Dataset Versions
v2
2024-11-16 12:36pm
Generated on Nov 16, 2024
994 Total Images
View All Images
Dataset Split
Train Set 94%
939Images
Valid Set 4%
36Images
Test Set 2%
19Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 224x224
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Grayscale: Apply to 10% of images
Hue: Between -21° and +21°
Saturation: Between -25% and +25%
Brightness: Between -24% and +24%
Noise: Up to 1.96% of pixels
Cutout: 4 boxes with 14% size each