GitLab

* 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 max_keep_ckpts parameters to save memory when saving models

* format

* format

* format

* fixed linting error
Co-authored-by: z-bingo <z-bingo@outlook.com>

Co-authored-by: xurui <xurui4@sensetime.com>

* 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

* add contains to registry
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* add contains to registry

* 'fix'

* 'minor'

* move
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>
Co-authored-by: lixuanyi <lixuanyi@sensetime.com>

* add pretty_text

* add test cases

* add argument indent

* change pretty_text to property

* fixed format end

Signed-off-by: lizz <lizz@sensetime.com>

* fix a potential bug of removing temp dir

* bump version to v0.4.2

* recursive scanndir

* support python3.5

* update and add tests

* resolve comments

* remove is_str

* add bia_init_with_prob

* add test_weight_init

* 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 iminvert

* DC now supports dimension other than 3

* add lmdb

* revise for lmdb

* add warmup_byepoch option

* remove not needed files

* add attribute active_loader for runner, revise code

* remove not used attrs

* remove not used attrs

* add docstring, else

* to pass CI

* to pass CI
Co-authored-by: zhaoyue-zephyrus <thuzhaoyue@gmail.com>

* add base for config

* fixed format

* rm terminal width

* support multiple & recursive base

* add test case

* fix format

* add test construct

* minor fix

* add more test, rewrite merge from opt

* avoid depulicate keys

* delete imported config as module

* rename merge_from_dict

* support pathlib in imageio

* support pathlib in imageio

* add Path type for the annotation.
Co-authored-by: xurui4 <xurui4@sensetime.com>

Signed-off-by: lizz <lizz@sensetime.com>

* Update tensorboard.py

* Update tensorboard.py

* add is_scalar in PaviLoggerHook and add unittest

* add docstrings

* update pre-commit hook config

* update the contributing guide

* fix wandb hook

* add pavi logger hook

* modify default value of init_kwargs to None

* add more features to PaviLoggerHook

* rm taskid (will cause bug)

* directly upload saved pth

* fix CI

* register PAVI Hooks

* add logging utils

* install torch (cpu only) for CI

* fix unittests for runner

* remove python 2.7 from testing list

* add a corner case

* add registry and build_from_cfg

* add some corner cases

* add some corner cases

* fix the unittest for python 3.5

* minor fix

* size hint
Signed-off-by: lizz <lizz@sensetime.com>

* add fast resize
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* fix
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* add tests
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* fix
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* Update tests/test_image.py

* change format
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* add tests
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* add docstring
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>
Co-authored-by: lizz <innerlee@users.noreply.github.com>

* add inplace normalization
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* fix
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* add dtype check

* modify inplace norm
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* fix
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* fix
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* fix line

* add implace flip

* add tests
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* modify tests
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* add inplace tests
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* add comments
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* fix
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* change image decoding method from cv2 to TurboJPEG
Signed-off-by: lixuanyi <lixuanyi@sensetime.com>

* add tests

* travis
Signed-off-by: lizz <lizz@sensetime.com>

* fix
Signed-off-by: lizz <lizz@sensetime.com>

* more doc
Signed-off-by: lizz <lizz@sensetime.com>

* change per request
Signed-off-by: lizz <lizz@sensetime.com>

* more doc
Signed-off-by: lizz <lizz@sensetime.com>

* baby-sitting
Signed-off-by: lizz <lizz@sensetime.com>
Co-authored-by: lizz <innerlee@users.noreply.github.com>

* Use cv2 to normalize img

* multipy faster than divide

* also do imdenormalize

* support more channels

* even faster
Signed-off-by: lizz <lizz@sensetime.com>

* add test
Signed-off-by: lizz <lizz@sensetime.com>

* even even faster
Signed-off-by: lizz <lizz@sensetime.com>

* fffaster
Signed-off-by: lizz <lizz@sensetime.com>

* okay
Signed-off-by: lizz <lizz@sensetime.com>

* guard against implace modification
Signed-off-by: lizz <lizz@sensetime.com>

* fix imdenormalize
Signed-off-by: lizz <lizz@sensetime.com>

* tidy
Signed-off-by: lizz <lizz@sensetime.com>

* clean
Signed-off-by: lizz <lizz@sensetime.com>

* Update test_image.py

* Update test_image.py

* Update test_image.py

* Update test_image.py

* Update test_image.py

* Update test_image.py

* fix an import error

* fix an import error

* log meta

* add type check

* Fix the missing numpy module for readthedocs

* fix the linting

* Speed up bbox_clip

* Update geometry.py