Dataset Versions
Versions
v1
2023-12-01 2:19pm
Generated on Dec 1, 2023
2600 Total Images
View All Images
Dataset Split
Train Set 92%
2400Images
Valid Set 4%
100Images
Test Set 4%
100Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Augmentations
Outputs per training example: 3
Flip: Horizontal
Rotation: Between -19° and +19°
Grayscale: Apply to 25% of images
Saturation: Between -82% and +82%
Exposure: Between -34% and +34%
Blur: Up to 3px
Noise: Up to 9% of pixels
Cutout: 5 boxes with 17% size each