Probability for all the classes, not just the most probable
30 views
Skip to first unread message
Alessandro Ferrari
Dec 3, 2015, 6:27:29 AM12/3/15
to pystruct
Hi,
if in EdgeGraphCRF, I would like to know the probability associated to each class for each node, and not just the most probable class, as it happens in the predict method, is there any way to do so? I would like to have the top 3 score.
Thank you to your community for the awesome work.
if in EdgeGraphCRF, I would like to know the probability associated to each class for each node, and not just the most probable class, as it happens in the predict method, is there any way to do so? I would like to have the top 3 score.
Thank you to your community for the awesome work.
Andy
Dec 5, 2015, 5:49:50 PM12/5/15
to pyst...@googlegroups.com
Hi Allesandro.
Unfortunately that is not really possible with the approach I'm taking.
There are some papers on "top k inference", maybe look at the stuff by Christoph Lampert.
PyStruct could be extended to do that but it is highly non-trivial.
Best,
Andy
Unfortunately that is not really possible with the approach I'm taking.
There are some papers on "top k inference", maybe look at the stuff by Christoph Lampert.
PyStruct could be extended to do that but it is highly non-trivial.
Best,
Andy
--
You received this message because you are subscribed to the Google Groups "pystruct" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pystruct+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Reply all
Reply to author
Forward
0 new messages