Dataset Versions
v2
2023-05-19 5:54pm
Generated on May 19, 2023
3901 Total Images
View All Images
Dataset Split
Train Set 86%
3352Images
Valid Set 9%
332Images
Test Set 6%
217Images
Preprocessing
Resize: Stretch to 224x224
Modify Classes: 0 remapped, 5 dropped
Filter Null: Require at least 20% of images to contain annotations.
Augmentations
Outputs per training example: 2
Flip: Horizontal, Vertical
90° Rotate: Clockwise, Upside Down
Shear: ±16° Horizontal, ±8° Vertical
Grayscale: Apply to 25% of images