You just need to exclude the weights from being treated as parameters for learning, without looking at lasagne's documentation its hard to know the proper way to do that, you might have better luck in their support forums.
On Monday, August 1, 2016 at 10:40:35 AM UTC-4, Geppetto Null wrote: > > Hi everyone, > > I would like to ask if there is a way of performing 2D convolution between > an input tensor and a filter, whose weights, though, I need to keep > constant (non-learnable). Actually, I'm using Lasagne, and I thought that > would be trivial, but I cannot find the way (after much googling). > > Thanks a lot, best, > Christos > -- --- 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.
