How to find sensitivity of feedforward net in matlab?

I am trying to find the sensitivity of a feedforward neural network. For example, to find the sensitivity of the following feedforwardnet

[x , t] = house_dataset;

net = feedforwardnet(10);

net = train(net, x, t);

 I used the formula 

 s = b2 + Lw * tanh(b1 + Iw * x)

But it has the following error

Error using + Matrix dimensions must agree.

I am trying to find another solution. Please Help me on how to find sensitivity of a neural network Feedforward net.

[Greg Heath]

Sensitivity to what?

The above equation has the right form however, you have to make the dimensions of the biases compatible using the function repmat.

Use the command

 whos

to make sure dimensions are compatible.

Hope this helps.

Greg

Answer 1

[Greg Heath]

You forgot to take into account that the input is 13 dimensional

[Andualem alemu]

so, would you please help me on how to find the sensitivity of feedforwardnet for this 13 dimensional data?

[Greg Heath]

You didn't answer my question.