Simple denoising autoencoder for 1D data

14 次查看（过去 30 天）
vikakise vikakise
vikakise vikakise 2017-8-29
评论： Georgios Papageorgiou 2019-10-19
I'm trying to set up a simple denoising autoencoder with Matlab for 1D data. As currently there is no specialised input layer for 1D data the imageInputLayer() function has to be used:
function net = DenoisingAutoencoder(data)
[N, n] = size(data);
%setting up input
X = zeros([n 1 1 N]);
for i = 1:n
for j = 1:N
X(i, 1, 1, j) = data(j,i);
end
end
% noisy X : 1/10th of elements are set to 0
Xnoisy = X;
mask1 = (mod(randi(10, size(X)), 7) ~= 0);
Xnoisy = Xnoisy .* mask1;
layers = [imageInputLayer([n 1 1]) fullyConnectedLayer(n) regressionLayer()];
opts = trainingOptions('sgdm');
net = trainNetwork(X, Xnoisy, layers, opts);
However, the code fails with this error message:
The output size [1 1 n] of the last layer doesn't match the response size [ n 1 1].
Any thoughts on how should the input / layers should be reconfigured? If the fullyConnectedLayer is left out then the code runs fine, but obviously then I'm left without the hidden layer.
  1 个评论
Georgios Papageorgiou
Georgios Papageorgiou 2019-10-19
I assume N is the number of data and n is you data_size. I think if you make:
  1. X 1xnx1xN
  2. The input layer: imageInputLayer([1 n])
  3. X_noisy of dimension Nxn and finally,
  4. net = trainNetwork(Xnoisy, X, layers, opts);
it should work. Make sure at the end that your inpout of your Denoising Autoencoder is the noisy data and the desired output is your "clean" data. A similar version that I implemented in MATLAB works fine for me and the dimensions match usign the regressionLayer like this.

请先登录，再进行评论。

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息

标签

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by