-
Notifications
You must be signed in to change notification settings - Fork 79
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
NeuralODE trajectory API is quite limiting #113
Comments
Hello, This seems to be a problem with Torchdyn. A workaround might be to use torchdiffeq instead. You could also write your own custom Euler integration method. Unfortunately, as the NeurIPS deadline is only one month away, I will not have time to look to this issue especially as it is not really related to TorchCFM but rather to Torchdyn. Best, |
I would implement this by inheriting from the SuperResModel class i.e.
I think we do this in the conditional example |
Hi,
I am trying to use your package with the
from torchcfm.models.unet.unet import SuperResModel
and other custom models that have kwargs in their forward method, but I think that the NeuralODE.trajectory method is not compatible with those models?Could you please try to add a model_kwargs parameters to NeuralODE.trajectory, NeuralODE.forward, etc?
Thanks!
The text was updated successfully, but these errors were encountered: