mirrors
/
mmcv
mirror of https://github.com/open-mmlab/mmcv.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
184 Commits
13 Branches
108 Tags
42 MiB
Commit Graph

184 Commits (5b72bd4bafd747994558a3ec3d9a94fd1f1ef011)
 

Author SHA1 Message Date
Kai Chen 5b72bd4baf
fix the compiling on mac os (#97) 2019-07-31 02:22:20 +08:00
Karlind 30985b2f76 add support for pathlib to load and dump (#91) 
* add support for pathlib to load and dump

* minor formatting
 2019-07-31 01:21:05 +08:00
HarryHan 5a5c94b68b Update lr_updater.py (#95) 
* Update lr_updater.py

"Add min_lr to PolyLrUpdaterHook, so that it can anneal to min_lr. When min_lr=0, it behaves as usual. "

* Update lr_updater.py
 2019-07-30 23:23:54 +08:00
Kai Chen eae81c1e86
Add yapf and isort to travis (#96) 
* add yapf and isort to travis

* minor formatting

* remove blank lines

* skip unit tests for progressbar when python2

* update travis to ubuntu 16.04

* use a newer version ffmpeg

* add -y to add-apt-repository
 2019-07-30 23:15:56 +08:00
Kai Chen 3e1e297d36
use torchvision:// instead of modelzoo:// (#89) 2019-07-13 13:43:10 +08:00
Kai Chen e5ca8846c5
Fix the compiling bug when using `pip install mmcv` (#88) 
* rename flowWarp to FlowWarp

* fix the compiling bug when using pip install

* update version to v0.2.10
 2019-07-07 21:41:31 +08:00
Kai Chen 0ad15a508e
Add mmcv._ext to autodoc_mock_imports (#86) 2019-07-07 17:32:40 +08:00
Kai Chen e60c7efde3
Improve the documentation (#85) 
* fix doc generation

* add docs for optical flow

* fix typo
 2019-07-07 16:54:22 +08:00
simon wu 492228794b Fixing build-errors on Windows (#84) 2019-07-07 13:55:51 +08:00
Jiangmiao Pang 258d4d320c fix one bug in vgg (#83) 2019-06-24 20:13:33 +08:00
Kai Chen ff5631003f
add mmaction to readme (#82) 2019-06-24 20:11:30 +08:00
Yue Zhao f9a750fe64 add BN pretrain model, add i3d pretrained on K-400 (#81) 
* add BN pretrain model, add i3d pretrained on K-400

* fix flake8
 2019-06-18 01:40:11 +08:00
lizz 28feffdb22 Ignore autogenerated flow_warp_module.cpp (#80) 
Signed-off-by: lizz <363664470@qq.com>
 2019-06-15 22:13:27 +08:00
Evgeny Nizhibitsky 395efbf2e9 Make latest checkpoint symlink relative (#75) 
* Make latest symlink relative. Add tests

* Skip test_save_checkpoint in the absense of torch

* Fix pep8 in test_runner

* Import runner only if torch import has succeeded
 2019-06-15 19:56:27 +08:00
lizz 123720b64e Download one copy in distributed environment (#76) 
* Download one copy in distributed environment

Signed-off-by: innerlee <363664470@qq.com>

* fix

Signed-off-by: innerlee <363664470@qq.com>

* format

Signed-off-by: innerlee <363664470@qq.com>
 2019-06-15 14:52:53 +08:00
Yue Zhao 92a81b62a6 Polish DataContainer (#79) 
* add iminvert

* DC now supports dimension other than 3

* fit flake8

* generalize pad_dim and beautify code

* fix bug; rename "pad_dim" to "pad_dims" for disambiguity

* fix bug

* fix flake8

* fix yapf format

* use deafault yapf

* beautify style
 2019-06-15 01:00:28 +08:00
nbei 9097957434 C++ implementation for Flow Warp Module (#71) 
* flow_warp_c

* flow_warp_c

* beautify format

* beautify format

* beautify format

* beautify format

* add Cython

* modify

* fix details

* fix details

* fix type
 2019-05-28 20:03:08 +08:00
Kai Chen 20d7cd30dc
Merge pull request #70 from hellock/master 
Update to v0.2.8
 2019-05-22 13:41:42 +08:00
Kai Chen 2051f4f4e6 update to v0.2.8 2019-05-22 13:35:52 +08:00
Kai Chen c234cdea80
add model urls for HRNet (#68) 2019-05-22 11:21:33 +08:00
Cao Yuhang 8e84059cee only display base lr when there are multi lrs (#65) 
* only display base lr when there are multi lrs

