我怎样才能在theano中获得一维卷积

Question

The only function I can find is for 2D convolutions described here ...

Is there any optimised 1D function ?

Answer 1

It looks as though this is in development . I've realised I can use the conv2d() function by specifying either width or height as 1...

For the function conv2d() , the parameter image_shape takes a list of length 4 containing:

([number_images,] height, width)

by setting height=1 or width=1 it forces it to a 1D convolution.

Answer 2

While I believe there's no conv1d in theano, Lasagne (a neural network library on top of theano) has several implementations of Conv1D layer. Some are based on conv2d function of theano with one of the dimensions equal to 1, some use single or multiple dot products. I would try all of them, may be a dot-product based ones will perform better than conv2d with width=1 .

https://github.com/Lasagne/Lasagne/blob/master/lasagne/theano_extensions/conv.py

Answer 3

Just to be a bit more specific, I found this to work nicely:

conv2d = T.signal.conv.conv2d

x = T.dmatrix()
y = T.dmatrix()
veclen = x.shape[1]

conv1d_expr = conv2d(x, y, image_shape=(1, veclen), border_mode='full')

conv1d = theano.function([x, y], outputs=conv1d_expr)

border_mode = 'full' is optional.