試験
通常
試験時間: 00:00問題時間: 00:00
問題一覧
問題 23 / 65
Which method is most appropriate for a space probe's AI to fairly compare its performance on text classification, summarization, translation, and other tasks with other models using public datasets?
①cross-validation
②Benchmark dataset
③Human evaluation
④Fine-tuning
⑤Ensemble learning