Dataset Versions
v8
p640.2
Generated on Apr 28, 2023
1004 Total Images
View All Images
Dataset Split
Train Set 70%
702Images
Valid Set 20%
198Images
Test Set 10%
104Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Modify Classes: 1 remapped, 0 dropped
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Hue: Between -5° and +5°
Saturation: Between -48% and +48%
Brightness: Between -22% and +22%
Exposure: Between -11% and +11%
Noise: Up to 7% of pixels