Dataset Versions

v5

2023-08-08 2:04am

Generated on Aug 7, 2023

Dataset Split

Train Set 94%
480Images
Valid Set 3%
15Images
Test Set 4%
18Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 224x224
Filter Null: Require all images to contain annotations.

Augmentations

Outputs per training example: 3
Flip: Horizontal
Crop: 0% Minimum Zoom, 20% Maximum Zoom
Rotation: Between -40° and +40°
Shear: ±15° Horizontal, ±15° Vertical