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
I understand that those usages are different, but it feels wrong to me to silently ignore a parameter. Also, I do not see why beta could not be applied in the IWAE case. I therefore suggest one of:
Raise a warning if K is set >1 and Beta is specified by the user to avoid misunderstanding
Change the IWAE loss to be computer as the beta-VAE one.
Let me know so I can propose a PR.
Thanks for the great software!
The text was updated successfully, but these errors were encountered:
Thank you so much for your attention to PyPOTS! You can follow me on GitHub
to receive the latest news of PyPOTS. If you find PyPOTS helpful to your work, please star⭐️ this repository.
Your star is your recognition, which can help more people notice PyPOTS and grow PyPOTS community.
It matters and is definitely a kind of contribution to the community.
I have received your message and will respond ASAP. Thank you for your patience! 😃
Issue description
In the backbone of GPVAE, a different path is taken depending on whether IWAE is used or not.
In the case K=1, the elbo is updated like this:
In the case K>1, the elbo is updated like that:
I understand that those usages are different, but it feels wrong to me to silently ignore a parameter. Also, I do not see why beta could not be applied in the IWAE case. I therefore suggest one of:
Let me know so I can propose a PR.
Thanks for the great software!
The text was updated successfully, but these errors were encountered: