Dataset Versions
v2
2024-11-02 7:56am
Generated on Nov 2, 2024
1231 Total Images
View All Images
Dataset Split
Train Set 87%
1077Images
Valid Set 8%
103Images
Test Set 4%
51Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Augmentations
Outputs per training example: 3
Flip: Horizontal
Crop: 0% Minimum Zoom, 34% Maximum Zoom
Shear: ±10° Horizontal, ±10° Vertical
Grayscale: Apply to 15% of images
Hue: Between -134° and +134°
Brightness: Between -15% and +15%
Blur: Up to 15.8px
Cutout: 3 boxes with 10% size each
Mosaic: Applied