Skip to content

Conversation

minsuking
Copy link

Summary

Add a Colab-specific setup and troubleshooting section to README to help users run MONAI tutorials reliably on Google Colab.

Changes

  • Document explicit installation of PyTorch 2.6.0 (CUDA 12.4) and MONAI 1.5 on Colab
  • Add troubleshooting notes (torchaudio mismatch, filelock conflict, num_workers warning)
  • Include a quick smoke test command to verify the setup

Motivation

Many users start with Colab and often hit dependency conflicts. This section improves first-time success and reproducibility.

Add a Colab-specific setup (PyTorch 2.6.0 + CUDA 12.4, MONAI 1.5) and troubleshooting notes (torchaudio mismatch, filelock conflict, num_workers guidance). Include a quick smoke test to verify the environment.

Signed-off-by: minsu <[email protected]>
@minsuking
Copy link
Author

Hi, this is my first contribution to MONAI. Thank you for reviewing 🙏

I added a Google Colab setup & troubleshooting section to the README.
This helps users resolve common dependency issues when running MONAI tutorials on Colab.

Verified installation with PyTorch 2.6.0 + CUDA 12.4 and MONAI 1.5

Tested using unet_training_array.py tutorial → training ran successfully and produced expected Dice score

Documented known issues: torchaudio mismatch, filelock conflict, and num_workers guidance

This section should improve first-time success and reproducibility for Colab users.

@ericspod
Copy link
Member

ericspod commented Sep 5, 2025

Hi @minsuking thanks for the contribution! I think it's fine as it is with a minor addition to change code in notebooks. I think notebook running in Colab is still fine with the environment that's installed, I just checked the MONAI 101 tutorial and I got Pytorch 2.8 that worked fine, but these instructions are good for install MONAI with other dependencies. It may be worth noting that people will usually run notebooks in Colab directly rather than install and run things on the command line.

@minsuking
Copy link
Author

@ericspod Thank you very much for your kind review and guidance.
I’ve applied the suggested change as you recommended.
This was my first contribution to MONAI, and I really appreciate the opportunity to learn through the review process.
I’d also like to continue contributing in the future, especially by helping with bug fixes and improving documentation where it may be helpful.
Please let me know if there are specific areas where I can support.

Copy link
Member

@ericspod ericspod left a comment

Choose a reason for hiding this comment

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

Looks good to me!

@ericspod
Copy link
Member

ericspod commented Sep 7, 2025

Thanks again @minsuking If you're looking to contribute further we're putting together efforts on implementing work from MICCAI if that's of interest. Other than that we're always open to updates and fixes here in the tutorials. We do plan to reorganise and streamline things here soon as well.

@ericspod ericspod enabled auto-merge (squash) September 7, 2025 20:20
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.

2 participants