Dataset Versions
v2
2023-08-26 2:26pm
Generated on Aug 26, 2023
7700 Total Images
View All Images
Dataset Split
Train Set 88%
6738Images
Valid Set 8%
642Images
Test Set 4%
320Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 640x640
Auto-Adjust Contrast: Using Adaptive Equalization
Grayscale: Applied
Modify Classes: 0 remapped, 0 dropped
Augmentations
Outputs per training example: 3
Flip: Horizontal
Crop: 0% Minimum Zoom, 20% Maximum Zoom
Shear: ±15° Horizontal, ±15° Vertical
Hue: Between -180° and +180°
Exposure: Between -25% and +25%
Blur: Up to 5px
Noise: Up to 24% of pixels