How to blacklist/whitelist by hand?
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 9k
- Forks
- 1.5k
- Avg merge
- 2d 4h
- Merged PRs (30d)
- 3
Description
In mode O2, my model goes nowhere, whereas without apex or in O1 training happens. Given that this is a variant of another code that works fine in O2, I can pinpoint which tensor is the root of the issue. Is there a way to exclude it from FP16, or do I have to run with O1?
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
No file, test, or entry point is named. Start by reading Apex's O2 mixed-precision documentation and locating the existing whitelist/blacklist configuration. Done would require establishing whether individual tensors can be excluded from FP16 and defining a validation case for the reported O2 training failure.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100