Dataset Versions

v8

2022-09-01 7:03pm

Generated on Sep 2, 2022

Dataset Split

Train Set 77%
1023Images
Valid Set 15%
204Images
Test Set 8%
102Images

Preprocessing

Static Crop: 33-67% Horizontal Region, 33-67% Vertical Region
Auto-Adjust Contrast: Using Adaptive Equalization
Modify Classes: 0 remapped, 1 dropped

Augmentations

Outputs per training example: 3
Flip: Horizontal
Hue: Between -16° and +16°
Saturation: Between -31% and +31%
Exposure: Between -10% and +10%