Dataset Versions
v5
2022-10-01 4:33pm
Generated on Oct 1, 2022
5226 Total Images
View All Images
Dataset Split
Train Set 92%
4818Images
Valid Set 4%
206Images
Test Set 4%
202Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 128x128
Modify Classes: 0 remapped, 1 dropped
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -25° and +25°
Shear: ±15° Horizontal, ±15° Vertical
Grayscale: Apply to 15% of images
Hue: Between -34° and +34°
Saturation: Between -35% and +35%
Brightness: Between -20% and +20%
Exposure: Between -18% and +18%
Blur: Up to 0.75px
Noise: Up to 5% of pixels