Dataset Versions
v1
2024-03-20 4:39pm
Generated on Mar 20, 2024
548 Total Images
View All ImagesDataset Split
Train Set 88%
483Images
Valid Set 7%
41Images
Test Set 4%
24Images
Preprocessing
Auto-Orient: Applied
Resize: Fill (with center crop) in 640x640
Auto-Adjust Contrast: Using Adaptive Equalization
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
Crop: 9% Minimum Zoom, 24% Maximum Zoom
Brightness: Between -14% and +14%