🧪TDD Challenge·intermediate·⏱️ 30–45m·⭐ 175 XP
M-009Build a Fine-Tuning Dataset Validator
Description
Nebula Corp is preparing training data for fine-tuning their customer support model. Before spending money on training, they need to validate the dataset quality. Build a validator that checks training examples for format compliance, detects contradictions, measures diversity, and produces a readiness report with a go/no-go recommendation.
Test Cases (3)
Catches invalid format
Should identify examples with bad format
Input:testFormatValidation()
Expected:CONTAINS:invalid
Detects contradictions
Should find contradictory responses for same question
Input:testContradictions()
Expected:CONTAINS:password
Produces readiness report
Small dataset with issues should not be ready
Input:testReadinessReport()
Expected:CONTAINS:NOT_READY
Related Lessons
Click Run / Check to validate your solution