Bump version to v0.3.0

[zytx121]

v0.3.0 (29/4/2022)

Highlight

• Support TorchServe ([Feature] support torchserve #160)
• Support Rotated ATSS (CVPR'20) ([Algorithm] Support Rotated ATSS-OBB (CVPR'20) #179)

New Features

• Update performance of ReDet on HRSC2016. ([Reimplementation] Update performance of ReDet on HRSC2016 #203)

• Upgrade visualization to custom colors of different classes ([Enhance] Refactor imshow_det_rbboxes #187)

• Update Stable KLD, which solves the Nan issue of KLD training. ([Algorithm] Update Stable KLD (NeurIPS'21) #183)

• Support setting dataloader arguments in config and add functions to handle config compatibility. ([Feature] Add dataloader arguments in config #215)
  The comparison between the old and new usages is as below.

  Before v0.2.0

  Since v0.3.0

  data = dict( samples_per_gpu=2, workers_per_gpu=2, train=dict(type='xxx', ...), val=dict(type='xxx', samples_per_gpu=4, ...), test=dict(type='xxx', ...), )

  # A recommended config that is clear data = dict( train=dict(type='xxx', ...), val=dict(type='xxx', ...), test=dict(type='xxx', ...), # Use different batch size during inference. train_dataloader=dict(samples_per_gpu=2, workers_per_gpu=2), val_dataloader=dict(samples_per_gpu=4, workers_per_gpu=4), test_dataloader=dict(samples_per_gpu=4, workers_per_gpu=4), ) # Old style still works but allows to set more arguments about data loaders data = dict( samples_per_gpu=2, # only works for train_dataloader workers_per_gpu=2, # only works for train_dataloader train=dict(type='xxx', ...), val=dict(type='xxx', ...), test=dict(type='xxx', ...), # Use different batch size during inference. val_dataloader=dict(samples_per_gpu=4, workers_per_gpu=4), test_dataloader=dict(samples_per_gpu=4, workers_per_gpu=4), )

• Add get_flops tool ([Feature] Support get flops tool #176)

Bug Fixes

• Fix bug about rotated anchor inside flags. ([Fix] Modify the use of rotated anchor inside flags #197)
• Fix Nan issue of GWD. ([Fix] Fix nan issue of GWD #206)
• Fix bug in eval_rbbox_map when labels_ignore is None. ([Fix] bug in eval_rbbox_map when labels_ignore is None #209)
• Fix bug of 'RoIAlignRotated' object has no attribute 'output_size' ([Fix] 'RoIAlignRotated' object has no attribute 'output_size' #213)
• Fix bug in unit test for datasets. ([Fix] Fix test code for datasets #222)
• Fix bug in rotated_reppoints_head. ([Fix] Fix bug in rotated_reppoints_head #246)

Improvements

• Update citation of mmrotate in README.md (Update citation of mmrotate in README.md #263)
• Update the introduction of SASM (AAAI'22) ([Docs] Update the introduction of SASM (AAAI'22) #184)
• Fix doc typo in Config File and Model Zoo. ([Docs] Fix typo in Config File and Model Zoo #199)
• Unified RBox definition in doc. ([Fix] Unified RBox definition #234)

Contributors

A total of 7 developers contributed to this release.
Thanks @nijkah @GamblerZSY @liuyanyi @yangxue0827 @jbwang1997 @zytx121 @ZwwWayne

[codecov]

Codecov Report

Merging #260 (427c2b2) into dev (e555fa2) will decrease coverage by 0.17%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##              dev     #260      +/-   ##
==========================================
- Coverage   29.30%   29.13%   -0.18%     
==========================================
  Files         113      113              
  Lines        7298     7298              
  Branches     1110     1110              
==========================================
- Hits         2139     2126      -13     
- Misses       5083     5094      +11     
- Partials       76       78       +2     

Flag	Coverage Δ
unittests	29.13% <100.00%> (-0.16%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
mmrotate/version.py	61.53% <100.00%> (ø)
mmrotate/datasets/builder.py	25.92% <0.00%> (-25.93%)	⬇️
mmrotate/utils/setup_env.py	68.00% <0.00%> (-24.00%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update e555fa2...427c2b2. Read the comment docs.

[yangxue0827]

LGTM