Shulou(Shulou.com)06/01 Report--

This article introduces how to use the neural network pattern recognition tool nprtool, the content is very detailed, interested friends can refer to it, I hope it can help you.

Pattern recognition is also called pattern classification. Generalized pattern recognition includes supervised recognition and unsupervised recognition, which correspond to the training process with target data and without target data respectively. The training data of the former belongs to unknown class, while the training data of the latter belongs to known class. The pattern recognition referred to in neural network pattern recognition tools is mainly the latter, i.e. supervised classification. For unsupervised classification problems, neural network clustering tools can be used to solve them.

In a pattern recognition problem, where input data is classified into a predetermined number of classes, and each input sample is ultimately classified into a predetermined class, neural network pattern recognition tools can be used to collect data, create and train neural networks, and evaluate the network using mean square error (MSE) and confusion matrices. The system uses a two-layer (excluding input and output layers) feedforward network

Enter nprtool on the command line to open the neural network pattern recognition tool

Construct test data

x=[0.1,4.2;-0.25,2.8;3,1.1;-0.9,1.2;-1.2,1;3.4,1;-2.5,-1.5;3,3.2;...

-2.5,2.7;3.1,-3.2;4,-1.2;3.9,-1;4,3;-4,3.5]';

y=[1,1,1,1,1,2,1,2,1,2,2,2,2,1];

y0=ind2vec(y);

Click Next in the clustering tool to enter the data selection interface. Pattern recognition requires target data, so here you need to specify input and target samples.

The target sample here needs to be represented in vector form

Click Next to enter the Validation and Test Data interface and divide the dataset into training samples, validation samples and test samples.

Click Next to enter the network structure interface and set the number of hidden layer neurons

Click Next to enter the network training interface, click Train to start training, the default iteration number is 1000 times

When training is complete, the mean square error and misclassification rate for the training, validation, and test samples are displayed.

Misclassification rate refers to the proportion of data in a sample that is incorrectly classified into another class

Click Next to enter the network evaluation interface and generate test data

xx=-4.4:.4:4.5;

N=length(xx);

for i=1:N

for j=1:N

xt(1,(i-1)*N+j)=xx(i);

xt(2,(i-1)*N+j)=xx(j);

end

end

yt=ones(1,529);

yt(1)=2;

yt=ind2vec(yt);

Select the test data and click Test Network for simulation test

test results

Click Next to enter the result saving interface, you can save the network and variables, or export the network as MATLAB script or Simulink model

Click Finish to complete classification

How to use the neural network pattern recognition tool nprtool to share here, I hope the above content can be of some help to everyone, you can learn more knowledge. If you think the article is good, you can share it so that more people can see it.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.