- Source created: April 25, 2022, 3:37 a.m.
- Default image annotation area: X: 0 - 100% / Y: 0 - 100%
- Annotation point generation: Simple random, 20 points
- Feature extractor: EfficientNet (default)
Confidence threshold: 100% (edit)
Dataset for training students on CoralNet, and how it can be used for measuring percentage cover of the benthic environment of coral reefs.
This source does not have a classifier yet. 20 images required to trigger training the first one.