MATLAB: Help with defining neural networks

errorMATLAB

Hi,

When I define the following layers:

decodingLayers_2=[ ...
    createUpsampleTransponseConvLayer(2,8), ...
    reluLayer, ...
    convolution2dLayer(3,1,'Padding','same'), ...
    clippedReluLayer(1.0), ...
    %clippedReluLayer(1.0), ...
    regressionLayer];

Matlab gives the following error:

Error using vertcat

Dimensions of arrays being concatenated are not consistent.

But when I modify the above just a bit as follows:

decodingLayers_2=[ ...
    createUpsampleTransponseConvLayer(2,8), ...
    reluLayer, ...
    convolution2dLayer(3,1,'Padding','same'), ...
    clippedReluLayer(1.0), ...
    clippedReluLayer(1.0), ...
    regressionLayer];

The error disappears.

Please let me know why is this happening and why do I need to put 2 relu layers at the end when I just need 1.

thanks

Best Answer

This error message may be due to the presence of a comment inside the “decodingLayers_2” array. This may interfere with the line continuation operator and eventually lead to an error during run-time.

One way to avoid this is to avoid commenting inside the array definition. So, the updated code can be as follows

decodingLayers_2 = [ ...
    createUpsampleTransponseConvLayer(2,8), ...
    reluLayer, ...
    convolution2dLayer(3,1,'Padding','same'), ...
    clippedReluLayer(1.0), ...
    regressionLayer];

Related Solutions

MATLAB: Error while transferring weights of a trained CNN network to an empty CNN network

If the function createUpsampleTransponseConvLayer is the helper function from the example Prepare Datastore for Image-to-Image Regression then change the 'NumChannels' Name-Value Pair Argument to 'auto' or don't mention it in the transposedConv2dLayer function.

% helper function from the example Prepare Datastore for Image-to-Image Regression
function out = createUpsampleTransponseConvLayer(factor,numFilters)
filterSize = 2*factor - mod(factor,2); 
cropping = (factor-mod(factor,2))/2;
numChannels = 1;
out = transposedConv2dLayer(filterSize,numFilters, ... 
    'NumChannels',numChannels,'Stride',factor,'Cropping',cropping);
end

Since the layer is defined with 'NumChannels' (number of channels of the input to this transposedConv2dLayer) as 1 hence it can accept wieghts of size "filterSize x filterSize x numFilters x numChannels" which is 4x4x8x1 in this case.

Change the function as follows:

function out = createUpsampleTransponseConvLayer(factor,numFilters)
filterSize = 2*factor - mod(factor,2); 
cropping = (factor-mod(factor,2))/2;
out = transposedConv2dLayer(filterSize,numFilters, ... 
    'Stride',factor,'Cropping',cropping);
end

and then define the layers.

MATLAB: Deep learning_alexnet

You have specified the image as single channel in your code. just change it to 3 channels (RGB).

 imageInputLayer([imageSize 1],....

 imageInputLayer([imageSize 3],....

Best Answer

Related Solutions

MATLAB: Error while transferring weights of a trained CNN network to an empty CNN network

MATLAB: Deep learning_alexnet

Related Question