Dataset Versions

v10

2024-07-16 4:18am

Generated on Jul 16, 2024

Dataset Split

Train Set 88%
108Images
Valid Set 8%
10Images
Test Set 4%
5Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Crop: 0% Minimum Zoom, 20% Maximum Zoom
Shear: ±2° Horizontal, ±9° Vertical
Grayscale: Apply to 15% of images
Hue: Between -44° and +44°
Brightness: Between -15% and +15%