Hi, Not all implementations of convolutions used in Theano can be modified. For instance, cuDNN ones are defined externally, and we do not have the source code.
In your case, you could probably copy / edit the CorrMM version, and add the normalization of patches before the call to im2col. You may actually have to reverse the role of image and kernel for that. Also, you would have to redefine the gradients as well. On Tue, Dec 20, 2016, Eric Huang wrote: > > > I wanna to normalize (substract mean, divide standard deviation) input in > each patch during convolution. > > > For example, > Input: (1,224, 224) > kernel: (64,5,5) > stride: 1 > During the calculation of the first feature map, I'd like to do the > following operations for each position: > feaMap[0, 0, 0] = conv( (Input[0, 0:5, 0:5] - mean)/std, kernel[0, :, :] ) > feaMap[0, 0, 1] = conv( (Input[0, 0:5, 1:6] - mean)/std, kernel[0, :, :] ) > feaMap[0, 0, 2] = conv( (Input[0, 0:5, 2:7] - mean)/std, kernel[0, :, :] ) > > ....... > where mean is a fix matrix of shape (1, 5, 5), std is a fix matrix of shape > (1, 5, 5). > > It is great cost if I write my own code to extract each patch, do > normalization then conduct convolution. So I just want to edit the conv > function to add "substract mean", "divide standard deviation" operation. > But it seems difficult to edit the conv function to achieve my goal. > > How can I achieve it ? Any idea? > > -- > > --- > 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. -- Pascal -- --- 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.
