In DeepnetworkDesigner, is it possible to make output as a matrix?

I want to predict output as a matrix in LSTM(regression).

for example,

x1 = 4, x2 = 5, x3 = 7 and y = [1 0 0 1]

Thanks.

Answer 1

[Muskan]

Hi Sierra,

As per my understanding of the question, you can follow the following steps to make output as a matrix in DeepNetworkDesigner:

i) Create a LSTM network, and configure it as per your requirements.

ii) Add an output layer.

iii) In the "Output" tab, set the number of outputs to be equal to the number of rows and columns of your desired output matrix and select "Regression" as the output type.

iv) Connect output layer to the last LSTM layer, and train your network.

v) After completing the training, the output will be a matrix with the dimensions as the output layer.

I hope the above information helps resolve your query.

Thanks.