deepspeedai / deepspeedai/DeepSpeed

Error with negative slope arg in leaky_relu

Open
#1,282 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
43.1k
Forks
5k
Avg merge
4d 15h
Merged PRs (30d)
112

Description

Hi, when I specify negative_slope=0.2 in torch.nn.functional.leaky_relu, I get the error TypeError: _relu_flops_compute() got an unexpected keyword argument 'negative_slope' while profiling. I get similar positional error while using torch.nn.LeakyReLU(negative_slope=0.2). Without this argument, it works fine. Please let me know if I am doing something wrong.

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

Reproduce the profiling error with torch.nn.functional.leaky_relu using negative_slope=0.2, then compare it with the torch.nn.LeakyReLU case described in the issue. Trace the profiling path that calls _relu_flops_compute and confirm that both forms accept the argument without raising a TypeError.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning, performance
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.