Dataset Versions

v6

2023-05-22 4:14pm

Generated on May 22, 2023

Dataset Split

Train Set 94%
2541Images
Valid Set 2%
50Images
Test Set 4%
100Images

Preprocessing

Auto-Orient: Applied

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
Shear: ±4° Horizontal, ±3° Vertical
Saturation: Between -8% and +8%
Brightness: Between -3% and +3%
Exposure: Between -3% and +3%
Blur: Up to 4px