Changelog¶
MMSelfSup¶
v1.0.0 (06/04/2023)¶
Highlight¶
Support
PixMIM
.Support
DINO
inprojects/dino/
.
New Features¶
Bug Fixes¶
Improvements¶
v1.0.0rc6 (10/02/2023)¶
The master
branch is still 0.x version and we will checkout a new 1.x
branch to release 1.x version. The two versions will be maintained simultaneously in the future.
We briefly list the major breaking changes here. Please refer to the migration guide for details and migration instructions.
Highlight¶
Support
MaskFeat
with video dataset inprojects/maskfeat_video/
Translate documentation to Chinese.
Bug Fixes¶
Improvements¶
v1.0.0rc5 (30/12/2022)¶
The master
branch is still 0.x version and we will checkout a new 1.x
branch to release 1.x version. The two versions will be maintained simultaneously in the future.
We briefly list the major breaking changes here. Please refer to the migration guide for details and migration instructions.
Highlight¶
Support
BEiT v2
,MixMIM
,EVA
Support
ShapeBias
for model analysisAdd Solution of FGIA ACCV 2022 (1st Place)
Refactor t-SNE
New Features¶
Bug Fixes¶
Improvements¶
v1.0.0rc4 (07/12/2022)¶
The master
branch is still 0.x version and we will checkout a new 1.x
branch to release 1.x version. The two versions will be maintained simultaneously in the future.
We briefly list the major breaking changes here. Please refer to the migration guide for details and migration instructions.
Highlight¶
Support
BEiT
andMILAN
Support low-level reconstruction visualization
New Features¶
Improvements¶
v1.0.0rc3 (01/11/2022)¶
The master
branch is still 0.x version and we will checkout a new 1.x
branch to release 1.x version. The two versions will be maintained simultaneously in the future.
We briefly list the major breaking changes here. Please refer to the migration guide for details and migration instructions.
Highlight¶
Support
MaskFeat
v1.0.0rc2 (12/10/2022)¶
The master
branch is still 0.x version and we will checkout a new 1.x
branch to release 1.x version. The two versions will be maintained simultaneously in the future.
We briefly list the major breaking changes here. Please refer to the migration guide for details and migration instructions.
Highlight¶
Full support of
MAE
,SimMIM
,MoCoV3
.
New Features¶
Improvements¶
v1.0.0rc1 (01/09/2022)¶
We are excited to announce the release of MMSelfSup v1.0.0rc1.
MMSelfSup v1.0.0rc1 is the first version of MMSelfSup 1.x, a part of the OpenMMLab 2.0 projects.
The master
branch is still 0.x version and we will checkout a new 1.x
branch to release 1.x version. The two versions will be maintained simultaneously in the future.
We briefly list the major breaking changes here. Please refer to the migration guide for details and migration instructions.
Highlight¶
New Features¶
Add
SelfSupDataSample
to unify the components’ interface.Add
SelfSupVisualizer
for visualization.Add
SelfSupDataPreprocessor
for data preprocess in model.
Improvements¶
Most algorithms now support non-distributed training.
Change the interface of different data augmentation transforms to
dict
.Run classification downstream task with MMClassification.
Docs¶
Refine all documents and reorganize the directory.
Add concepts for different components.
v0.6.0 (02/02/2022)¶
Highlight¶
Bug Fixes¶
Improvements¶
Cancel previous runs that are not completed in CI (#145)
Enhance MIM function (#152)
Skip CI when some specific files were changed (#154)
Add
drop_last
when building eval optimizer (#158)Deprecate the support for “python setup.py test” (#174)
Speed up training and start time (#181)
Upgrade
isort
to 5.10.1 (#184)
v0.5.0 (16/12/2021)¶
Highlight¶
Released with code refactor.
Add 3 new self-supervised learning algorithms.
Support benchmarks with MMDet and MMSeg.
Add comprehensive documents.
Refactor¶
Merge redundant dataset files.
Adapt to new version of MMCV and remove old version related codes.
Inherit MMCV BaseModule.
Optimize directory.
Rename all config files.
New Features¶
Add SwAV, SimSiam, DenseCL algorithms.
Add t-SNE visualization tools.
Support MMCV version fp16.
Benchmarks¶
More benchmarking results, including classification, detection and segmentation.
Support some new datasets in downstream tasks.
Launch MMDet and MMSeg training with MIM.
Docs¶
Refactor README, getting_started, install, model_zoo files.
Add data_prepare file.
Add comprehensive tutorials.
OpenSelfSup (History)¶
v0.3.0 (14/10/2020)¶
Highlight¶
Support Mixed Precision Training
Improvement of GaussianBlur doubles the training speed
More benchmarking results
Bug Fixes¶
Fix bugs in moco v2, now the results are reproducible.
Fix bugs in byol.
New Features¶
Mixed Precision Training
Improvement of GaussianBlur doubles the training speed of MoCo V2, SimCLR, BYOL
More benchmarking results, including Places, VOC, COCO
v0.2.0 (26/6/2020)¶
Highlights¶
Support BYOL
Support semi-supervised benchmarks
Bug Fixes¶
Fix hash id in publish_model.py
New Features¶
Support BYOL.
Separate train and test scripts in linear/semi evaluation.
Support semi-supevised benchmarks: benchmarks/dist_train_semi.sh.
Move benchmarks related configs into configs/benchmarks/.
Provide benchmarking results and model download links.
Support updating network every several iterations.
Support LARS optimizer with nesterov.
Support excluding specific parameters from LARS adaptation and weight decay required in SimCLR and BYOL.