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

2812 enhance swish and efficientnet #2813

Merged
merged 18 commits into from
Aug 26, 2021

Conversation

yiheng-wang-nv
Copy link
Contributor

@yiheng-wang-nv yiheng-wang-nv commented Aug 20, 2021

Signed-off-by: Yiheng Wang vennw@nvidia.com

Fixes #2812 .

This PR also enhances efficientnet via:

  1. adding EfficientNetBNFeatures class, which is used to build a backbone for segmentation and object detection tasks.
  2. remove unnecessary code (load fc) for _load_state_dict, and remove the hard requirement of in_channels = 3. Now, the network with in_channels != 3 will load weights for all layers except the first conv layer.
  3. add more weights urls.

The enhancement refers to: https://github.com/rwightman/pytorch-image-models

Status

Ready

Types of changes

  • Non-breaking change (fix or new feature that would not break existing functionality).
  • Breaking change (fix or new feature that would cause existing functionality to change).
  • New tests added to cover the changes.
  • Integration tests passed locally by running ./runtests.sh -f -u --net --coverage.
  • Quick tests passed locally by running ./runtests.sh --quick --unittests.
  • In-line docstrings updated.
  • Documentation updated, tested make html command in the docs/ folder.

Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
yiheng-wang-nv and others added 2 commits August 23, 2021 16:44
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
@yiheng-wang-nv yiheng-wang-nv changed the title 2812 enhance swish 2812 enhance swish and efficientnet Aug 23, 2021
@yiheng-wang-nv yiheng-wang-nv marked this pull request as ready for review August 23, 2021 09:00
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
Signed-off-by: Yiheng Wang <vennw@nvidia.com>
@wyli
Copy link
Contributor

wyli commented Aug 23, 2021

/build

@wyli
Copy link
Contributor

wyli commented Aug 24, 2021

/build

monai/networks/nets/efficientnet.py Outdated Show resolved Hide resolved
monai/networks/nets/efficientnet.py Outdated Show resolved Hide resolved
@wyli
Copy link
Contributor

wyli commented Aug 25, 2021

/build

Copy link
Contributor

@wyli wyli left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks, there are some minor style issues reported here https://github.com/Project-MONAI/MONAI/pull/2813/checks?check_run_id=3419685439 please help fix them here or in another PR

@wyli wyli enabled auto-merge (squash) August 25, 2021 19:52
@wyli
Copy link
Contributor

wyli commented Aug 25, 2021

/build

1 similar comment
@wyli
Copy link
Contributor

wyli commented Aug 25, 2021

/build

@wyli wyli merged commit 415f994 into Project-MONAI:dev Aug 26, 2021
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.

Enhance MemoryEfficientSwish with optional import of SiLU
2 participants