Dataset Versions
v5
small 5tr-2v-2te-pr512-aug_v3_reds
Generated on Sep 1, 2024
704 Total Images
View All ImagesDataset Split
Train Set 82%
576Images
Valid Set 9%
64Images
Test Set 9%
64Images
Preprocessing
Resize: Fit (black edges) in 512x512
Augmentations
Outputs per training example: 3
Rotation: Between -11° and +11°
Shear: ±0° Horizontal, ±11° Vertical
Saturation: Between -22% and +22%
Brightness: Between -11% and +11%
Noise: Up to 0.22% of pixels