L T

L 3

Instance Segmentation

L 3 Image Dataset

v7

2023-11-19 1:00pm

Generated on Nov 19, 2023

Dataset Split

Train Set 87%
975Images
Valid Set 8%
93Images
Test Set 5%
57Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 128x70
Auto-Adjust Contrast: Using Contrast Stretching

Augmentations

Outputs per training example: 3
Rotation: Between -5° and +5°
Shear: ±5° Horizontal, ±5° Vertical
Hue: Between -25° and +25°
Saturation: Between -25% and +25%
Brightness: Between -10% and +10%
Exposure: Between -10% and +10%
Cutout: 1 boxes with 5% size each
Mosaic: Applied
Bounding Box: Rotation: Between -5° and +5°
Bounding Box: Shear: ±5° Horizontal, ±5° Vertical
Bounding Box: Exposure: Between -25% and +25%