Dataset Versions

v3

2024-11-01 12:53pm

Generated on Nov 1, 2024

Dataset Split

Train Set 78%
1347Images
Valid Set 22%
370Images
Test Set %
0Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
Crop: 0% Minimum Zoom, 12% Maximum Zoom
Shear: ±15° Horizontal, ±10° Vertical
Hue: Between -8° and +8°
Saturation: Between -4% and +4%
Blur: Up to 0.9px
Cutout: 3 boxes with 7% size each