EleutherAI/lm-evaluation-harness

Add better test coverage for models

Open

#1,613 opened on Mar 20, 2024

View on GitHub
 (3 comments) (0 reactions) (0 assignees)Python (3,306 forks)auto 404
feature requestgood first issuehelp wanted

Repository metrics

Stars
 (12,755 stars)
PR merge metrics
 (Avg merge 15d 7h) (11 merged PRs in 30d)

Description

We should improve our test coverage of the various model classes supported.

Ideally we'd have at least https://github.com/EleutherAI/lm-evaluation-harness/blob/main/tests/models/test_huggingface.py tests for every model class. This would require, for API models, mocking out the API call with a dummy one, and probably something similar for VLLM, which currently requires CUDA to be run so we don't have the tests run on the CI.

Help or contributions on this welcome! Else, I'll look into this hopefully soonish.

Contributor guide

Research direction
Examine the existing test file tests/models/test huggingface.py to understand the testing pattern, then create analogous tests for other model classes, mocking API calls and handling CUDA dependencies by using CPU only or mock modules.
Tech stack
pythonpytorch
Domain
backend
Issue type
Test
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Mostly clear
Prerequisites
GitPythonpytest
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60