Documents Computer Vision Model
1526 Total Images
View All Images
Dataset Split
Train Set 88%
1350Images
Valid Set 10%
150Images
Test Set 2%
26Images
Preprocessing
Auto-Orient: Applied
Resize: Stretch to 960x960
Augmentations
Outputs per training example: 3
Rotation: Between -11° and +11°
Grayscale: Apply to 15% of images
Brightness: Between -15% and +15%