[Feature] Support STDC Network (old, deleted) #865

xiexinch · 2021-09-09T09:18:39Z

Motivation

Support a new real-time algorithm, Rethinking BiSeNet For Real-time Semantic Segmentation (CVPR 2021).

Modification

Add a new backbone modified from https://github.com/MichaelFan01/STDC-Seg
Add config files.

Update(2021-10-25):

Aligning inference metric.
Add Unit Test.
Aligning training metric.

codecov · 2021-09-09T09:45:49Z

Codecov Report

Merging #865 (d548fb2) into master (2825efe) will decrease coverage by 1.47%.
The diff coverage is 56.08%.

❗ Current head d548fb2 differs from pull request most recent head b762c9b. Consider uploading reports for the commit b762c9b to get more accurate results

@@            Coverage Diff             @@
##           master     #865      +/-   ##
==========================================
- Coverage   88.90%   87.42%   -1.48%     
==========================================
  Files         110      112       +2     
  Lines        5928     6187     +259     
  Branches      950      989      +39     
==========================================
+ Hits         5270     5409     +139     
- Misses        465      582     +117     
- Partials      193      196       +3

Flag	Coverage Δ
unittests	`87.42% <56.08%> (-1.46%)`	⬇️

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

Impacted Files	Coverage Δ
mmseg/core/evaluation/metrics.py	`90.42% <ø> (-0.20%)`	⬇️
mmseg/datasets/pipelines/formating.py	`63.82% <ø> (ø)`
mmseg/models/backbones/cgnet.py	`94.63% <ø> (ø)`
mmseg/models/backbones/fast_scnn.py	`97.08% <ø> (ø)`
mmseg/models/backbones/hrnet.py	`84.37% <ø> (ø)`
mmseg/models/backbones/mit.py	`91.53% <ø> (ø)`
mmseg/models/backbones/mobilenet_v2.py	`71.08% <ø> (ø)`
mmseg/models/backbones/resnest.py	`83.72% <ø> (ø)`
mmseg/models/backbones/resnet.py	`99.28% <ø> (ø)`
mmseg/models/backbones/resnext.py	`100.00% <ø> (ø)`
... and 19 more

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 2825efe...b762c9b. Read the comment docs.

…n-mmlab#866) * [Fix] Fix iter bug when resuming checkpoint in distributed train * fix lint error Signed-off-by: FreyWang <wangwxyz@qq.com>

* update dockerfile * Update docker/Dockerfile Co-authored-by: Jerry Jiarui XU <xvjiarui0826@gmail.com> * add ARG * change dash to bash Co-authored-by: Jerry Jiarui XU <xvjiarui0826@gmail.com>

mmseg/models/backbones/stdc.py

configs/stdc/fcn_stdc-813_512x1024_40k_cityscapes.py

mmseg/models/backbones/stdc.py

…nd (open-mmlab#867) * support load gt for evaluation from multi-backend * move some code from get_gt_seg_maps to get_one_gt_seg_map * rename gt_seg_map_loader_conf to gt_seg_map_loader_cfg * fix doc str * rename get_one_gt_seg_map to get_gt_seg_map_by_idx

* [Fix] Convert SyncBN to BN when training on DP. * Modify SyncBN2BN. * Add SyncBN2BN unit test. * Resolve some comments. * use mmcv official revert_sync_batchnorm * Remove local syncbn2bn unit tests. * Update mmcv version. * Fix bugs of gather model tools. * Modify warnings. * Modify docker mmcv version. * Update mmcv version table.

