Dataset Versions
v3
2023-10-03 7:46am
Generated on Oct 3, 2023
875 Total Images
View All ImagesDataset Split
Train Set 88%
769Images
Valid Set 8%
70Images
Test Set 4%
36Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 512x512
Filter Null: Require at least 95% of images to contain annotations.
Augmentations
Outputs per training example: 3
Flip: Horizontal
Grayscale: Apply to 25% of images
Noise: Up to 1% of pixels