A question about scaled_loss.backward. It cost me about 1s when i use a single GPU. However

Open
#1,005 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
25/100
Issue type
Bug
Clarity
Needs clarification
Activity status
Stale
Tech stack
python, pytorch

Research direction

Start with scaled_loss.backward in NVIDIA/apex and compare the reported single-GPU and two-GPU timings. Done means explaining the roughly 1-second versus 10-second cost, but the issue provides no model, environment, reproduction, or test to run.

Written by the indexing model from the issue text.

Description

I try to train my model with 2 gpus, this function cost me about 10s. How to explain this phenomenon?

Dominant language
Python
Stars
9k
Forks
1.5k
Avg merge
2d 4h
Merged PRs (30d)
3

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from NVIDIA/apex

All issues in NVIDIA/apex

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.