Dataset Versions
v2
2024-03-19 6:39pm
Generated on Mar 19, 2024
689 Total Images
View All ImagesDataset Split
Train Set 90%
618Images
Valid Set 7%
47Images
Test Set 3%
24Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Augmentations
Outputs per training example: 3
Crop: 0% Minimum Zoom, 27% Maximum Zoom
Rotation: Between -1° and +1°
Grayscale: Apply to 1% of images
Hue: Between -2° and +2°
Brightness: Between -3% and +3%
Exposure: Between -7% and +7%
Blur: Up to 0.4px