✨ add groundtruth-based scoring and evaluation prompt #151
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Introduced groundtruth/compliance metrics and new evaluation prompt file.
Contributor
There was a problem hiding this comment.
Pull Request Overview
This PR introduces groundtruth-based scoring in the evaluation prompt and extends test data structures to include groundtruth information.
- Add a new prompt file (use_groundtruth_rules.prompty) that defines evaluation guidelines incorporating a groundtruth value.
- Update testrun.mts and resolvers.mts to propagate groundtruth and groundtruthModel properties.
Reviewed Changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.
| File | Description |
|---|---|
| src/prompts/use_groundtruth_rules.prompty | New evaluation prompt file defining groundtruth-based scoring and compliance checks |
| src/genaisrc/src/testrun.mts | Added optional groundtruth and groundtruthModel fields to the test result structure |
| src/genaisrc/src/resolvers.mts | Extended resolver arguments to include groundtruth and groundtruthModel properties |
Improved test result consistency, prompt file error, and command args.
Standardizes metric key separator with METRIC_SEPARATOR and adds a test script.
pelikhan
reviewed
Jun 6, 2025
src/genaisrc/promptpex.genai.mts
Outdated
| scenario, | ||
| input, | ||
| output, | ||
| groundtruth: groundtruth, |
Member
There was a problem hiding this comment.
grouptruth,
if the variable and the field are the same, you can just put the variable and typescript does the rest
pelikhan
reviewed
Jun 6, 2025
pelikhan
reviewed
Jun 6, 2025
| dbg(`evaluating ${metrics.length} metrics with eval model(s) %O`, eModel) | ||
| for (const metric of metrics) { | ||
| const key = metricName(metric)+"|em|"+eModel | ||
| const key = metricName(metric)+METRIC_SEPARATOR+eModel |
Member
There was a problem hiding this comment.
why not using JSON to serialize a key?
Member
|
Refactor groundtruth parsing / mangling into functions (probably new file) to avoide repeating parsing code in various places. |
Added full CLI and script parameter documentation, improved model suggestions, fixed groundtruth model handling, and clarified output formatting in reports.
Refactored runTests options for clarity; normalized groundtruth checks.
- Introduces example-use and test-collection-review docs, updates scripts, fixes CLI usage.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Introduced groundtruth/compliance metrics and new evaluation prompt file.