Dataset Versions

v1

2023-10-25 6:06pm

Generated on Oct 25, 2023

Dataset Split

Train Set 94%
17604Images
Valid Set 4%
782Images
Test Set 2%
389Images

Preprocessing

Auto-Orient: Applied
Grayscale: Applied
Modify Classes: 11 remapped, 1 dropped
Filter Null: Require all images to contain annotations.

Augmentations

Outputs per training example: 3
Noise: Up to 15% of pixels
Bounding Box: Noise: Up to 10% of pixels