Dataset Versions
v4
2024-06-14 4:26pm
Generated on Jun 14, 2024
78 Total Images
View All Images
Dataset Split
Train Set 88%
69Images
Valid Set 4%
3Images
Test Set 8%
6Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Modify Classes: 2 remapped, 0 dropped
Augmentations
Outputs per training example: 3
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Counter-Clockwise
Crop: 0% Minimum Zoom, 25% Maximum Zoom
Grayscale: Apply to 18% of images
Hue: Between -11° and +11°
Saturation: Between -25% and +25%
Brightness: Between -13% and +13%
Exposure: Between -8% and +8%
Blur: Up to 1.2px