Dataset Versions

v19

2024-08-14 6:30am

Generated on Aug 14, 2024

Dataset Split

Train Set 97%
3871Images
Valid Set 2%
99Images
Test Set 1%
40Images

Preprocessing

Auto-Orient: Applied
Resize: Fit (white edges) in 1000x1000

Augmentations

Outputs per training example: 7
Flip: Horizontal, Vertical
90° Rotate: Applied
Crop: 0% Minimum Zoom, 30% Maximum Zoom
Rotation: Between -12° and +12°
Shear: ±10° Horizontal, ±21° Vertical
Grayscale: Apply to 15% of images
Saturation: Between -34% and +34%
Brightness: Between -16% and +16%
Exposure: Between -12% and +12%
Blur: Up to 0.8px
Noise: Up to 0.65% of pixels

Similar Projects

See More
9.9k images 1 model