metlit

hand

Classification

hand Image Dataset

v2

model2

Generated on May 25, 2024

Dataset Split

Train Set 92%
924Images
Valid Set 4%
44Images
Test Set 3%
34Images

Preprocessing

Auto-Orient: Applied
Resize: Stretch to 300x300

Augmentations

Outputs per training example: 3
Flip: Horizontal, Vertical
Crop: 0% Minimum Zoom, 29% Maximum Zoom
Grayscale: Apply to 15% of images
Hue: Between -18° and +18°
Brightness: Between -24% and +24%
Blur: Up to 1.7px
Noise: Up to 1.8% of pixels