.


:




:

































 

 

 

 


layerConnect, outputConnect, tartegtConnect).

, , - . , .

:

numImputs = 2 - ;

numLayers = 3 - ;

biasConnect = [1; 0; 0] - numLayers * 1;

imputConnect = [1 1; 0 0; 0 0] - numLayers * numImputs;

layerConnect = [0 0 0;1 0 0 0; 0 1 0] - numLayers * numLayers;

outputConnect = [0 0 1] - 1* numLayers;

tartegtConnect = [0 0 1] - 1 * numLayers.

:

 

1. ,

net = network (2, 3, [1; 0; 0], [1 1; 0 0; 0 0],[0 0 0; 1 0 0; 0 1 0], [0 0 1])

 

2. net .

3. ,

numOutputs = 1 - ;

numInputDelays = 0 -

numLayerDelays = 0 - .

, . , . . numInputDelays NumLayerDelays - .

4. , gensim(net) .

 

. 2.

 

NNT :

) Neural Network - x{1}, x{2}, y{1};

) Input1, x{1} Input2, x{2};

) y{1};

) Layer 1, Layer 2, Layer 3, x{1}, x{2], a{1}, a{2}, a{1}, a{2}, a{3}, , y{1};

) TDL - (Time Delay) Delays1, Delays2,..., 1, 2, 3, ;

) Weights - ; S × R, S - , R - , ; j i S × R, S - i, R - j, ;

) dotprod - , , . . ;

) mux - , , ;

) netsum - : , . .;

) hardlim, purelin . . - ;

) pd{1, 1}, pd{1, 2}, ad{2, 1},... - (d - delay);

) iz{1, 1}, iz{1, 2}, lz{2, 1}, lz{3, 2} - - ;

) bias - ;

) IW - : IW{i, j} - i j;

) LW - : LW{i, j} - i j.

5. NNT.

6.

P = [0 0; 0 0]

A = sim(net, P)

7. :

 

net.inputs{1}.range = [0 1];

net.inputs{2}.range = [0 1];

net.b{1} = - 0,25;

net.IW{1, 1} = [0.5];

net.IW{1, 2} = [0.5];

net.LW{2, 1} = [0.5];

net.LW{3, 2} = [0.5].

 

gensim(net) .

 


. 3.

 

8. , celldisp. .

ans{1} =

exampleInput: [0 1]

name: 'Input'

processFcns: {}

processParams: {}

processSettings: {1x0 cell}

ans{2} =

exampleInput: [0 1]

name: 'Input'

processFcns: {}

processParams: {}

processSettings: {1x0 cell}

 

9. , . . ,

PG = [0.5 1; 1 0.5];

PS= {[0.5 1] [1 0.5]};

PG1 = sim(net, PG);

PS1 = sim(net, PS)

, .

10. 2- 3- , :

net.inputWeights{1, 1}.delays = [0 1];

net.inputWeights{1, 2}.delays = [0 1];

net.layerWeights{3, 2}.delays = [0 1 2].

11. :

net.IW{1, 1} = [0.5 0.5];

net.IW{1, 2} = [0.5 0.25];

net.LW{3, 2} = [0.5 0.25 1].

12. , :

AG = sim(net, PG);

AS= sim(net, PS).

, ( ) .

 

. 4.

 

13. , celldisp.

ans{1} =

exampleInput: [0 0,5]

name: 'Input'

processFcns: {0,5 0,5}

processParams: {0,5 0,25}

processSettings: {1x0 cell}

ans{2} =

exampleInput: [0 0,5]

name: 'Input'

processFcns: {0,5 0,5}

processParams: {0,5 0,25 1}

processSettings: {1x0 cell}

 

14. - .

2.

asgnet, network . net asgnet.

.

 

.

