Dataset Versions
v7
2024-03-04 8:47pm
Generated on Mar 4, 2024
824 Total Images
View All ImagesDataset Split
Train Set 87%
720Images
Valid Set 6%
52Images
Test Set 6%
52Images
Preprocessing
Auto-Orient: Applied
Resize: Fit (reflect edges) in 128x128
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
Rotation: Between -2° and +2°
Grayscale: Apply to 15% of images
Hue: Between -15° and +15°
Blur: Up to 3px
Noise: Up to 0.42% of pixels