Dataset Versions
v3
2023-01-07 3:53am
Generated on Jan 7, 2023
310 Total Images
View All Images
Dataset Split
Train Set 87%
270Images
Valid Set 8%
25Images
Test Set 5%
15Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Augmentations
Outputs per training example: 3
Crop: 0% Minimum Zoom, 10% Maximum Zoom
Rotation: Between -3° and +3°
Shear: ±3° Horizontal, ±3° Vertical
Hue: Between -5° and +5°
Saturation: Between -5% and +5%
Brightness: Between 0% and +30%
Exposure: Between -5% and +5%
Blur: Up to 1.5px
Noise: Up to 1% of pixels