How to use the trained neural network to make predictions
Google AI Writing Project: Neural Network Pseudo-Original
How Artificial Neural Networks Can Predict the Next Value
The newff function establishes a BP neural network, historical data is used as samples, for example, the first n data are used as input, and the input nodes are n writing cats . The current data is taken as p, and the output node is 1. Hidden layer nodes are obtained by trial and error.
Through the train function of matlab, the trained BP neural network is obtained. Then the first n data of the current prediction point are used as input, and the output is the current prediction value.
How to use the neural network toolbox in matlab? And how to predict new data after the neural network is trained?
Use the neural network in weka to make predictions, there are ready-made ones, how to predict new data?
Weka seems to be able to predict only class attributes, that is, you can provide training set data with known class attributes and prediction sets with unknown class attributes.
After selecting the neural network algorithm under the Explorer, select the training test and supplied test set under the test option.
Help: Use a neural network to make a data prediction
The following code is the code for the BP neural network to predict the sales volume of 37-56 weeks: % x is the original sequence load sales volume.matdata=Cx=data';t=1:length(x);lag=2;fn=length( t);[f_out,iinput]=BP(x,lag,fn);%Forecast year or a certain time period t1=fn:fn+20;n=length(t1);t1=length(x)+1: length(x)+n;%The number of prediction steps is fnfn=length(t1); [f_out,iinput]=BP(x,lag,fn);P=vpa(f_out,5);[t1' P']% Draw the forecast figure figure(6), plot(t,x,'b*-'),hold onplot(t(end):t1(end),[iinput(end),f_out],'rp-'), grid onxlabel('weeks'), ylabel('sales');str=['BP neural network forecast',num2str(length(x)+1),'-',num2str(length(x)+20) ,'weekly sales'];title(str)str1=['1-',num2str(length(x)),'weekly sales'];str2=[num2str(length(x)+1), '-',num2str(length(x)+20),'weekly forecast sales'];legend(str1,str2) operation results.
How to use spss to classify data and predict probability according to neural network model
This prediction must have historical data, that is, you need to have a lot of historical churn data in the past and data of some influencing factors that may affect churn. To put it bluntly, it is still necessary to have historical data similar to independent variables and dependent variables, and then it is possible to build model predictions.
If you don't have historical data to support the establishment of a model, there is no way to make predictions.
How to build a bp neural network forecasting model
To establish a BP neural network prediction model, the following steps can be followed: 1. Provide raw data 2. Extraction and normalization of training data prediction data 3. BP network training 4. BP network prediction 5. Results analysis Now use a practical example, To predict the population of an area in 2015 and 2016.
It is known that the population of a certain area from 2009 to 2014 is 3583, 4150, 5062, 4628, 5270, and 53.4 million people respectively. Execute the BP_main program, and get [2015, 5128. Hehe 3946380615234375] [ 2016, 5100.579732564277946949005112695] The code and graphics are as follows.