Na test wey you fit repeat to measure how well AI model or app dey perform for selected task or criteria.
English definition
A repeatable test that measures how well an AI model or application performs on selected tasks or criteria.
AI Evaluation & Testing
4 terms
AI terms in AI Evaluation & Testing
Na set of data or tasks wey dem define to compare how different models or systems perform.
English definition
A defined dataset or set of tasks used to compare the performance of models or systems.
Na data wey get correct reference answers or labels wey dem use to check AI output.
English definition
Data containing reference answers or labels used to check an AI system’s outputs.
Na when people check AI answers with clear criteria like correctness, clarity, or usefulness.
English definition
Assessment of AI outputs by people using defined criteria such as correctness, clarity, or usefulness.