* fix comment
* updata ut
* fix link and comment
* fix comment
2023-01-10 17:05:39 +08:00
a3cb068590
[Refactor] Use the new interface of fileio from mmengine ( #2468 )
...
* [Refactor] Use new interface of fileio
* update ut
* remove file_client_args
* add * before backend_args
* refine backend_args description
2022-12-28 00:09:05 +08:00
d510b8b174
[Feature] Support LayerScale in FFN ( #2451 )
...
* add layer scale
* add layer scale
* add layer scale
* Update mmcv/cnn/bricks/transformer.py
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* Update mmcv/cnn/bricks/transformer.py
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* add layer scale
* move LayerScale
* add layer_scale_init_value
* add typehint
* fix for tensor with any dim
* fix layer scale rule
* fix layer scale rule
* fix test
* add docs
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
2022-12-11 17:58:19 +08:00
433607030a
[Enhancement] Support value_proj_ratio in MultiScaleDeformableAttention ( #2452 )
...
* add ratio in ms_deform_attn_
* add ratio in ms_deform_attn
* Update mmcv/ops/multi_scale_deform_attn.py
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* Update tests/test_ops/test_ms_deformable_attn.py
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
* add ratio in ms_deform_attn
* add ratio in ms_deform_attn
* add ratio in ms_deform_attn
* add ratio in ms_deform_attn
* add ratio in ms_deform_attn
* add ratio in ms_deform_attn
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
2022-12-11 17:48:21 +08:00
a0939977a3
[Feature] Support MultiScaleDeformableAttn with cambricon MLU backend
2022-11-23 16:57:35 +08:00
huhongsun
wujiadi1
Annarine
sherie
RRaoyzee
DaGaiBa
Chai
huaweiZJX
lihao7212148
Yinlei Sun
q.yao
youkaichao
qipengh
MrShadowY
long11111111
Danielmic
CokeDong
tudejiang79
xiabo123
bdf
Mashiro
liuhw
ckirchhoff
Zaida Zhou
q.yao
Hakjin Lee
enemy1205
Shanghua Gao
Yifei Yang
liukuikun
takuoko
ZShaopeng