Dataset Versions

v11

2024-12-29 6:51pm

Generated on Dec 29, 2024

Dataset Split

Train Set 92%
12625Images
Valid Set 5%
633Images
Test Set 3%
475Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Modify Classes: 32 remapped, 0 dropped

Augmentations

Outputs per training example: 5
Crop: 0% Minimum Zoom, 20% Maximum Zoom
Rotation: Between -10° and +10°
Shear: ±6° Horizontal, ±6° Vertical
Brightness: Between -20% and +20%
Exposure: Between -15% and +15%
Blur: Up to 0.7px