NVIDIA / NVIDIA/apex

What is the difference between training without apex.amp and with opt level = O0?

Open
#911 0 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

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

Description

What is the difference between training without apex.amp and with opt level = O0?
Apex documentation shows "O0: FP32 training / Your incoming model should be FP32 already, so this is likely a no-op." .
But I compared those on M2Det, and I found training with opt level=O0 was much faster.
Why is this happening?

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.

Research direction

No source files or tests are identified in the issue. Start by reproducing the M2Det comparison with training without apex.amp and with opt level O0, then inspect the Apex and PyTorch training paths involved; done means explaining the observed speed difference with supporting measurements.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.