Dataset Versions
v3
2023-06-11 11:01pm
Generated on Jun 11, 2023
2553 Total Images
View All ImagesDataset Split
Train Set 88%
2235Images
Valid Set 8%
208Images
Test Set 4%
110Images
Preprocessing
Auto-Orient: Applied
Resize: Fit (reflect edges) in 416x416
Modify Classes: 0 remapped, 2 dropped
Filter Null: Do not filter any null images.
Augmentations
Outputs per training example: 3
Rotation: Between -15° and +15°
Saturation: Between -10% and +10%
Exposure: Between -15% and +15%
Noise: Up to 5% of pixels