huggingface/lighteval

[BUG] Optimize tokenization

Open

#732 opened on May 15, 2025

View on GitHub
 (1 comment) (0 reactions) (1 assignee)Python (514 forks)auto 404
buggood first issuehelp wantedscience-team

Repository metrics

Stars
 (2,496 stars)
PR merge metrics
 (PR metrics pending)

Description

Describe the bug

For now tokenization is bing made in a for loop, making the whole process very expensive for large benchmarks.

To Reproduce

run any big benchmarks

Expected behavior

Use batch tokenize to speed it up.

Version info

Please provide your operating system, lighteval version or commit if you installed from main, and pip/conda environment if your problem concerns dependencies.

Contributor guide

Research direction
Locate the tokenization loop in the benchmark code and replace it with batch tokenization using the tokenizer's batch encode plus or similar method.
Tech stack
python
Domain
machine learningai
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-3 hours
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.
Clear
Prerequisites
PythonGit
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
75