Benjamin-Lee / Benjamin-Lee/deep-rules
mlf-core: a framework for deterministic machine learning
- Dominant language
- HTML
- Stars
- 226
- Forks
- 44
- PR merge metrics
- No merged PRs in 30d
Description
This paper dives into sources of non-determinism in machine learning frameworks: https://arxiv.org/abs/2104.07651
It would be a great reference for our paragraph that starts
> One specific reproducibility pitfall that is often missed in applying deep learning is the default use of non-deterministic algorithms by CUDA/CuDNN backends when using GPUs.
I could add it now or wait until the next version if we're still considering this content frozen for submission.
Contributor guide
Research direction
Review the linked paper and locate the manuscript paragraph beginning “One specific reproducibility pitfall.” Confirm whether the content is frozen for submission; done means adding the paper as a reference if edits are still allowed, or recording that it should wait for the next version.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- machine-learning
- Domain
- documentation, machine-learning
- Issue type
- Documentation
- Difficulty
- 1/5
- Estimated time
- 1-3 hours
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100