* change comment: base lr to lr of the first param group
 2019-05-13 20:56:42 -07:00
Cao Yuhang 573aaa785f fix resnet-x101-32x4d-gn+ws hash error (#67) 
* fix resnet-x101-32x4d-gn+ws hash error

* fix hash error of other models in jhu
 2019-05-12 17:46:31 -07:00
Kai Chen c728193f8d bug fix for the latest torchvision 2019-05-09 00:16:04 -07:00
Kai Chen e2e109e4a5
Merge pull request #66 from wangg12/master 
add bgr2hls and hls2bgr
 2019-05-06 21:28:15 -07:00
Gu Wang d4b91b8376 test bgr2hls 2019-05-07 12:04:01 +08:00
Gu Wang a092de341c add bgr2hls and hls2bgr 2019-04-30 22:52:14 +08:00
Kai Chen 0d77f61450
Merge pull request #64 from ycxioooong/master 
video fps should be float
 2019-04-23 21:06:24 -07:00
xiongyu d8bdc8679c video fps should be float 2019-04-24 11:58:50 +08:00
Kai Chen 5d88696a8a
Merge pull request #63 from hellock/master 
Add jhu gn model urls and update to v0.2.7
 2019-04-23 18:07:54 -07:00
Kai Chen bb089030a9 remove wrong license argument 2019-04-23 12:52:02 -07:00
Kai Chen cc207c4523 add jhu gn model urls and update to v0.2.7 2019-04-23 00:15:27 -07:00
Kai Chen 4abb916d60
Merge pull request #60 from wangg12/master 
fix possible encoding error when install
 2019-04-21 19:49:46 -07:00
Gu Wang 11ad40b560 fix possible encoding error when install 2019-04-22 10:43:33 +08:00
Kai Chen d99c6f8ddf
Merge pull request #62 from hellock/collections-abc 
Use collections.abc instead of collections
 2019-04-21 19:33:54 -07:00
Kai Chen 528bf72ce8
Merge pull request #61 from hellock/weight-init 
Add caffe2_xavier_init()
 2019-04-21 19:16:24 -07:00
Kai Chen b7c924cb96 use collections.abc instead of collections 2019-04-21 19:13:28 -07:00
Kai Chen 5e3d09bc5d add caffe2_xavier_init() because of different implementations of caffe2 and pytorch 2019-04-21 18:51:38 -07:00
Yue Zhao 944da49fcb add iminvert (#58) 
* add iminvert

* fix for flake8

* minor improvement

* add test for iminvert

* fix for flake8

* fix for flake8

* revise __init__

* revise __init__
 2019-04-20 00:27:59 -07:00
Kai Chen 38fa314226
Add GN+WS pretrained model urls (#59) 
* add GN+WS pretrained model urls

* fix linting errors
 2019-04-19 21:27:34 -07:00
Cao Yuhang 8266d877fe add json log support (#55) 
* add json log support

* add some comment, minor fix

* rename variable vv to item

* mv round float to a function
 2019-04-19 00:18:33 -07:00
Kai Chen 506455af0b
Update version to 0.2.6 (#54) 
* fix the time estimation when resuming from a checkpoint

* fix the time estimation when resuming from a checkpoint

* support str type for loggers

* fix yaml dump tests

* update version to v0.2.6
 2019-04-14 22:55:58 -07:00
Cao Yuhang b85136772d fix bug that stuck at evaluation (#53) 
* fix bug that stuck at evaluation

* remove mode variable
 2019-04-14 22:28:05 -07:00
Kai Chen bef7c13a95
Merge pull request #47 from yhcao6/memory_statistic 
reduce max memory in dist training
 2019-04-10 22:38:45 -07:00
Cao Yuhang 5726795c7a reduce max memory in dist training 2019-04-08 14:07:43 +08:00
Kai Chen 4a1ddfc8a1
Merge pull request #44 from yhcao6/master 
fix memory order in log text
 2019-04-06 19:32:53 -07:00
Cao Yuhang f3d5d97919 fix memory order in log text 2019-04-07 10:20:52 +08:00
Cao Yuhang 1263c91e5c add memory statistic (#41) 
* add memory statistic

* fix import order and cuda check
 2019-04-05 23:11:35 -07:00
Kai Chen a0ce998ecd update to v0.2.5 2019-03-25 22:54:42 -07:00
Kai Chen d426585999
Support str type for loggers (#39) 
* fix the time estimation when resuming from a checkpoint

* fix the time estimation when resuming from a checkpoint

* support str type for loggers

* fix yaml dump tests
 2019-03-25 21:59:02 -07:00
Aitical b0f40396ad Update the documentation (#36) 
* update the documentation

* update the documentation

* Update io.md
 2019-03-09 16:16:08 -08:00