pytorch / pytorch/benchmark

Compare some PyTorch models against corresponding JAX models in the suite

Open
#345 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
1k
Forks
346
PR merge metrics
No merged PRs in 30d

Description

This was brought up by @bertmaher offline, I am just writing an issue for it. It would be nice if we could have a set of models where JAX is more competitive than PyTorch and be able to run comparisons on both in torchbench.

Some prior work: https://github.com/TheSalon/fast-dpsgd benchmarked various differential private models between PT, JAX, and some other frameworks.

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

Start by reviewing the TorchBench suite and the prior fast-dpsgd benchmark linked in the issue. Identify suitable corresponding models and define how both implementations would be run and compared in TorchBench. Done means the suite contains reproducible comparisons showing models where JAX is more competitive than PyTorch.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning, performance, testing-qa
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.