* Improve docs style * update lists * update the size of image * modify duplicate mmdet3d

* Use Chinese menu * replace \ with /

* support coco stuff-10k/164k * update docs * fix docs * update docs * fix import lints * Update docs/dataset_prepare.md * Update docs/dataset_prepare.md * Update tools/convert_datasets/coco_stuff164k.py * Update tools/convert_datasets/coco_stuff10k.py * Update tools/convert_datasets/coco_stuff10k.py * Update tools/convert_datasets/coco_stuff10k.py * Update tools/convert_datasets/coco_stuff10k.py * Update coco_stuff.py fix the description of the dataset * Update dataset_prepare.md fix the doc tree of coco stuff 10k * Update coco_stuff10k.py fix img_dir * Update coco_stuff.py fix descriptions * Update coco_stuff164k.py fix out_dir * Update coco_stuff10k.py fix save file name * Update coco_stuff.py fix seg_map_suffix * Update dataset_prepare.md fix -p * Update dataset_prepare.md fix doc tree * modify coco stuff convertor * Remove redundant code * fix 164k convert bug * remove redundant comment * add deeplabv3 configs and more iterations * replace shutil.move with shtil.copyfile * Update deeplabv3_r50-d8_512x512_4x4_80k_coco_stuff10k.py fix wrong config * Update deeplabv3_r101-d8_512x512_4x4_80k_coco_stuff164k.py fix wrong config * fix wrong configs * fix wrong configs * fix wrong path for coco stuff 10k * fix convert bugs * fix seg_filename bug * when nproc=0, use track progress * rename configs: coco_stuff --> coco-stuff * add coco-stuff 10k and 164k to README.md * update configs * add deeplabv3 benchmark * add pspnet benchmark * remove redundant comma Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn>

* test_torchserver1.1 * test_torchserver1.2 * update * update mmseg_handler.py * update docs * update torchserver * tranfer torchserver to torchserve * update docs

* multiple losses * fix lint error * fix typos * fix typos * Adding Attribute * Fixing loss_ prefix * Fixing loss_ prefix * Fixing loss_ prefix * Add Same * loss_name must has 'loss_' prefix * Fix unittest * Fix unittest * Fix unittest * Update mmseg/models/decode_heads/decode_head.py Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn>

* Update config.md * Update config.md

* [Fix] Fix loss parse in val_step * Add val_step unittest * Add train_step unittest

* BiSeNetV2 first commit * BiSeNetV2 unittest * remove pytest * add pytest module * fix ConvModule input name * fix pytest error * fix unittest * refactor * BiSeNetV2 Refactory * fix docstrings and add some small changes * use_sigmoid=False * fix potential bugs about upsampling * Use ConvModule instead * Use ConvModule instead * fix typos * fix typos * fix typos * discard nn.conv2d * discard nn.conv2d * discard nn.conv2d * delete **kwargs * uploading markdown and model * final commit * BiSeNetV2 adding Unittest for its modules * BiSeNetV2 adding Unittest for its modules * BiSeNetV2 adding Unittest for its modules * BiSeNetV2 adding Unittest for its modules * BiSeNetV2 adding Unittest for its modules * BiSeNetV2 adding Unittest for its modules * BiSeNetV2 adding Unittest for its modules * Fix README conflict * Fix unittest problem * Fix unittest problem * BiSeNetV2 * Fixing fps * Fixing typpos * bisenetv2

…b#882) * Use docker to skip CUDA installation in CI * Update .github/workflows/build.yml Co-authored-by: Jerry Jiarui XU <xvjiarui0826@gmail.com> * Remove Install system dependencies * Add Install system dependencies for cuda usage * Add Install system dependencies and requirements * fix requirements error Co-authored-by: Jerry Jiarui XU <xvjiarui0826@gmail.com>

* [Enhancement] Change readme style and prepare for metafiles update. * Update apcnet github repo url. * add code snippet. * split code snippet & official repo. * update md2yml hook. * Update metafiles. * Add converted from attribute. * process conflict. * Put defualt variable value. * update bisenet v2 metafile. * checkout to ubuntu environment. * pop empty attribute & make task attribute list. * update readme style * update readme style * update metafiles Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn>

* First Commit * fix typos * fix typos * Fix assertion bug * Adding Assert * Adding Unittest * Fixing typo * Uploading models & logs * Fixing unittest error * changing README.md * changing README.md

* [Improvement] Refactor Swin-Transformer * fixed swin test * update patch emebd, add more tests * fixed test * remove pretrain_style * fixed padding * resolve coments * use mmcv 2tuple * refactor init_cfg Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn>

