Dataset Versions
v2
2024-12-01 12:42pm
Generated on Dec 1, 2024
1039 Total Images
View All Images
Dataset Split
Train Set 94%
980Images
Valid Set 4%
39Images
Test Set 2%
20Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 256x256
Modify Classes: 0 remapped, 3 dropped
Augmentations
Outputs per training example: 7
Flip: Horizontal, Vertical
Crop: 0% Minimum Zoom, 30% Maximum Zoom
Rotation: Between -15° and +15°
Shear: ±15° Horizontal, ±15° Vertical
Brightness: Between -25% and +25%
Blur: Up to 2.5px
Noise: Up to 1.88% of pixels