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
Hello everyone and thank you for your great work on Laplace library.
Based on the documentation for sharing of a fitted Laplace() instance, it is better to use state_dict():
torch.save(la.state_dict(), "model_state_dict.bin")
but unfortunately, it results to thies error:
AttributeError: 'FullLLLaplace' object has no attribute 'state_dict'
Context information:
It is also the same for other classes like KronLLLaplace.
In another experiment I used 'pip install laplace-torch' but the error is still there :( .
That would be perfect if you could help me to resolve this issue in my implementation.
Regards,
Keyvan
The text was updated successfully, but these errors were encountered:
Hello everyone and thank you for your great work on Laplace library.
Based on the documentation for sharing of a fitted Laplace() instance, it is better to use state_dict():
torch.save(la.state_dict(), "model_state_dict.bin")
but unfortunately, it results to thies error:
AttributeError: 'FullLLLaplace' object has no attribute 'state_dict'
Context information:
That would be perfect if you could help me to resolve this issue in my implementation.
Regards,
Keyvan
The text was updated successfully, but these errors were encountered: