mattpocock / mattpocock/evalite
Threshold as evalite config param
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- TypeScript
- Stars
- 1.7k
- Forks
- 102
- PR merge metrics
- No merged PRs in 30d
Description
We have different evals that require different thresholds. For one eval a score of 80% is good enough, but for a different eval we need 95%.
Right now there's only a global threshold. Just as a specific eval can override the trialCount we'd want to override the threshold per eval.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by locating the global threshold configuration and the per-evaluation trialCount override in the TypeScript source. Follow how evaluation configuration is read, then make the threshold overridable per eval while preserving the global threshold as the default; done means different evals can use different thresholds.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- ai, testing
- Issue type
- Feature
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100