v3

2024-06-23 4:04am

Generated on Jun 23, 2024

Dataset Split

Train Set 71%
1672Images
Valid Set 19%
449Images
Test Set 9%
220Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 2
Rotation: Between -15° and +15°
Grayscale: Apply to 10% of images
Brightness: Between -15% and +15%
Noise: Up to 0.5% of pixels