CompactRegressionChainEnsemble

Compact multiresponse regression model

Since R2024b

Description

CompactRegressionChainEnsemble is a compact version of a RegressionChainEnsemble model object. The compact model does not include the data used for training the model.

Creation

Create a full RegressionChainEnsemble object and then compact it by using the compact object function.

Properties

Chain Ensemble Properties

ChainOrders — Order of response variables in regression chains

Read-only: positive integer matrix

This property is read-only.

Order of the response variables in the regression chains, specified as a positive integer matrix. Row i indicates the order of the response variables in regression chain i.

Data Types: double

Learners — Compact regression models trained as part of regression chains

Read-only: cell array of regression model objects

This property is read-only.

Compact regression models trained as part of the regression chains, specified as a cell array of regression model objects. Each row of Learners corresponds to one regression chain.

This table lists the possible compact regression models.

Regression Model Type	Model Object
Bagged or boosted ensemble of trees	CompactRegressionEnsemble
General additive model (GAM)	CompactRegressionGAM
Gaussian process regression (GPR)	CompactRegressionGP
Kernel model	RegressionKernel
Linear model	RegressionLinear
Support vector machine (SVM)	CompactRegressionSVM
Decision tree	CompactRegressionTree

Data Types: cell

NumChains — Number of regression chains

Read-only: positive integer scalar

This property is read-only.

Number of regression chains in the chain ensemble, specified as a positive integer scalar. NumChains indicates the number of rows in ChainOrders and Learners.

Data Types: double

Data Properties

CategoricalPredictors — Categorical predictor indices

Read-only: positive integer vector | []

This property is read-only.

Categorical predictor indices, specified as a positive integer vector. Each index value in CategoricalPredictors indicates that the corresponding predictor listed in PredictorNames is categorical. If none of the predictors are categorical, then this property is empty ([]).

Data Types: double

NumPredictors — Number of predictor variables

Read-only: positive integer scalar

This property is read-only.

Number of predictor variables, specified as a positive integer scalar. NumPredictors does not include response variables that are used as predictors by some models in Learners.

To see all the predictors used by a specific compact regression model in Learners, use the properties of the compact regression model. For an example, see Specify Multiresponse Regression Model Properties.

Data Types: double

NumResponses — Number of response variables

Read-only: positive integer scalar

This property is read-only.

Number of response variables, specified as a positive integer scalar.

Data Types: double

PredictorNames — Predictor variable names

Read-only: cell array of character vectors

This property is read-only.

Predictor variable names, specified as a cell array of character vectors. The order of the elements in PredictorNames corresponds to the order of the predictors in the data used to train the model.

Data Types: cell

ResponseName — Response variable names

Read-only: string array

This property is read-only.

Response variable names, specified as a string array. The order of the elements in ResponseName corresponds to the order of the response variables in the data used to train the model.

Data Types: string

Object Functions

loss	Loss for multiresponse regression model
predict	Predict responses using multiresponse regression model

Examples

Reduce Size of Multiresponse Regression Model

Reduce the size of a full multiresponse regression model by removing the training data from the model. You can use a compact model to improve memory efficiency.

Load the carbig data set, which contains measurements of cars made in the 1970s and early 1980s. Create a table containing the predictor variables Displacement, Horsepower, and so on, as well as the response variables Acceleration and MPG. Display the first eight rows of the table.

load carbig
cars = table(Displacement,Horsepower,Model_Year, ...
    Origin,Weight,Acceleration,MPG);
head(cars)

    Displacement    Horsepower    Model_Year    Origin     Weight    Acceleration    MPG
    ____________    __________    __________    _______    ______    ____________    ___

        307            130            70        USA         3504           12        18 
        350            165            70        USA         3693         11.5        15 
        318            150            70        USA         3436           11        18 
        304            150            70        USA         3433           12        16 
        302            140            70        USA         3449         10.5        17 
        429            198            70        USA         4341           10        15 
        454            220            70        USA         4354            9        14 
        440            215            70        USA         4312          8.5        14 

Categorize the cars based on whether they were made in the USA.

cars.Origin = categorical(cellstr(cars.Origin));
cars.Origin = mergecats(cars.Origin,["France","Japan",...
    "Germany","Sweden","Italy","England"],"NotUSA");

Remove observations with missing values.

cars = rmmissing(cars);

Train a multiresponse regression model by passing the cars data to the fitrchains function. Use regression chains composed of regression SVM models with standardized numeric predictors.

Mdl = fitrchains(cars,["Acceleration","MPG"], ...
    Learner=templateSVM(Standardize=true))

Mdl = 
  RegressionChainEnsemble
           PredictorNames: {'Displacement'  'Horsepower'  'Model_Year'  'Origin'  'Weight'}
             ResponseName: ["Acceleration"    "MPG"]
    CategoricalPredictors: 4
                NumChains: 2
            LearnedChains: {2×2 cell}
          NumObservations: 392


  Properties, Methods

Mdl is a trained RegressionChainEnsemble model object. The model contains information about the training data set, such as the training data properties X and Y.

Reduce the size of the model by using the compact object function.

compactMdl = compact(Mdl)

compactMdl = 
  CompactRegressionChainEnsemble
           PredictorNames: {'Displacement'  'Horsepower'  'Model_Year'  'Origin'  'Weight'}
             ResponseName: ["Acceleration"    "MPG"]
    CategoricalPredictors: 4
                NumChains: 2
            LearnedChains: {2×2 cell}


  Properties, Methods

compactMdl is a CompactRegressionChainEnsemble model object. compactMdl contains fewer properties than the full model Mdl.

Display the amount of memory used by each model.

whos("Mdl","compactMdl")

  Name            Size             Bytes  Class                                                    Attributes

  Mdl             1x1             125951  RegressionChainEnsemble                                            
  compactMdl      1x1              95825  classreg.learning.regr.CompactRegressionChainEnsemble              

The full model is larger than the compact model.

Version History

Introduced in R2024b