Get amp handler in a more decent way
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- Python
- Stars
- 9k
- Forks
- 1.5k
- Avg merge
- 2d 4h
- Merged PRs (30d)
- 3
Description
Usually, the location of amp.init() is far from loss.backward(). While it is possible to pass the handler as a parameter to the function which calls loss.backward(), but it is not very decent. I wonder if we can do something like:
import torch
import apex
apex.amp.apex.init()
........
def backward(loss):
with apex.amp.get_default_handler().scale_loss(loss, optimizer) as scaled_loss:
scaled_loss.backward()
optimizer.step()
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
Start by tracing the existing amp.init() and scale_loss usage; the issue names no files, tests, or implementation entry points. Before coding, establish the intended default-handler API and define tests that show how initialization and backward usage should work; the issue currently provides no concrete completion criteria.
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
- 25/100