Use
theano.tensor.ge() 

rather than 
self.y_pred >= 0.7

About 2/3 of the way down in 
http://deeplearning.net/software/theano/library/tensor/basic.html

On Tuesday, March 7, 2017 at 2:47:27 AM UTC-6, Feras Almasri wrote:
>
>
> How to fix this issue. I had this result after using this indexing but 
> then I remove it from the code and I still have the same message ? 
>
> # idxs = (self.y_pred >= 0.7).nonzero()
> # self.y_pred = T.set_subtensor(self.y_pred[idxs], 1)
> # idxs = (self.y_pred < 0.7).nonzero()
> # self.y_pred = T.set_subtensor(self.y_pred[idxs], 0)
>
>
>
>  
>

-- 

--- 
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