Sikhin V C

text

Classification

text Image Dataset

v1

augmented

Generated on Nov 25, 2021

Dataset Split

Train Set 87%
5163Images
Valid Set 8%
487Images
Test Set 4%
254Images

Preprocessing

Auto-Orient: Applied

Augmentations

Outputs per training example: 3
Crop: 0% Minimum Zoom, 47% Maximum Zoom
Rotation: Between -10° and +10°
Shear: ±15° Horizontal, ±15° Vertical
Cutout: 3 boxes with 12% size each