Dataset Versions
Versions
v1
2024-10-17 4:56pm
Generated on Oct 17, 2024
1062 Total Images
View All ImagesDataset Split
Train Set 89%
944Images
Valid Set 9%
94Images
Test Set 2%
24Images
Preprocessing
Auto-Orient: Applied
Augmentations
Outputs per training example: 2
Cutout: 1 boxes with 10% size each
Bounding Box: Noise: Up to 0.1% of pixels