Project-MONAI / Project-MONAI/MONAI
Should we support random operation of PyTorch Tensor
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 8.7k
- Forks
- 1.6k
- Avg merge
- 5d 1h
- Merged PRs (30d)
- 20
Description
Is your feature request related to a problem? Please describe.
Hi @wyli @ericspod @rijobro ,
Currently, all the random operations in transforms are based on numpy, especially for deterministic purpose.
Most of them are light computation, but some are heavy to generate a random array, for example:
https://github.com/Project-MONAI/MONAI/blob/dev/monai/transforms/intensity/array.py#L1910
It may block the GPU Tensor transform.
Should we consider to support Tensor random operations?
What do you guys think?
Thanks.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Read monai/transforms/intensity/array.py around line 1910 and identify the NumPy-based random operations that can generate large arrays. Determine the scope and requirements for Tensor-based random operations, including how deterministic behavior should be preserved. Done should mean the project has an agreed implementation path for avoiding the reported GPU-transform blocking concern.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning, performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100