Dataset Versions

v5

2024-05-31 5:05pm

Generated on May 31, 2024

Dataset Split

Train Set 87%
1059Images
Valid Set 8%
103Images
Test Set 4%
50Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Crop: 0% Minimum Zoom, 30% Maximum Zoom
Rotation: Between -19° and +19°
Brightness: Between -31% and +31%
Blur: Up to 4.5px
Noise: Up to 2.24% of pixels