Dataset Versions

v12

2023-07-18 5:49pm

Generated on Jul 18, 2023

Dataset Split

Train Set 94%
5163Images
Valid Set 4%
223Images
Test Set 2%
115Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Adaptive Equalization
Modify Classes: 2 remapped, 3 dropped

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
Rotation: Between -15° and +15°
Exposure: Between -25% and +25%
Blur: Up to 2.5px
Noise: Up to 10% of pixels