MATLAB: How to include the model created using TreeBagger in a user-defined function to predict responses using the model

callfunctionincludemodelStatistics and Machine Learning Toolboxtreebagger

How to include the model created using TreeBagger in a user-defined function to predict responses using the model?

I have used TreeBagger to create a random forest model called "Mdl". Now, I want to include "Mdl" in a separate user-defined function.

The function takes some input, "inputX" and performs a predict(Mdl,inputX).

The result from "predict" is the output for the function.

How can the function call the "Mdl"?

Best Answer

To include the model, "Mdl", in the function that you are interested in, you can use any of the following solutions:

1. You can pass the model as an input argument to the function. For example, if you would like to pass the model to the function, "foo", you can define the function to have the model as an input argument, along with any other inputs, like:

function output = foo( Mdl, inputX )
  output = predict(Mdl,inputX);
end

Now, you can call this function from the script or the function that creates a random forest model called "Mdl" using the TreeBagger class, as follows:

>> result = foo( Mdl, inputX);

Where, "inputX" is the predictor input that you would like to predict the responses for. Please ensure that the defined function, "foo", is on the MATLAB path.

2. You can save the created random forest model, "Mdl", in a MAT-file by executing the following command in the Command Window:

>> save('model.mat','Mdl');

Then, you can load the MAT-file in the function that you are interested in, by including the following command in the function definition:

>> load model.mat

Please ensure that the saved MAT-file (model.mat) is on the MATLAB path.

Related Solutions

MATLAB: How to choose the best kernel for a Gaussian process regression, possibly using “bayesopt” function? How to predict unseen data with the resulting model

There are two ways you could do this. One of them involves "bayesopt" function.

(1) The easiest way is to use the "fitrgp" function. There is an example of how to optimize a fitrgp hyperparameter (Sigma) on the "fitrgp" function Documentation page:

https://www.mathworks.com/help/stats/fitrgp.html?searchHighlight=fitrgp

To optimize ‘KernelFunction’, using default values for all other hyperparameters, you could do the following (using data from the documentation example):

load(fullfile(matlabroot,'examples','stats','gprdata2.mat'))
model1 = fitrgp(x,y,'OptimizeHyperparameters',{'KernelFunction'});

Because fitrgp tends to be sensitive to the Sigma parameter, you will probably get a better result if you simultaneously optimize KernelFunction and Sigma. Also, searching categorical variables inherently requires more function evaluations, so you should probably run it longer:

model2 = fitrgp(x,y,'OptimizeHyperparameters',{'KernelFunction','Sigma'},...
    'HyperparameterOptimizationOptions', struct('MaxObjectiveEvaluations',100));

The "fitrgp" function documentation page lists the hyperparameters that are eligible for optimization, and what values are searched:

https://www.mathworks.com/help/stats/fitrgp.html#input_argument_namevalue_OptimizeHyperparameters.

You can use this model to make predictions on unseen data directly with the "predict" function:

model2 = fitrgp(x,y,'OptimizeHyperparameters',{'KernelFunction','Sigma'},...
    'HyperparameterOptimizationOptions', struct('MaxObjectiveEvaluations',100));
% Predict on unseen data

xNew = 0.2
ypred = predict(model2, xNew)

(2) A second approach would be to use "bayesopt" function directly, defining your own objective function. Here, the objective function is the 5-fold crossvalidation loss of the model.

load(fullfile(matlabroot,'examples','stats','gprdata2.mat')) 
% Define kernel variable
kernel = optimizableVariable('KernelFunction',{'exponential','squaredexponential','matern32','matern52',...
'rationalquadratic','ardexponential','ardsquaredexponential','ardmatern32','ardmatern52','ardrationalquadratic'},...
'Type','categorical') 
% Call bayesopt, capturing x and y in the objective function

bo = bayesopt(@(T)objFcn(T,x,y), kernel)
% Define an objective function
function Loss = objFcn(Vars, x, y)
m = fitrgp(x, y, 'KernelFunction', char(Vars.KernelFunction), 'KFold', 5);Loss = kfoldLoss(m);
end

(2.1) To optimize both KernelFunction and Sigma using "bayesopt" function, you can do the following:

% Define 2 variables
kernel = optimizableVariable('KernelFunction',{'exponential','squaredexponential','matern32','matern52',...
'rationalquadratic','ardexponential','ardsquaredexponential','ardmatern32','ardmatern52','ardrationalquadratic'},...
'Type','categorical')
sigma = optimizableVariable('Sigma',[1e-4,10],'Transform','log')
 
% Call bayesopt, capturing x and y in the objective function
bo = bayesopt(@(T)objFcn(T,x,y), [sigma, kernel], 'MaxObjectiveEvaluations', 100)
 
function Loss = objFcn(Vars, x, y)
m = fitrgp(x, y, 'KernelFunction', char(Vars.KernelFunction), ...
                 'Sigma', Vars.Sigma, 'ConstantSigma', true,...
                 'KFold', 5);
Loss = kfoldLoss(m);
end

To get a model that can make predictions, you need to fit a model without passing the ‘kfold’ argument, but keeping the optimal kernel function obtained from "bayesopt" function.

kernel= bo.XAtMinObjective.KernelFunction
model = fitrgp(x, y, 'KernelFunction', char(kernel));
% Predict on unseen data
xNew = 0.2
ypred = predict(model2, xNew)

MATLAB: Am I getting an error “Time series process Y and specification structure are non-conformable.” when using vgxvarx with data of certain dimensions

There is no actual limitation to the number of dimensions "vgvarx" accepts.

You are getting this error because of the way you are calling the function "vgxvarx". According to the documentation, you can call this function in two ways:

1. With exactly two arguments, "Spec" and "Y". In your code, these are the variables "Mdl" and "EstY".

2. With four or more arguments, "Spec", "Y", "X", "Y0", and additional name/value pair arguments.

In your code, you use a mix of these two syntaxes, by providing variables "Mdl" and "EstY" and the name/value pair 'IgnoreMA', 'yes':

>> vgxvarx(Mdl, EstY, 'IgnoreMA','yes');

Using this syntax, the word 'yes' is interpreted as the argument "Y0", a vector of length 3 (because of the three characters in the word). That is why your code does not return an error when your data has 3 columns, but returns an error for any other number of columns. However, the results of the function will not be the ones you expect, even if your data indeed has 3 columns.

To work around this error, please use the following syntax instead:

>> vgxvarx(Mdl, EstY, [ ], [ ], 'IgnoreMA','yes');

This way, you are calling the function and setting "X" and "Y0" to the empty vector [ ], which enables you to then set the 'IgnoreMA' argument to 'yes'.

Please find the documentation page in question in the following link:

http://uk.mathworks.com/help/releases/R2016b/econ/vgxvarx.html

Best Answer

Related Solutions

MATLAB: How to choose the best kernel for a Gaussian process regression, possibly using “bayesopt” function? How to predict unseen data with the resulting model

MATLAB: Am I getting an error “Time series process Y and specification structure are non-conformable.” when using vgxvarx with data of certain dimensions

Related Question