Dataset Versions

v1

2023-09-01 3:27pm

Generated on Sep 1, 2023

Dataset Split

Train Set 85%
1275Images
Valid Set 9%
131Images
Test Set 7%
98Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Crop: 0% Minimum Zoom, 63% Maximum Zoom
Blur: Up to 7.75px
Noise: Up to 25% of pixels
Cutout: 8 boxes with 49% size each