Dataset Versions

v1

2023-12-25 4:41pm

Generated on Dec 25, 2023

Dataset Split

Train Set 89%
426Images
Valid Set 6%
30Images
Test Set 4%
20Images

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, 31% Maximum Zoom
Rotation: Between -27° and +27°
Shear: ±15° Horizontal, ±15° Vertical
Blur: Up to 2.5px
Noise: Up to 5% of pixels
Cutout: 8 boxes with 15% size each