Dataset Versions

v3

2024-04-09 5:01pm

Generated on Apr 9, 2024

Dataset Split

Train Set 88%
198Images
Valid Set 9%
20Images
Test Set 4%
8Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Grayscale: Apply to 21% of images
Noise: Up to 0.97% of pixels
Cutout: 7 boxes with 9% size each
Mosaic: Applied
Bounding Box: Rotation: Between -14° and +14°