Dataset Versions
v3
3
Generated on Nov 22, 2024
35 Total Images
View All ImagesDataset Split
Train Set 86%
30Images
Valid Set 11%
4Images
Test Set 3%
1Images
Preprocessing
Auto-Orient: Applied
Resize: Fill (with center crop) in 512x512
Modify Classes: 0 remapped, 2 dropped
Augmentations
Outputs per training example: 3
Rotation: Between -15° and +15°
Blur: Up to 1.8px
Noise: Up to 0.1% of pixels