Dataset Versions

v6

2023-11-09 1:07pm

Generated on Nov 9, 2023

Dataset Split

Train Set 88%
99Images
Valid Set 8%
9Images
Test Set 4%
5Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640

Augmentations

Outputs per training example: 3
Rotation: Between -15° and +15°
Shear: ±15° Horizontal, ±17° Vertical
Grayscale: Apply to 10% of images
Noise: Up to 5% of pixels
Cutout: 4 boxes with 10% size each

Similar Projects

See More
7.9k images 2 models
6.4k images