GitLab

* [feature] Use cu92 & ubuntu1604 for torch 1.3.1

* [fix]Use ubuntu-16.04 rather than latest for cu92

* [fix]: soft link cuda path

* Debug cu101 bug

* Update usr/include for cublas with cu101

* Check default nvcc version

* rm :

* add cuda root

* Try to locate cublas_v2.h

* Set cublas version

* Force cublas version

* check usr/local/cuda-10.2

* Add tree

* cp cublas files

* cp cublas files to lib64

* do not tree cuda

* fix path cp error

* cp dir

* Keep using CUDA10.1 only

* recover empty line

* update impad

* fix docstring

* add shape for impad

* fix unit test

* remove old version & fix doc

* fix linting

* fix doc

* add linear decay learning rate scheduler

* fix impad

* fix setup.cfg

* fix linting

* add yapf

* add swish

* fix lr_updater

* fix lr_updater.py

* update swish

* add swish

* fix inplace

* fix typo
Co-authored-by: lixiaojie <lixiaojie@sensetime.com>

* move unit tests to specific folders

* fix path error

* remove some assertions

* fix ignore path

* bump version to 1.1.1

* add more version utils

* move parse_version_info to version.py

* Custom imports

* Resolve comments

* Add unittest

* Add unittest

* Rename custom_imports to import_modules_from_strings

* Move import_modules_from_strings ito misc.py and allow failed imports

* small change

* small change

* change mmcv.runner to os.path

* [feature] Add collect_env() to collect environment info

* fix unit tests

* fix the case when ops are not compiled

* fix docstring

* Fix function types

* Remove const

* Support to split batched_nms when box number is too large

* mv data from gpu to cpu

* Set split_thr through nms_cfg

* clean code

* Update motivation in docstring

* fix typos

* support ONNX adaptive average pooling

* fix double quotes
Co-authored-by: Kai Chen <chenkaidev@gmail.com>

* fix mdconv addmm bug for parrots

* fix mdconv ctv save tensor
Co-authored-by: hanyachao <hanyachao@sensetime.com>

* Fix mmcv-dataparallel

* Fix (parallel): fix CPU inference with MMDataParallel

* Update docstrings

* Doc (parallel): refine docstrings

* Fix (parallel): fix missing changes of train/val step function

* resolve comments

* Fix (data_parallel): fix bug when single gpu test return None

* log mode by using runner.mode

* add IterTimerHook to fix unittest

* correct the logic

* display fix when using EvalHook in train mode

* simplify logic

* fix saconv

* add parrots condition

* add unittest

* fix torch version

* fix some pavi logger hooks

* fix unittest

* fix small bugs

* small change

* fix unittest

* Add EpochBasedRunner conditions

* Add session text

* fix small bug

* fetch runner mode from log buffer

* Add max_iter to pavi session text

* change yaml.dump to yamp.dump(yaml.load(mmcv.dump))

* Directly use by_epoch

* fix unittest

* add comments

* Use runner.epoch + 1 in pavi log

* fix runner.epoch issue for runner.mode=='val'

* fix runner.epoch issue for runner.mode=='val'

* Use abspath instead of realpath

* Add meta dump unittest

* small change

* Add comments

* add tin shift

* add unittest

* add docstring

* add docstring

* parrots for tin_shift

* fix lint

* fix lint
Co-authored-by: jiaomenglei <jiaomenglei@sensetime.com>

* Add json set_default for unserializable values

* kwargs.setdefault

* add op trouble shooting

* update trouble_shooting.md

* clean ops.md

* add trouble shooting to index.rst

* reorder

* add troubleshooting in readme

* fix the bug (#472)

* fix the bug (#472)

* fix the bug (#472)
Co-authored-by: hezijian <hezijian@dm-ai.cn>

* add pytorch 1.6 to CI

* fix typo

* fix ci error
Co-authored-by: Cao Yuhang <yhcao6@gmail.com>

* bump version to 1.0.5

* fix pytorch name

* fix version

* disable the cache

* Add git hash

* Add fallback

* Add func in mmcv.utils