how to initialize the neural network to a set of weights ???

I created my NN with patternet ??

Answer 1

[Greg Heath]

Unlike the older nets (e.g., newfit, newpr, newff,...), you cannot assign weights to the newer networks (e.g., fitnet, patternnet, feedforwardnet,...) unless the net is configured.

There are two ways to configure the net before manually assigning your own initial weights. Both will assign initial weights that you can overwrite:

1. help/doc configure.

net = configure(net, x, t );

2. Train the net for 1 epoch

net.trainParam.epochs= 1.

net = train(net,x,t);

Hope this helps.

Thank you for formally accepting my answer.

Greg

[Mariem Harmassi]

ok i will try to cinfigure the net before training cauz the second solution is not a good one i need to train the net according to a specifical set of weignts .

[Samisam]

@Greg Heath can I do a manual weight initialization before I train the net???

I mean if I have an optimal weight from a spesific algorithm and I want to create a NN to test data using these weights is there any way to do this without training the net again??

Answer 2

[Greg Heath]

net = patternet;

will default to H = 10 hidden nodes. For other values use

net = patternnet(H);

If

size(input) = [I N ]

size(target) = [O N ]

the node topology is I-H-O.

For a manual weight initialization, first configure the net:

net = configure(net,x,t);

For a random weight initialization, initialize the random number generator. Then generate and assign the weights:

rng(0)

IW = 0.01*randn(H,I);

b1 = 0.01*randn(H,1);

LW = 0.01*randn(O,H);

b2 = 0.01*randn(O,1);

then

net.IW{1,1} = IW;

net.b{1,1} = b1;

net.LW{2,1} = LW;

net.b{2,1} = b2;

Hope this helps.

Thank you for formally accepting my answer.

Greg

[Mariem Harmassi]

i tried to initialize the IW but matlag genrate error because the net.IW is an empty matrix and the same for the second layer net.LW {2,1} net=patternnet(10); net.IW

ans =

    [10x0 double]
    []

>> net.LW

ans =

               []    []
    [0x10 double]    []

suppose taht i have 4 input nodes i tried this : net.IW{1}=rand(10,4); Error using network/subsasgn>network_subsasgn (line 536) net.IW{1,1} must be a 10-by-0 matrix.

Error in network/subsasgn (line 13) net = network_subsasgn(net,subscripts,v,netname); which means the network does not accept initialization of IW let try whith the layer suppose that i want to set the number of output nodes to 1 : net.LW{2,1}=rand(1,10); Error using network/subsasgn>network_subsasgn (line 536) net.LW{2,1} must be a 0-by-10 matrix.

Error in network/subsasgn (line 13) net = network_subsasgn(net,subscripts,v,netname); what s the hell ?

[Greg Heath]

As I said before, you have to configure the net before manually assigning weights.

help configure

doc configure

[Heather Zhang]

Thank you Greg. "configure" works really well.

[LukasJ]

Dear Greg Heath,

unfortunately configuring the net doesn't do the trick for me:

https://de.mathworks.com/matlabcentral/answers/589957-neural-network-not-updating-weights-after-configuration#answer_536465?s_tid=prof_contriblnk

I tried setting the inital weights manually e.g.

net.iw{1,1} = zeros(...

and via

net.initFcn = 'initlay';
net.layers{1,1}.initFcn = 'initwb';
net.layers{2,1}.initFcn = 'initwb';
net.InputWeights{1,1}.initFcn = 'midpoint';
net.LayerWeights{2,1}.initFcn = 'midpoint';

initFcn to call for midpoint initialization. The first won't update any weights after training, the former won't do anything (still random weights when I check before training, training results after fixed epochs are not comparable).

Your help would be very much appreciated!

Best regards,

Lukas

Answer 3

[renz]

net.IW{1} = %input weights
net.LW{2} = %layer weights
% biases:
net.b{1}  = 
net.b{2}  =

Answer 4

[Sara Perez]

You can specify your own function for the initialization of the weights with 'WeightsInitializer' in convolution2dLayer.

layer = convolution2dLayer(filterSize,numFilters, ...

'WeightsInitializer', @(sz) rand(sz) * 0.0001, ...

'BiasInitializer', @(sz) rand(sz) * 0.0001)

info here:

https://es.mathworks.com/help/deeplearning/ref/nnet.cnn.layer.convolution2dlayer.html