We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
Welcome to the prompt wiki!
These are example of how to use this platform with datasets currently in Tech4Dev workspace. Throughout, we assume
You are in eval data directory:
git clone https://github.com/ProjectTech4DevAI/file-search-eval.git cd file-search-eval
You have followed the general setup described in the README
Kunji data is split across two Github repositories differentiated by phase:
They have the same structure, so using either can be reduced to variable naming. For this example:
KUNJI_DATA=veddis-eval-data # KUNJI_DATA=kunji-haryana-data # also valid, for Phase 2 Haryana data
From there we can clone the repository and run the scripts:
git clone https://github.com/ProjectTech4DevAI/${KUNJI_DATA}.git gt="-g $KUNJI_DATA/assessments/correct" ./bin/run-prompts.sh $gt -p $KUNJI_DATA/prompts -d $KUNJI_DATA/docs \ | ./bin/run-evals.sh $gt -u $KUNJI_DATA/prompts/user > evaluations.jsonl