Project-MONAI / Project-MONAI/MONAI

Should we support random operation of PyTorch Tensor

Open
#3,180 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

question
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

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. 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

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.