Fie

CHAR

Classification

CHAR Image Dataset

v1

2023-06-26 2:03am

Generated on Jun 25, 2023

Dataset Split

Train Set 92%
2442Images
Valid Set 8%
204Images
Test Set %
0Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 416x416

Augmentations

Outputs per training example: 3
90° Rotate: Clockwise, Counter-Clockwise
Rotation: Between -40° and +40°
Shear: ±25° Horizontal, ±25° Vertical
Brightness: Between -30% and +30%
Blur: Up to 2.5px
Noise: Up to 10% of pixels
Cutout: 5 boxes with 12% size each