* add icnet backbone * add icnet head * add icnet configs * nclass -> num_classes * Support ICNet * ICNet * ICNet * Add ICNeck * Add ICNeck * Add ICNeck * Add ICNeck * Adding unittest * Uploading models & logs * Uploading models & logs * add comment * smaller test_swin.py * try to delete test_swin.py * delete test_unet.py * delete test_unet.py * temp * smaller test_unet.py Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn>

* FastFCN first commit * FastFCN first commit * Fixing lint error * Fixing lint error * use for loop on JPU * Use For Loop * Refactor FastFCN * FastFCN * FastFCN * temp * Uploading models & logs (4x4) * Fixing typos * fix typos * rename config * change README.md * use _delete_=True * change configs * change start_level to 0 * change start_level to 0 * jpu * add unittest for start_level!=0

open-mmlab#934) * replace the unavailable pspnet model download address with the updated address * replace the unavailable pspnet model download address with the updated address

* bump to v0.18.0 * replace \ with /

…en-mmlab#923) * [Fix] open-mmlab#916 expection string type classes * add unittests for string path classes * fix double quote string in test_dataset.py * move the import to the top of the file * fix isort lint error fix isort lint error when move the import to the top of the file

* test_torchserver1.1 * test_torchserver1.2 * update * update mmseg_handler.py * update docs * update torchserver * tranfer torchserver to torchserve * update docs * updata torchserve support

…#920) * add codespell pre-commit hook and fix typos * Update mmseg/models/decode_heads/dpt_head.py * Update mmseg/models/backbones/vit.py * Update mmseg/models/backbones/vit.py * fix typos * skip formating typo * deprecate formating * skip ipynb * unstage ipynb changes * unstage ipynb changes * fix typos in ipynb * unstage ipynb changes

* Uploading FastFCN on ADE20K * fixing lint error

* remove python-dev in CI * remove pycocotools installation * remove python 3.9-dev installation dependencies

* [Fix] Fix the broken link * [Fix] Fix the broken link * Update docs/dataset_prepare.md Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn> * Update docs_zh-CN/dataset_prepare.md Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn> * [Docs] Fix the broken link Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn>

MengzhangLI · 2021-10-25T15:10:44Z

The inference metrics of STDC1-Seg50(72.17) and STDC2-Seg50 (74.16) have been aligned.

STDC1-Seg50 (72.17)

STDC2-Seg50 (74.16)

xiexinch added 4 commits September 7, 2021 18:49

refactor stdc code

d7c2168

update key

021eb65

fix backbone inference

499a07f

remove comments

b189253

xiexinch requested a review from MengzhangLI September 9, 2021 09:19

Junjun2016 changed the title ~~WIP support STDC net~~ [Feature] Support STDC Network Sep 9, 2021

Junjun2016 added the WIP Work in process label Sep 9, 2021

FreyWang and others added 2 commits September 11, 2021 14:33

[Fix] Fix iter bug when resuming checkpoint in distributed train (ope…

c4c2fdc

…n-mmlab#866) * [Fix] Fix iter bug when resuming checkpoint in distributed train * fix lint error Signed-off-by: FreyWang <wangwxyz@qq.com>

[Fix] Update mmcv installation in dockerfile (open-mmlab#860)

56e18ba

* update dockerfile * Update docker/Dockerfile Co-authored-by: Jerry Jiarui XU <xvjiarui0826@gmail.com> * add ARG * change dash to bash Co-authored-by: Jerry Jiarui XU <xvjiarui0826@gmail.com>

MengzhangLI reviewed Sep 14, 2021

View reviewed changes

mmseg/models/backbones/stdc.py Outdated Show resolved Hide resolved

MengzhangLI reviewed Sep 14, 2021

View reviewed changes

configs/stdc/fcn_stdc-813_512x1024_40k_cityscapes.py Outdated Show resolved Hide resolved

MengzhangLI reviewed Sep 14, 2021

View reviewed changes

configs/stdc/fcn_stdc-813_512x1024_40k_cityscapes.py Outdated Show resolved Hide resolved

MengzhangLI reviewed Sep 14, 2021

View reviewed changes

mmseg/models/backbones/stdc.py Show resolved Hide resolved

uni19 and others added 16 commits September 14, 2021 19:16

[Docs] Improve docs style (open-mmlab#879)

