How do I see the weight and biases variables in neural networks, please show me on simple example code?

How do I see the weight and biases variables in neural networks, please show me on simple example code?

[Matthew Eicholtz]

It depends. Are you using the built-in Neural Network Toolbox or building one from scratch?

Answer 1

[Matthew Eicholtz]

You can find some code examples using the Neural Network Toolbox here.

Copying from there, one example deals with classifying the sex of crabs.

[x,t] = crab_dataset; %load the dataset
net = patternnet(10); %create the network
[net,tr] = train(net,x,t); %train the network

Now, you can find the weights and biases by looking at net .

w1 = net.IW{1} %the input-to-hidden layer weights
w2 = net.LW{2} %the hidden-to-output layer weights
b1 = net.b{1} %the input-to-hidden layer bias
b2 = net.b{2} %the hidden-to-output layer bias

Hope this helps.

[Radek]

Many thanks Matt

[Subhadarshini Jati]

this provides only one set of weights associated to a single input. what if i need all weights associated with all inputs?

Answer 2

[Ava harry]

How about havng 2 hidden layers? how can I extract bias and weights between them?

[Rui Guilhôto]

Having the same doubt.

Answer 3

[Gene Pennello]

https://www.mathworks.com/matlabcentral/answers/398531-how-to-manually-calculate-a-neural-network-output