Dataset Versions

v8

2024-01-03 2:53pm

Generated on Jan 3, 2024

Dataset Split

Train Set 82%
2502Images
Valid Set 12%
359Images
Test Set 6%
178Images

Preprocessing

Auto-Orient: Applied
Resize: Fit (black edges) in 640x640
Modify Classes: 0 remapped, 3 dropped
Filter Null: Require all images to contain annotations.

Augmentations

Outputs per training example: 2
Blur: Up to 1px
Noise: Up to 0.5% of pixels