Update evals aggregations when updating results

[snopoke]

When importing results back into OCS we need to re-run the aggregation logic to update the metrics.

• Create a dataset for testing
• Create a python evaluator that returns {"manual_score": -1}
• Create an evaluation config that connects the dataset and the evaluator
  • Do not select a bot for generation
• Run the evals which will create a result set with the manual_score field set to -1.
• Click the 'edit results' button at the top left
• Download the template
• Update the scores in the CSV file
• Re-upload the updated CSV file back to OCS
• Navigate back to the results view

Result

• The manual scores are shown
• The aggregations are not updated 👈🏻 this is what needs to be fixed

See https://drive.google.com/file/d/1sSrNKoUy95xKtMcGz17RPe8Kw129IQJ3/view for a video of this workflow

[coderabbitai]

📝 CodeRabbit Plan Mode

Generate an implementation plan and prompts that you can use with your favorite coding agent.

• Create Plan

Examples

• Example 1
• Example 2

_{You can configure auto-planning by selecting labels in the issue_enrichment configuration.}

🔗 Similar Issues

Related Issues

• [evals] Update dataset CSV import & export to support participant data and session state #2273

👤 Suggested Assignees

• snopoke

---

🧪 Issue enrichment is currently in open beta.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord or schedule a call!