[theano-users] Intel MKL ERROR: Parameter 4 was incorrect on entry to DGELSD.

[S Nothing]

I am training a neural net and I'm running into the above error message.

Im using Lasagne to implement the neural net. It seems to throw this error
when I change my layer to return the square of some value e.g.,


class RandomLayer(MergeLayer):

def __init__(self, incomings, **kwargs):

super(RandomLayer, self).__init__(incomings, **kwargs)


def get_output_for(self, inputs, **kwargs):
x = inputs[0]

return x**2