deepspeedai / deepspeedai/DeepSpeed

How to finetune certain portion of the whole parameter

Open
#5,487 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

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

Description

I have to add some LoRA layers by hand(without left) to a pre-trained Multi-modal model, to finetune the model for new data. I want Deepspeed to optimize ONLY the parameters from the LoRA layer rather than all the parameters. Like this
image

The platform is hugging face's transformers and Deepspeed.

Therefore I decorate the Trainer from HF's transformers, as below:
image

Unfortunately, it doesn't work, both LoRA and non-LoRa's weights are not changed during training. It seems that the optimizer in Deepspeed is not the same as that from Pytorch.

My question is, are there any ways that allow me to ONLY finetune certain subnet (LoRA) parameters with Deepspeed+Transformer's Trainer?

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 repository file or test is named. Start by tracing the Hugging Face Trainer and DeepSpeed optimizer integration, then verify how parameter groups and trainable parameters are passed through it. Done means a supported configuration updates only the manually added LoRA parameters while leaving the other model weights unchanged.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
distributed-systems, machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.