Dataset Versions

v7

2023-11-12 5:21pm

Generated on Nov 12, 2023

Dataset Split

Train Set 79%
75Images
Valid Set 13%
12Images
Test Set 8%
8Images

Preprocessing

Auto-Orient: Applied
Filter Null: Require at least 80% of images to contain annotations.

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Saturation: Between -56% and +56%