deepspeedai / deepspeedai/DeepSpeed

[REQUEST] When training a FP16 model, the ability to set some of the layers to FP32

Open
#2,100 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement
Dominant language
Python
Stars
43.1k
Forks
5k
Avg merge
4d 15h
Merged PRs (30d)
112

Description

When training a FP16 model, I wonder if it's possible to set some of the layers to FP32.

I can add .to(torch.float16) and .to(torch.float32) to do the conversion between layers.

So the process will be like:

...
x = f(x)    # x and f(x) are float16
x = x.to(torch.float32)
x = g(x)   # x and g(x) are float32
x = x.to(torch.float16)
x = f(x)   # x and f(x) are float16
...

Contributor guide

Open the contributing guide

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 file, test, or entry point is named. Start by locating the FP16 training path and determine how selected layers could run in FP32 while surrounding tensors are converted as shown; done means a defined way to choose FP32 layers and verify the example behavior.

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

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.