Skip to content

Get Coefficients of a Model

get_model_coef-libml.Rd

Extract the coefficients for an arbitrary model.

Usage

# S3 method for class 'glm'
get_model_coef(model, ...)

# S3 method for class 'lm'
get_model_coef(model, ...)

# S3 method for class 'lda'
get_model_coef(model, ...)

# S3 method for class 'libml_nb'
get_model_coef(model, ...)

# S3 method for class 'naiveBayes'
get_model_coef(model, ...)

# S3 method for class 'kknn'
get_model_coef(model, ...)

# S3 method for class 'randomForest'
get_model_coef(model, ...)

# S3 method for class 'gbm'
get_model_coef(model, ...)

# S3 method for class 'glmnet'
get_model_coef(model, lambda = NULL, ...)

# S3 method for class 'cv.glmnet'
get_model_coef(model, lambda = NULL, ...)

# S3 method for class 'train'
get_model_coef(model, lambda = NULL, ...)

# S3 method for class 'svm'
get_model_coef(model, ...)

Arguments

model

A model object, currently one of:

#>  [1] get_model_coef.cv.glmnet
#>  [2] get_model_coef.default*
#>  [3] get_model_coef.gbm
#>  [4] get_model_coef.glm
#>  [5] get_model_coef.glmnet
#>  [6] get_model_coef.kknn
#>  [7] get_model_coef.lda
#>  [8] get_model_coef.libml_nb
#>  [9] get_model_coef.lm
#> [10] get_model_coef.naiveBayes
#> [11] get_model_coef.randomForest
#> [12] get_model_coef.svm
#> [13] get_model_coef.train
#> see '?methods' for accessing help and source code

...

Additional parameters for extensibility.

lambda

The value of the penalty parameter lambda which can either be NULL or numeric. If NULL, the default value depends on the underlying class:

glmnet

The first value of lambda.

cv.glmnet

The lambda where the cross validated error is minimized.

train

The optimal value of lambda. When given a numeric value of lambda, the closest value of lambda within the model will be used.

Value

A named numeric vector of the coefficients of the model. If the model is non-linear (e.g. random forest), NULL.

Functions

  • get_model_coef(glm): S3 method for glm models.

  • get_model_coef(lm): S3 method for lm models.

  • get_model_coef(lda): S3 method for lda models.

  • get_model_coef(libml_nb): S3 method for libml_nb models.

  • get_model_coef(naiveBayes): S3 method for naiveBayes models.

  • get_model_coef(kknn): S3 method for kknn models.

  • get_model_coef(randomForest): S3 method for randomForest models.

  • get_model_coef(gbm): S3 method for gbm models.

  • get_model_coef(glmnet): S3 method for glmnet models.

  • get_model_coef(cv.glmnet): S3 method for cv.glmnet models.

  • get_model_coef(train): S3 method for train models.

  • get_model_coef(svm): S3 method for SVM models. If the model kernel = linear, coefficients are returned. Otherwise, NULL.

See also

coef()

Examples

# set up training and test data:
iris2 <- droplevels(iris[iris$Species != "setosa", ])

# Logistic Regression
stats::glm(Species ~ ., data = iris2, family = "binomial") |>
  get_model_coef()
#>  (Intercept) Sepal.Length  Sepal.Width Petal.Length  Petal.Width 
#>   -42.637804    -2.465220    -6.680887     9.429385    18.286137