4d8229c

* Improve docs style * update lists * update the size of image * modify duplicate mmdet3d

[Docs] Use Chinese menu (open-mmlab#880)

4e6f2eb

* Use Chinese menu * replace \ with /

Fixing README (open-mmlab#896)

cff01b3

[Fix] fix_torchserver1.1 (open-mmlab#844)

0b11d58

* test_torchserver1.1 * test_torchserver1.2 * update * update mmseg_handler.py * update docs * update torchserver * tranfer torchserver to torchserve * update docs

The path of the annotation is wrong in the tutorial (open-mmlab#907)

e171e80

* Update config.md * Update config.md

[Fix] Fix loss parse in val_step (open-mmlab#906)

29c82ea

* [Fix] Fix loss parse in val_step * Add val_step unittest * Add train_step unittest

[Feature] Support BiSeNetV1 (open-mmlab#851)

ab12009

* First Commit * fix typos * fix typos * Fix assertion bug * Adding Assert * Adding Unittest * Fixing typo * Uploading models & logs * Fixing unittest error * changing README.md * changing README.md

change name to EncNet (open-mmlab#929)

c1dcf91

Junjun2016 and others added 12 commits September 30, 2021 07:50

fix load ckpt bug in swin (open-mmlab#928)

10886b0

replace the unavailable pspnet model download address with the update… (

d7924ba

open-mmlab#934) * replace the unavailable pspnet model download address with the updated address * replace the unavailable pspnet model download address with the updated address

Bump to v0.18.0 (open-mmlab#940)

8d49dd3

* bump to v0.18.0 * replace \ with /

[Update] Update torchserve about (open-mmlab#951)

47052f7

* test_torchserver1.1 * test_torchserver1.2 * update * update mmseg_handler.py * update docs * update torchserver * tranfer torchserver to torchserve * update docs * updata torchserve support

fix the wrong description in the Chinese documentation (open-mmlab#954)

c403177

first commit (open-mmlab#946)

2676027

Change BiSeNetV2 to Method (open-mmlab#945)

d4d64eb

[Benchmark] Uploading FastFCN on ADE20K (open-mmlab#972)

6d35d76

* Uploading FastFCN on ADE20K * fixing lint error

MengzhangLI self-assigned this Oct 20, 2021

Junjun2016 and others added 3 commits October 24, 2021 11:49

[Fix] Fix bug in CI with py3.9 (open-mmlab#994)

5fdb769

* remove python-dev in CI * remove pycocotools installation * remove python 3.9-dev installation dependencies

vit contiguous (open-mmlab#992)

93f8b14

MengzhangLI added 2 commits October 25, 2021 23:13

Adding STDC for training

0bb88c1

Merge branch 'stdc' of github.com:xiexinch/mmsegmentation into stdc

b762c9b

MengzhangLI mentioned this pull request Oct 25, 2021

[Feature] Support STDC Network (new) #995

Merged

6 tasks

MengzhangLI removed the WIP Work in process label Oct 25, 2021

MengzhangLI changed the title ~~[Feature] Support STDC Network~~ [Feature] Support STDC Network (old, deleted) Oct 25, 2021

MengzhangLI closed this Oct 25, 2021

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Feature] Support STDC Network (old, deleted) #865

[Feature] Support STDC Network (old, deleted) #865

xiexinch commented Sep 9, 2021 •

edited by MengzhangLI

Loading

codecov bot commented Sep 9, 2021 •

edited

Loading

MengzhangLI commented Oct 25, 2021

[Feature] Support STDC Network (old, deleted) #865

[Feature] Support STDC Network (old, deleted) #865

Conversation

xiexinch commented Sep 9, 2021 • edited by MengzhangLI Loading

Motivation

Modification

codecov bot commented Sep 9, 2021 • edited Loading

Codecov Report

MengzhangLI commented Oct 25, 2021

xiexinch commented Sep 9, 2021 •

edited by MengzhangLI

Loading

codecov bot commented Sep 9, 2021 •

edited

Loading