Dataset Versions
v4
2024-07-02 6:42am
Generated on Jul 2, 2024
801 Total Images
View All ImagesDataset Split
Train Set 95%
759Images
Valid Set 3%
23Images
Test Set 2%
19Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Contrast Stretching
Augmentations
Outputs per training example: 3
90° Rotate: Applied
Crop: 0% Minimum Zoom, 5% Maximum Zoom
Grayscale: Apply to 1% of images
Hue: Between -2° and +2°
Saturation: Between -6% and +6%
Brightness: Between -3% and +3%
Exposure: Between -7% and +7%
Blur: Up to 0.4px