Hi everyone I've been having this issue lately and can't figure out what
seems to be the problem.
In the main body of the code I have two theano.functions, one for the
training and one for the prediction.
p_y_x = model(X, W)
p_y_x_max = tt.argmax(p_y_x, axis=1)
cost = tt.mean(tt.nnet.categorical_crossentropy(p_y_x, Y))
updates = optimize(cost, W)
1.) train = theano.function(inputs=[X, Y], outputs=cost,
updates=updates,
allow_input_downcast=True,
mode='DebugMode', profile=True)
2.) predict = theano.function(inputs=[X], outputs=p_y_x_max,
updates=updates,
allow_input_downcast=True,
mode='DebugMode')
Everything seems to be working fine until the 1.) train function and in a
reasonable amount of time since I've not introduced the main training loop
yet.
After that everything halts for eternity in the 2.) predict function,
without any debug message or bug printed out.
I've tried pretty much everything and all the previous steps up until step
2.) the predict function, work fine and produce the expected outcomes.
I really don't know why is this happening anyone has any ideas or similar
experience in the past?
Thank you in advance!
--
---
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.
