Hi,

I have coded up two models in theano and I want to compare their 
performances on a test data. I have fixed random seeds for parameter 
initializations. 

But when I run the models I get different results each time (although the 
variance is low but it makes comparison almost impossible) , which is 
counter-intuitive. With random seeds fixed, on same input result should be 
same.

is this due to theano's internals? Is there any randomization in the 
internals of theano?

-- 

--- 
You received this message because you are subscribed to the Google Groups 
"theano-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to