You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Could you please clarify how the recovered latent variables are derived from the trained model? Is there a specific process or transformation applied to the encoder's output to align or correlate it with the labels for the MIC calculation?
Hello, the first author of CausalVAE has completed her internship with us. For a more detailed and accurate reply, you can email your questions to Mengyue Yang ([email protected]).
This previous discussion may also be of some help: #64
From the code running results, the decoded z is not corresponding to the label (permutated)?
The text was updated successfully, but these errors were encountered: