GitLab

* add ops (furthest point sample) in mmdet3d

* refactor code

* refactor code

* fix typo

* fix typo

* fix typo

* refactor code

* fix typo

* define DIVUP in common_cuda_helper.hpp

* add onnx dcn

* replace gemm with torch C++ api

* small update

* fix cpp clang format

* pefer generic GEMM than torch ATen library

* addressing comments

* add ops path check

* Update deform_conv.py

* polish the docstring of reset_parameters

* change the initialization of `self.weight`

switch the initialization of `self.weight` to the standard kaiming method
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* Add code-spell hook to pre-commit

* Add code-spell hook to pre-commit and fix typos

* fix setup.cfg

* [Fix] Revise unit test of correlation

* rename

* lint

* lint

* lint

* lint

* [Feature]: support empty tensor in MMSyncBN

* refine code

* resolve comments

* clean unnecessary comments

* fix inaccurate statistics when empty tensor

* resolve comments and add docstrings

* update unit tests

* rephrase, ready for merge

* Reimplement cc_attention using pure pytorch

* fix: avoid BC-Breaking

* delete cc_attention related cpp and cuda files

* delete cc_attention related lines in pybind.cpp

* make out Tensor contiguous.

* remove unneeded lines.

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

* Update TestCrissCrossAttention

* passing pre-commit

* Update docstring of CrissCrossAttention

* Update docstring of CrissCrossAttention

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

* [docs]polish the docstring

* [Docs] Polish the docstring
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>
Co-authored-by: Zaida Zhou <58739961+zhouzaida@users.noreply.github.com>

* DCN cpu version

* add modulated dcn cpu version

* move deform_conv_shape_check to deform conv utils

* add inline to deform_conv_shape_check

* add tests

* run linter

* add newline at file end

* run pre-commit against modulated deform conv cpp

* update saconv test

* run clang-format

* remove cuda device inline

* refactor dcn cuda/cpu functions

* remove DCN util

* remove DCN util hpp from all included files

* Addressing PR comment by refactoring modulated-DCN

* fix lint in cpp files

* add modulated_deform_conv in onnxruntime support

* Add docs descriptions

* Add gpu test in test_onnx.py

* code format

* remove new usage and move if outside for loop

* use memset when bias is nullptr

* [Fix] Fix pdf error when building docs

* fix docs

* fix pdf

* fix pdf

* fix pdf

* [Refactor] Refactor the csrc directory

* update MANIFEST.in

* fix hip

* add csrc readme

* trailing whitespace

* fix syntax error in setup.py

* add compatibility docs

* move parrots_cudawarpfunction.cuh to common/cuda

* fix grammar, update directory tree

* fix MANIFEST.in

* Add new structre of csrc in compatibility.md

* Add original structre of csrc in compatibility.md

* fix typo

* remove TODO

* modify according to comment

* format
Co-authored-by: grimoire <yaoqian@sensetime.com>

* [Docs] Add header for files

* change to OpenMMLab

* add headers for .cpp, .cu, .h, .cuh

* replace Open-MMLab with OpenMMLab

* improve digit_version & use it for version_checking

* more testing for digit_version

* setuptools >= 50 is needed

* fix CI

* add debuging log

* >= to ==

* fix lint

* remove

* add failure case

* replace

* fix

* consider TORCH_VERSION == 'parrots'

* add unittest

* digit_version do not deal with the case if 'parrots' in version name.

* windows build doc

* fix floor ceil

* fix lint

* change floor/ceil to floorf/ceilf

* Update build.md

* recover onnx and parrots

* fix clang-lint
Co-authored-by: WRH <12756472+wangruohui@users.noreply.github.com>

* revise the interface of upfirdn2d function

* adopt to_2tuple

* porting mmcv for hip

* add nvcc

* fix format

* fix format

* fix bug for carafe

* fix test_utils because rocm_torch not allow set torch.backends.cudnn.benchmark to false

* add LOOSEVERSION

* fix format

* fix format of version

* fix code format

* test for yaml

* fix bug for citest

* fix bug for how to get torch._version_ at setup.py

* fix saconv

* update

* update

* fix

* use LooseVersion

* optimize criss cross attention

* optimize criss cross attention

* optimize criss cross attention

* fix lint

* fix ci, remove useless variable

* better ca_forward_kernel
Co-authored-by: wondervictor <victorchanchina@gmail.com>

* add modulated dcn, better dcn plugin

* clangformat

* update documentation

* minor changes

* change to modulist

* change to Sequential

* replace dropout with attn_drop and proj_drop in MultiheadAttention

* add operation_name for attn

* add drop path and move all ffn args to ffncfgs

* fix typo

* fix a bug when use default value of ffn_cfgs

* fix ffns

* add deprecate warning

* fix deprecate warning

* change to pop kwargs

* support register FFN of transformer

* support batch first

* fix batch first wapper

* fix forward wapper

* fix typo

* fix lint

* add unitest for transformer

* fix unitest

* fix equal

* use allclose

* fix comments

* fix comments

* change configdict to dict

* move drop to a file

* add comments for drop path

* add noqa 501

* move bnc wapper to MultiheadAttention

* move bnc wapper to MultiheadAttention

* use dep warning

* resolve comments

* add unitest:

* rename residual to identity

* revert runner

* msda residual to identity

* rename inp_identity to identity

* fix name

* fix transformer

* remove key in msda

* remove assert for key
Co-authored-by: HIT-cwh <2892770585@qq.com>
Co-authored-by: bkhuang <congee524@gmail.com>
Co-authored-by: Wenwei Zhang <40779233+ZwwWayne@users.noreply.github.com>

* Fix export to onnx for PointRend

* Fix codestyle

* Fix codestyle

* Fix type in docstring

* Minor fix

* Fix export with custom ops

* Fix codestyle

* Add tests for bilinear_grid_sample function

* Remove redundant operation and rename variables

* Fix bug in bilinear_grid_sample and update test

* Fix getting batch size

* skip torch==1.3.1

* remove unused import

* fix lint

* support export with batch

* fix dynamic clip

* skip test for torch<1.5.0

* Add docstrings and comments

* Minor fix

* Recover clipping code

* Fix clamping in pytorch 1.7.0

* Fix bilinear_grid_sampler

* Minor fix
Co-authored-by: maningsheng <maningsheng@sensetime.com>