AWS Certification Exam AI Practitioner (Practice Questions) set3

問題 35 / 65

What problems should you be aware of when heavily using data augmentation in image classification?

The computational cost being reduced too much by the extension.
Augmented data is always beneficial to the model.
That data augmentation will necessarily increase the number of labels.
The expansion operation must always increase the resolution of the original image.
There is a risk that excessive transformations will change the meaning of labels and increase noise.

当サイトでは、ユーザー体験の向上を目的としてCookieを使用しています。サイトの利用を継続することで、Cookieの使用に同意したものとみなされます。