aws / aws/sagemaker-huggingface-inference-toolkit

How to pass device_id to overriden functions?

Open
#66 9 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
270
Forks
60
PR merge metrics
No merged PRs in 30d

Description

I want to select the gpu that my custom model is being loaded into. Can't find a way to do this, as when we do self.load_fn = load_fn we lose the reference to self inside load_fn

Contributor guide

Open the contributing guide

Research direction

Start by tracing the custom model loading path around `self.load_fn = load_fn` and inspect how overridden functions are invoked. Determine how `device_id` could remain available to the custom loader, and consider the work done when a custom model can select the GPU it is loaded into.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.