 

1. asgnet network

asgnet = network();

 

 
 

2.

asgnet.numInputs = 2

 
 

asgnet.numLayers = 3

 
 

asgnet.biasConnect = [1; 0; 0]

asgnet.inputConnect = [1 1; 0 0; 0 0]

asgnet.layerConnect = [0 0 0; 1 0 0; 0 1 0]

asgnet.outputConnect = [0 0 1]

 
 

. 5. asgnet

P1 = [0 0; 0 0]

A1 = sim(asgnet, P1)

asgnet.inputs{1}.range = [0 1];

asgnet.inputs{2}.range = [0 1];

PG1 = [0.5 1; 1 0.5];

PS1= {[0.5 1]; [1 0.5]};

AG1 = sim(asgnet, PG1);

AS1= sim(asgnet, PS).

asgnet.b{1} = - 0,25;

asgnet.inputWeights{1, 1}.delays = [0 1];

asgnet.inputWeights{1, 2}.delays = [0 1];

asgnet.layerWeights{3, 2}.delays = [0 1 2].

asgnet.IW{1, 1} = [0.5 0.5];

asgnet.IW{1, 2} = [0.5 0.25];

asgnet.LW{2, 1} = [0.5];

asgnet.LW{3, 2} = [0.5 0.25 1]

 

 
 

. 6. asgnet

. 7. net

.

 

3. Simulink MATLAB, asgnet, , net Report Generator.

Simulink, .

4. network , , :

) , ;

task1 = network (3, 1, [1], [1 1 1],[1], [1], [1])

 

. 8.

) ; - , ; ; - ;

 

task2 = network (3, 3, [1; 1; 1], [1 0 0; 1 0 0; 1 0 0],[1 0 0; 1 0 0; 1 0 0], [1 0 0], [1 0 0])

 

. 9.

) , ; - ; - ; ;

 

task3 = network (3, 3, [1; 0; 0], [1 1 1; 1 1 1; 1 1 1],[1 1 1; 1 1 1; 1 1 1], [1 0 0], [1 0 0])

. 11. , ;

 

) ; - , ; ; , ;

 

task3 = network (3, 3, [1; 0; 0], [1 1 1; 1 1 1; 1 1 1],[1 1 1; 1 1 1; 1 1 1], [1 0 0], [1 0 0])

 

 

. 12.

 

task3.inputWeights{1, 1}.delays = [0 1];

task3.inputWeights{1, 2}.delays = [0 1];

task3.inputWeights{1, 3}.delays = [0 1];

task3.inputWeights{2, 1}.delays = [0 1];

task3.inputWeights{2, 2}.delays = [0 1];

task3.inputWeights{2, 3}.delays = [0 1];

task3.inputWeights{3, 1}.delays = [0 1 2];

task3.inputWeights{3, 2}.delays = [2 1 2];

 

task3.IW{1, 1} = [0.5 0.8];

task3.IW{1, 2} = [0.5 0.2];

task3.IW{1, 3} = [0.2 1];

task3.IW{2, 1} = [0.2 1];

task3.IW{2, 2} = [0.5 0.2];

task3.IW{2, 3} = [0.5 1];

task3.LW{3, 1} = [0.5 0.2 0.1];

task3.LW{3, 2} = [0.1 0.2 0.1];

task3.LW{3, 3} = [0 0.2 1];

 

. 13.

 

) ; ; ; .

P1 = [0; 0; 0; 0; 0; 0; 0; 0; 0; 0];

P2 = {

[1 1 1 1 1 1 1 1 1 1]

[ 1 1 1 1 1 1 1 1 1 1]

[ 1 1 1 1 1 1 1 1 1 1]

[ 1 1 1 1 1 1 1 1 1 1]

[ 1 1 1 1 1 1 1 1 1 1]

[ 1 1 1 1 1 1 1 1 1 1]

[ 1 1 1 1 1 1 1 1 1 1]

[ 1 1 1 1 1 1 1 1 1 1]

[ 1 1 1 1 1 1 1 1 1 1]

[ 1 1 1 1 1 1 1 1 1 1] };

P3 = [1 1 1 1 1 1 1 1 1 1];

task5 = network(10, 10, P1, P2, P2, P3, P3)

 

, task5.mdl.

 

 

, .

:

, ;

, .

.

, , , .

MatLab , , .

, .
.



<== | ==>
. | .
:


: 2017-02-11; !; : 244 |


:

:

- , , .
==> ...

1763 - | 1548 -


© 2015-2024 lektsii.org - -

: 0.055 .