Dataset Versions
v2
2023-07-07 9:58pm
Generated on Jul 7, 2023
384 Total Images
View All Images
Dataset Split
Train Set 88%
336Images
Valid Set 9%
36Images
Test Set 3%
12Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Grayscale: Applied
Tile: 2 rows x 2 columns
Augmentations
Outputs per training example: 3
Flip: Horizontal
Rotation: Between -16° and +16°
Shear: ±7° Horizontal, ±45° Vertical
Hue: Between -54° and +54°
Saturation: Between -71% and +71%
Noise: Up to 2% of pixels
Cutout: 6 boxes with 14% size each