Compact generalized linear regression model class
CompactGeneralizedLinearModel is a compact version of a full
generalized linear regression model object
GeneralizedLinearModel. Because a compact model does not store the input
data used to fit the model or information related to the fitting process, a
CompactGeneralizedLinearModel object consumes less memory than a
GeneralizedLinearModel object. You can use still use a compact
model to predict responses using new input data, but some
GeneralizedLinearModel object functions do not work with a compact
when you work with tall arrays, and returns
you work with in-memory tables and arrays.
Fit a generalized linear regression model to data and reduce the size of a full, fitted model by discarding the sample data and some information related to the fitting process.
largedata4reg data set, which contains 15,000 observations and 45 predictor variables.
Fit a generalized linear regression model to the data using the first 15 predictor variables.
mdl = fitglm(X(:,1:15),Y);
Compact the model.
compactMdl = compact(mdl);
The compact model discards the original sample data and some information related to the fitting process, so it uses less memory than the full model.
Compare the size of the full model
mdl and the compact model
vars = whos('compactMdl','mdl'); [vars(1).bytes,vars(2).bytes]
ans = 1×2 15518 4382502
The compact model consumes less memory than the full model.
Usage notes and limitations:
Code generation does not support categorical predictors. You cannot
supply training data in a table that contains a logical vector,
character array, categorical array, string array, or cell array of
character vectors. Also, you cannot use the
'CategoricalVars' name-value pair argument. To include categorical predictors
in a model, preprocess the categorical predictors by using
fitting the model.
Inverse fields of the
name-value pair argument cannot be anonymous functions. That is, you
cannot generate code using a generalized linear model that was created
using anonymous functions for links. Instead, define functions for link
For more information, see Introduction to Code Generation.