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 MONAI quickstart example #2773

Merged
merged 29 commits into from
Feb 28, 2024
Merged

Add MONAI quickstart example #2773

merged 29 commits into from
Feb 28, 2024

Conversation

charlesbvll
Copy link
Member

@charlesbvll charlesbvll commented Jan 4, 2024

Issue

Description

MONAI (Medical Open Network for AI) is a PyTorch-based, open-source framework for deep learning in healthcare imaging, part of the PyTorch Ecosystem and we don't yet have an example for it.

Related issues/PRs

N/A

Proposal

Explanation

Add a quickstart example using Flower and MONAI.

Checklist

  • Implement proposed change
  • Update the changelog entry below
  • Make CI checks pass
  • Ping maintainers on Slack (channel #contributions)

Changelog entry

MONAI (Medical Open Network for AI) is a PyTorch-based, open-source framework for deep learning in healthcare imaging, part of the PyTorch Ecosystem. You can find the example at https://github.com/adap/flower/tree/main/examples/quickstart-monai.

Any other comments?

N/A

@charlesbvll charlesbvll marked this pull request as ready for review January 4, 2024 21:16
Copy link
Contributor

@jafermarq jafermarq left a comment

Choose a reason for hiding this comment

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

Hi @charlesbvll, some minor comments below. More concerning is the fact that I'm getting segmentation faults when running this on Ubuntu x64 (no matter if I use or don't use GPU) (but no issues on mac arm64). Seems to be related to the dataloaders...


Update: setting num_workers=0 in the dataloaders seems to fix the issue in Ubuntu. But this slows down training so I don't think it's an ideal solution

@jafermarq jafermarq added the part: examples Add or update a Flower example label Jan 4, 2024
@charlesbvll
Copy link
Member Author

Hi @charlesbvll, some minor comments below. More concerning is the fact that I'm getting segmentation faults when running this on Ubuntu x64 (no matter if I use or don't use GPU) (but no issues on mac arm64). Seems to be related to the dataloaders...

Update: setting num_workers=0 in the dataloaders seems to fix the issue in Ubuntu. But this slows down training so I don't think it's an ideal solution

@jafermarq I don't know of any other solutions to fix num_workers > 0 related errors, even if the training is slower, I think it's better to set num_workers=0 for stability.

Copy link
Contributor

@jafermarq jafermarq left a comment

Choose a reason for hiding this comment

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

Let's keep the default num_workers

jafermarq
jafermarq previously approved these changes Jan 17, 2024
@jafermarq
Copy link
Contributor

jafermarq commented Jan 17, 2024

@charlesbvll, I added a link to this example in the top-level readme. Also, adding MONAI to the list of supported frameworks. Looks good?

@danieljanes danieljanes enabled auto-merge (squash) February 28, 2024 19:13
@danieljanes danieljanes merged commit 42f13ad into main Feb 28, 2024
28 checks passed
@danieljanes danieljanes deleted the add-monai-quickstart branch February 28, 2024 19:23
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
part: examples Add or update a Flower example
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants