Dock inspectors have a modest labeled plank-photo set and a JumpStart vision model that already knows generic image features. Random-initialized training would waste that head start. What should they do?
Select an answer to reveal the explanation.
Short Explanation
The JumpStart vision model already knows generic image features. Fine-tune it on the plank photos so that head start transfers. Training from random weights throws the head start away, and Transcribe or Personalize never painted a vision model.
Full Explanation
Fine-tune a SageMaker JumpStart pretrained model on the custom labeled set so generic image features transfer. Random-initialized training wastes that head start. Transcribe and Personalize are not vision fine-tunes.