Dataset Versions

v4

2024-02-08 11:43pm

Generated on Feb 9, 2024

Dataset Split

Train Set 91%
310Images
Valid Set 6%
20Images
Test Set 3%
10Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 5
Hue: Between -7° and +7°
Saturation: Between -8% and +8%
Brightness: Between -10% and +10%
Exposure: Between -5% and +5%
Blur: Up to 0.2px
Noise: Up to 0.47% of pixels