Dataset Versions
v3
Balanced data
Generated on Oct 23, 2023
448 Total Images
View All ImagesDataset Split
Train Set 87%
390Images
Valid Set 8%
38Images
Test Set 4%
20Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 128x128
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise
Rotation: Between -25° and +25°
Shear: ±13° Horizontal, ±23° Vertical
Saturation: Between -25% and +25%
Noise: Up to 2% of pixels