Dataset Versions

v8

2023-12-06 8:36pm

Generated on Dec 6, 2023

Dataset Split

Train Set 92%
684Images
Valid Set 4%
32Images
Test Set 4%
28Images

Preprocessing

Auto-Orient: Applied
Auto-Adjust Contrast: Using Contrast Stretching
Tile: 2 rows x 2 columns
Modify Classes: 2 remapped, 0 dropped

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise
Shear: ±15° Horizontal, ±15° Vertical
Saturation: Between -50% and +50%
Brightness: Between -36% and +36%
Exposure: Between -27% and +27%