AWS Certification Exam AI Practitioner (Practice Questions) set19
通常
試験時間: 00:00問題時間: 00:00
問題一覧
問題 4 / 65
Which method is most reliable for evaluating the final usefulness and naturalness of generated text and images?
①Automatic evaluation metrics such as BLEU scores
②Automatic metrics such as ROUGE scores.
③Image quality metrics such as FID
④Scoring using large language models for automatic evaluation.
⑤Human evaluation