Dataset Versions
v3
2024-02-08 3:34pm
Generated on Feb 8, 2024
1154 Total Images
View All Images
Dataset Split
Train Set 88%
1011Images
Valid Set 8%
96Images
Test Set 4%
47Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Augmentations
Outputs per training example: 3
Shear: ±13° Horizontal, ±14° Vertical
Grayscale: Apply to 37% of images
Hue: Between -115° and +115°
Brightness: Between -22% and +22%
Blur: Up to 2.4px
Noise: Up to 2.39% of pixels