mmsegmentation
mmsegmentation copied to clipboard
[Feature] Support COCO-LVIS dataset
Motivation
Support COCO-LVIS dataset
Paper: Scaling Semantic Segmentation Beyond 1K Classes on a Single GPU Repo: ESSNet
Modification
- add convert tool to generate gt_mask
- add COCOLVISDataset class
- modify LoadAnnotations pipeline to support loading 16-bit label
BC-breaking (Optional)
Just add a optional parameter to LoadAnnotations
method.
Checklist
TODO
- Pre-commit or other linting tools are used to fix the potential lint issues.
- The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness.
- If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D.
- The documentation has been modified accordingly, like docstring or example tutorials.
Codecov Report
Merging #977 (c94069d) into master (54bd4bd) will decrease coverage by
0.24%
. The diff coverage is47.36%
.
@@ Coverage Diff @@
## master #977 +/- ##
==========================================
- Coverage 89.85% 89.61% -0.25%
==========================================
Files 118 120 +2
Lines 6558 6596 +38
Branches 1019 1023 +4
==========================================
+ Hits 5893 5911 +18
- Misses 464 484 +20
Partials 201 201
Flag | Coverage Δ | |
---|---|---|
unittests | 89.61% <47.36%> (-0.25%) |
:arrow_down: |
Flags with carried forward coverage won't be shown. Click here to find out more.
Impacted Files | Coverage Δ | |
---|---|---|
mmseg/datasets/coco_lvis.py | 41.17% <41.17%> (ø) |
|
mmseg/datasets/__init__.py | 100.00% <100.00%> (ø) |
|
mmseg/datasets/misc.py | 100.00% <100.00%> (ø) |
|
mmseg/datasets/pipelines/loading.py | 98.55% <100.00%> (+0.02%) |
:arrow_up: |
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 54bd4bd...c94069d. Read the comment docs.
Hi @xiexinch The CI has been fixed, please merge the master branch.
Please merge the master branch into your branch, thank you.
Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
谢昕辰 seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
You have signed the CLA already but the status is still pending? Let us recheck it.