Dataset Versions

v4

2024-03-27 7:00pm

Generated on Mar 27, 2024

Dataset Split

Train Set 94%
4193Images
Valid Set 4%
171Images
Test Set 2%
85Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 7
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Crop: 0% Minimum Zoom, 50% Maximum Zoom
Rotation: Between -45° and +45°
Shear: ±10° Horizontal, ±10° Vertical