Dataset Versions

v11

2024-10-28 6:27am

Generated on Oct 28, 2024

Dataset Split

Train Set 80%
12288Images
Valid Set 19%
2930Images
Test Set 1%
156Images

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, 15% Maximum Zoom
Rotation: Between -10° and +10°
Saturation: Between -15% and +15%
Brightness: Between -15% and +15%
Exposure: Between -5% and +5%
Blur: Up to 0.6px
Noise: Up to 1.25% of pixels