About training precision
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 13.7k
- Forks
- 1.3k
- PR merge metrics
- No merged PRs in 30d
Description
Dear authors,
thanks for your great work!
I would like to know do you use mixed precise for training? e.g. torch.amp = True or torch.cuda.amp.autocast(enabled=use_amp)?
I found that mip-splatting rasterization only support fp32 so I guess you didn't enable amp? But flash attention only support fp16. So I'm confused.
Best,
Contributor guide
No contributing guide indexed for this repository
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
The issue names no file, test, or entry point. Start by locating the training configuration and the mip-splatting and flash-attention setup; done means documenting whether mixed precision is enabled and how the fp32 and fp16 requirements are handled.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 1/5
- Estimated time
- Under an hour
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100