Does ot functions returns in torch tensor? #810
Unanswered
98srkulkarni98
asked this question in
Q&A
Replies: 1 comment
-
|
Hello, most POT functions now work on multiple backends that is if you give the and pytorch tensor they will return a pytorch tensor. This is true of ot.sinkhorn, ot.emd and ot.solve that is a wrapper over multiple solvers. ot.unif returns a numpy array by default but it is easy to convert them to pytorch or to compute a uniform vector with pytorch only. Finally Note that ot.solve does not require the weights to be given when they are uniform. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
As I know this package initially uses numpy.
And I am working with torch tensor, I would like to know all the function calls like ot.unif and ot.sinkhorn do they return tensor?
is there a simple work around
Beta Was this translation helpful? Give feedback.
All reactions