AWS Certification Exam AI Practitioner (Practice Questions) set16
通常
試験時間: 00:00問題時間: 00:00
問題一覧
問題 55 / 65
I want to adapt a foundation model to ocean data with limited computational resources. Which efficient method allows learning only a small number of additional parameters while keeping most of the parameters fixed?
①Parameter-efficient fine-tuning methods such as adapters and LoRA.
②Retrain the full model with a large amount of labeled data.
③Quantize the model to improve accuracy.
④Rule-based adjustments for post-processing model outputs
⑤Make it lightweight by removing the fully connected layers.