GitLab

* add windows CI

* clean versions

* only allow pt1.7 on windows

* fix windows install issue

* add win cpu

* fix win command

* clean unnecessary command

* resolve turbojpeg & tempfile on win

* replace os.readlink with os.path.realpath

* fix windows ci

* close file before removing it

* fix windows ci

* fix symlink on windows

* fix windows ci

* fix windows ci

* fix windows ci

* fix windows ci

* fix windows ci

* fix windows ci

* fix windows ci

* fix windows ci

* fix windows ci

* modify according to comment
Co-authored-by: zhouzaida <zhouzaida@163.com>

* Add gradient cumulative optimizer

fixes #190

* Update optimizer.py

* Update optimizer.py

* fix loss scale improperly in last equivalent_iter

* Add `GradientCumulativeOptimizerHook` in `__init__.py`.

* Add docstring of `GradientCumulativeOptimizerHook`.

* Add type check, BN warning and resume warning. And fix typo, lint the
code.

* Add unit test

* Update docstring example.

* Change GradientCumulativeOptimizerHook `__init__` arguments.

* Add GradientCumulativeOptimzierHook unit tests with IterBasedRunner.

* Add GradientCumulativeFp16OptimizerHook.

* Add unit tests of GradientCumulativeFp16OptimizerHook

* Use '!=' instead of '>' to determine resume
Co-authored-by: Zhiyuan Chen <this@zyc.ai>

* add flat cosine lr updater

* add test

* add doc

* update doc

* reformat

* update unittest

* update test flat cos

* remove momentum hook test

* update test

* change assert to ValueError

* fix unittest

* add by_epoch=True unittest

* change to start_percent

* change to start_percent in test

* support print using hooks before running.

* Support to print hook trigger stages.

* Print stage-wise hook infos. And make `stages` as class attribute of
`Hook`

* Add util function `is_method_overriden` and use it in
`Hook.get_trigger_stages`.

* Add unit tests.

* Move `is_method_overriden` to `mmcv/utils/misc.py`

* Improve hook info text.

* Add base_class argument type assertion, and fix some typos.

* Remove `get_trigger_stages` to `get_triggered_stages`

* Use f-string.

* Refine default hooks and custom hooks priority rank.

* Add unit tests for custom hooks with string priority.

* Use priority `ABOVE_NORMAL` and `BELOW_NORMAL` instead of `HIGHER` and
`LOWER`.

And add unit tests for custom hook with the same priority as
default hooks.

* Add dvclive logger hook

* Move docstring to class

* docstring updates

* add neptune.ai logger hook

* add docstring

* move docstring

* update docstr

* update before_run

* Assign different priority to default hooks, and add custom hook register in base runner.

* Add custom hook register in example train file

* Add unittest of custom hook

* Code format

* add StepMomentumUpdaterHook

* add unit test

* fix typos

* refactor step updater

* replace stage with exp

* fix linting error

* use all() operation

* [Fix] lr add triangular

* [Fix] lr_triangular

* lr add docstring

* lr add docstring

* add unittest of triangular lr update

* support clipping min_lr in StepLrUpdaterHook

* add docstring for StepLrUpdaterHook

* fix small bugs

* add unit test for StepLrUpdaterHook

* fix linting error

* [Fix] OneCycleLrUpdaterHook interface

* revise according to comments

* revise according to comments

* add test

* fix lint

* revise according to comments

* minors

* add pytest param

* fix lint

* ci

* [Fix] fix bug of lr update hook

* [Fix] fix bug of lr update hook

* [Fix] Fix bug of lr updater hook

* [Feature] Add OneCycleLrUpdaterHook

* fix docstring

* fix docstring

* Remove redundant code

* Simplified the code.

* Improved chkpt compatibility.

* One may modify the checkpoint via adding keywords.

* Tiny.

* Following reviewer's suggestion.

* Added unit_test.

* Fixed.

* Modify the state_dict  with  construction.

* Added test.

* Modified。

* Mimimalised the modification.

* Added the docstring.

* Format.

* Improved.

* Tiny.

* Temp file.

* Added assertion.

* Doc string.

* Fixed.

* [Enhance] Refactor logger

* fixed test

* make commit optional

* remove debug info

* fixed test

* [Feature] Add hook msg for ckpt_hook

* [Feature] Add hook msg for ckpt_hook

* add test

* Update lr_updater.py

since epoch/iteration in  runner starts with 0, we shouldn't leave the latter iteration to former (12th epoch for example, with first period equal to 12) period.

* Update lr_updater.py

* Update test_hooks.py

* Refactor tags for consistency

* Fix missing runner

* Fix missing runner

* Fix missing runner

* Fix missing runner

* Fix momentum runner hook inner iter

* Fix tests

* pre-commit run

* Add build_runner

* Parametrize test_runner

* Add imports to runner __init__

* Refactor max_iters and max_epochs from run to init

* Add assertion error messages

* Add test_builder

* Make change retro-compatible

* Raise ValueError if max_epochs and max_iters

* 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 ema hook

* add ema hook resume

* add ema hook test

* fix typo

* fix according to comment

* delete logger

* fix according to comment

* fix unitest

* fix typo

* fix according to comment

* change to resume_from

* typo

* fix isort

* More robust sync buffer hook

* More robust sync buffer hook

* Reformat

* reformat

* reformat

* Add register hook from cfg

* docstring

* change according to comments

* update doc formats

* update docstring

* feat: add CosineRestartLrUpdaterHook

* style: rename period to periods

* fix: bug in period 0

* feat: rename eta_min to min_lr and add min_lr_ratio

* docs: fix docstring of restart lr updater

* refactor: use annealing_cos

* docs: add docstring to annealing_cos

* feat: cosine restart lr update hook

* refactor: modify code order for unittest

* add a BaseRunner and rename Runner to EpochBasedRunner

* fix the train/val step

* bug fix

* update unit tests

* fix unit tests

* raise an error if both batch_processor and train_step are set

* add a unit test

* refactor unittests

* split test_video.py to two files

* use tmp dir as work_dir of runner

* only run codecov for python 3.7

* remove useless comments

* track progress of iter&enum

* restore

* add momentum scheduler

* fix small bug

* cyclic scheduler"

* fix bug

* fix second phase's bug

* reformat

* feature (cosine lr): use relative ratio for more flexible scheduler

* Fix (runner): fix bugs in runner

* Refactor (hook): refactor cosing/cyclic LR/momentum hook with unittest

* Clean unnecessary files and reformat

* Fix memory key error when GPU is not avaliable

* Resolve comments

* Do not print momentum in text log

* Change hook register order

* Refactor max_iter

* Fix max_iter bugs in runner

* Enforce target_ratio to be either tuple or float

* Add MLflowLoggerHook

* Add MLflowLoggerHook to __all__

* Update name

* Fix tracking.MlflowClient setup

* Fix log_metric

* Fix mlflow_pytorch import

* Handle active_run

* Fix self.mlflow reference

* Simplify using high level API

* Fix set_experiment

* Add only_if_torch_available decorator and test_mlflow_hook

* Add missing import in hooks

* Fix torch available check

* Patch mlflow.pytorch in test

* Parametrize log_model

* Fix log_model parametrize

* Add docstring

* Move wand patch

* Fix flake8

* Add regression test for non numeric metric

* Only log numbers

* Rename experiment_name-> exp_name

* Remove pytest skip

* remove supports for python 2.7

* fix the unit test for python 3.5

* add python 3.8 in CI

* try ubuntu 18.04 as the environment

* add is_scalar in PaviLoggerHook and add unittest

* add docstrings