Skip to content

Latest commit

 

History

History
8 lines (5 loc) · 500 Bytes

File metadata and controls

8 lines (5 loc) · 500 Bytes

Approaches to Training the Model

FishNET V1 used Teachable Machine to train the back-end AI model. This process uses transfer learning to fine-tune an image classification model on our dataset of fish images.

This approach does not allow for customizing training parameters.

Possible Improvements

  • FishNET V2 will implement YOLO V8 which will give the system image segmentation capabilities, and object tracking.