Dataset Versions
v2
2023-02-27 8:11pm
Generated on Feb 27, 2023
236 Total Images
View All Images
Dataset Split
Train Set 81%
192Images
Valid Set 7%
16Images
Test Set 12%
28Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Modify Classes: 2 remapped, 1 dropped
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise, Upside Down
Rotation: Between -18° and +18°
Shear: ±15° Horizontal, ±15° Vertical
Hue: Between -49° and +49°
Blur: Up to 4.25px