balkian
commented
11 months ago Thanks for your interest in senpy!
The only conversion method supported at the moment is based on the distance in VAD space to a set of centroids calculated from data labelled with the two emotion models.
The code for the computation is here.
The specific centroids for Ekman to VAD (and vice-versa) is here.
I hope that answers your question.
Hello,
Some deep learning models provide VAD values in 3D space.
However, the Ekman models is more intuitive to share the results with users.
It seems that senpy can perform the conversion between both models.
Could you please point me to the chunks of code performing this conversion?
Alternatively, are you aware of a straightforward approach to perform this conversion?
Best,
Ed