Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add mmseg2torchserve tool #552

Merged
merged 10 commits into from
Jul 5, 2021
Merged

Conversation

daavoo
Copy link
Contributor

@daavoo daavoo commented May 13, 2021

This P.R. enables serving MMSegmentation models with TorchServe. A new section Model Serving has been added in useful_tools.md.

@codecov
Copy link

codecov bot commented May 13, 2021

Codecov Report

Merging #552 (7eee01e) into master (8558163) will decrease coverage by 1.61%.
The diff coverage is 74.64%.

❗ Current head 7eee01e differs from pull request most recent head 32907fd. Consider uploading reports for the commit 32907fd to get more accurate results
Impacted file tree graph

@@            Coverage Diff             @@
##           master     #552      +/-   ##
==========================================
- Coverage   86.76%   85.14%   -1.62%     
==========================================
  Files         101      105       +4     
  Lines        5250     5670     +420     
  Branches      849      923      +74     
==========================================
+ Hits         4555     4828     +273     
- Misses        535      663     +128     
- Partials      160      179      +19     
Flag Coverage Δ
unittests 85.14% <74.64%> (-1.62%) ⬇️

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

Impacted Files Coverage Δ
mmseg/datasets/pipelines/transforms.py 97.12% <ø> (ø)
mmseg/models/decode_heads/psp_head.py 100.00% <ø> (ø)
mmseg/models/segmentors/cascade_encoder_decoder.py 100.00% <ø> (ø)
mmseg/models/utils/inverted_residual.py 100.00% <ø> (ø)
mmseg/models/utils/ckpt_convert.py 4.28% <4.28%> (ø)
mmseg/datasets/ade.py 40.54% <24.13%> (-59.46%) ⬇️
mmseg/models/segmentors/encoder_decoder.py 88.35% <25.00%> (-2.50%) ⬇️
mmseg/models/backbones/hrnet.py 82.45% <52.94%> (-1.85%) ⬇️
mmseg/models/backbones/cgnet.py 94.63% <53.84%> (-2.13%) ⬇️
mmseg/models/backbones/unet.py 94.87% <53.84%> (-0.93%) ⬇️
... and 31 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 8558163...32907fd. Read the comment docs.

@xvjiarui
Copy link
Collaborator

Hi @daavoo
Thanks for the nice PR! We will review it recently.

@Junjun2016
Copy link
Collaborator

Hi @daavoo
Sorry for the late reply.
@AllentDan will help with the review.

@AllentDan
Copy link
Member

Hi, it worked all fine for me except the image. We may move 3dogs_mask.png to the folder resources here for better management.

@Junjun2016
Copy link
Collaborator

Hi @AllentDan
@daavoo has updated this PR.
Could you please review it again?

docs/useful_tools.md Outdated Show resolved Hide resolved
docs/useful_tools.md Show resolved Hide resolved
docs/useful_tools.md Show resolved Hide resolved
@Junjun2016
Copy link
Collaborator

Hi @daavoo
Could you please add some unittest?

@AllentDan
Copy link
Member

Hi, @Junjun2016. For codes in tools other than mmseg, we may not consider unittests.

@Junjun2016
Copy link
Collaborator

Hi, @Junjun2016. For codes in tools other than mmseg, we may not consider unittests.

Got it.

@Junjun2016 Junjun2016 merged commit 420783d into open-mmlab:master Jul 5, 2021
bowenroom pushed a commit to bowenroom/mmsegmentation that referenced this pull request Feb 25, 2022
* Add docker/serve

* Add handler

* Add mmseg2torchserve

* Fix mmv minimum version

* Update docs with model serving section

* Update useful_tools.md

* pre-commit

* Update useful_tools.md

* Add 3dogs to resources

* Move mask to resources
aravind-h-v pushed a commit to aravind-h-v/mmsegmentation that referenced this pull request Mar 27, 2023
* * Stabe Diffusion img2img using onnx.

* * Stabe Diffusion inpaint using onnx.

* Export vae_encoder, upgrade img2img, add test

* updated inpainting pipeline + test

* style

Co-authored-by: anton-l <anton@huggingface.co>
aravind-h-v pushed a commit to aravind-h-v/mmsegmentation that referenced this pull request Mar 27, 2023
…pen-mmlab#1191)

* Restore compatibility with old ONNX pipeline.

I think it broke in open-mmlab#552.

* Add missing attribute `vae_encoder`
wjkim81 pushed a commit to wjkim81/mmsegmentation that referenced this pull request Dec 3, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants