1
2
3
Currently, there are two pre-trained models to try
- v1: example-based model which takes animation examples for conditioning on the style. 
- v2: label-based model which takes labels for conditioning on the style. This model was trained on a subset of styles