michaelfeil / michaelfeil/infinity

Abstraction for `resolve_torch_dtype_device(dtype: Dtype, device: Device) -> tuple[quantization_type, torch.device, torch.dtype]`

Open
#424 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
2.9k
Forks
206
PR merge metrics
No merged PRs in 30d

Description

### Feature request

Too much boilerplate template:

Resolves loading, quantization, and device

Eg. if
device: auto -> torch.cuda.is_available() -> cuda or mps.
dtype: float32 -> float32, no quantization
dtype: float16 -> float16, no quantization
dtype: bfloat16 -> float16, no quantization
dtype: auto -> (bfloat16 if possible else float16) if device is cuda else float32, no quantization
dtype: int8 -> float32, int8 quantization
dtype: fp8 -> float32, fp8 quantization

### Motivation

-

### Your contribution

-

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 files, tests, or entry points are named. Start by locating the existing loading, quantization, and device-resolution paths, then verify that the requested dtype and device cases are defined and covered by tests.

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.