Dataset Versions

v6

Remapping classes

Generated on Apr 18, 2023

Dataset Split

Train Set 65%
530Images
Valid Set 23%
187Images
Test Set 12%
95Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 640x640
Modify Classes: 7 remapped, 0 dropped
Filter Null: Require all images to contain annotations.

Augmentations

Outputs per training example: 1
Crop: 0% Minimum Zoom, 20% Maximum Zoom
Rotation: Between -10° and +10°
Saturation: Between -25% and +25%
Exposure: Between -25% and +25%