Sourya Dey

Results 4 comments of Sourya Dey

@Jakob-Unfried : I am facing the same problem and like your `safe_inverse` solution, but I'm not sure how to apply such a thing when doing backpropagation in Pytorch. Do you...

> > @Jakob-Unfried : I am facing the same problem and like your `safe_inverse` solution, but I'm not sure how to apply such a thing when doing backpropagation in Pytorch....

Thank you for your reply. Yes, returning a list of axes / figures would be a great solution. That way they can be used downstream.