Dataset Versions

v2

2023-11-10 2:26pm

Generated on Nov 10, 2023

Dataset Split

Train Set 82%
11160Images
Valid Set 9%
1240Images
Test Set 9%
1240Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 160x160

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
Rotation: Between -45° and +45°
Saturation: Between -30% and +30%
Brightness: Between -30% and +30%
Blur: Up to 1.5px
Noise: Up to 3% of pixels
Cutout: 10 boxes with 5% size each