not sure if this is related, but a got a (similar?) error:

"The convolution would produce an invalid shape (dim[3] <= 0)"

apparently some of my input data was smaller than the filter_size of the 
convolution layer. In my case, padding these few unusually small inputs 
helped

-- 

--- 
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