deepspeedai / deepspeedai/DeepSpeed
Error with negative slope arg in leaky_relu
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
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
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