model.interface-class: (Internal) An interface for model.adapter and...
In Marchen/model.adapter: Abstraction Layer for Statistical/Machine Learning Models
model.interface-class R Documentation
(Internal) An interface for model.adapter and statistical/machine learning models.
Description
This class provides unified interface for calls/objects of
statistical/machine learning models.
model.adapter-class{model.adapter} internally using object
of this class to get information from model calls/objects.
Methods
get.family(x, type = c("character", "family"), envir)
Get family from the model.
xcall or model object.
type = c("character", "family")-
a character literal specifying the type of data returned.
If "character", this returns character vector of family name.
If "family", this returns family object.
For some family specific for certain model (e.g., categorical
family of MCMCglmm), their family object is not implimented.
For such family, this method raise stop error.
envirenvironment where x is evaluated.
get.call(x)
This method returns call by which the object is made. If call is not
available, this returns NULL. To distinguish the returned value of NULL
is intended action or not, inherited classes are encouraged to
inherit this method to explicitly return NULL if x does not have call.
xcall or model object.
get.data(x, envir, package = "", ...)
Get a data.frame containing the data used for modeling.
If data is not available this method returns NULL.
x-
a model object/call from which data is extracted.
enviran environment in which call is evaluated.
get.formula(x, envir, package = "")
Extract formula from model object/call.
If couldn't retrieve formula from x, this method returns NULL.
x-
a model object/call from which formula is extracted.
envir-
an environment in which call in x is evaluated.
package-
name of the package having the modeling function.
expand.formula(f, d, specials = NULL, package = NULL)
Expand . in formula.
fa formula to expand.
da data.frame used to expand . in formula.
specials = NULL-
special characterss passed to
terms.formula.
package = NULL-
a character literal of package name having the model function.
predict(object, newdata, type, random, ...)
Calculate predictions.
objecta model object used for prediction.
newdata-
a data.frame containing data used for prediction.
type-
the type of prediciton. This should be a type specific for
each modeling functions.
random = ~0-
the random effect to use.
Tentatively, ~0 means don't use random effects.
...other variables passed to predict methods.
get.link(x, envir)
Get link function. If the model does not have link function, this
function returns identity function.
x-
an object of statistical model or a call of model function.
envir-
an environment where call in x is evaluated.
get.linkinv(x, envir)
Get inverse function of link function. If the model does not have
link function, this function returns identity
function.
x-
an object of statistical model or a call of model function.
envir-
an environment where call in x is evaluated.
get.model.type(x, envir, package = "", ...)
Return a character vector specifying model type
(regression or classification).
If the model is regression model, it returns 'regression'.
If the model is classification model, it returns 'classification'.
x-
an object of statistical model or a call of model function.
envir-
an environment where call in x is evaluated.
has.offset.argument(x, envir, package = "")
Check whether the model has offset argument.
x-
an object of statistical model or a call of model function.
envir-
an environment where call in x is evaluated.
package = NULL-
a character literal of package name having the model function.
has.offset.in.formula(x, envir, package = "")
Check whether the model has offset term(s) in formula.
x-
an object of statistical model or a call of model function.
envir-
an environment where formula and data in x is evaluated.
package = NULL-
a character literal of package name having the model function.
Active bindings
predict.typesa named character vector representing the conversion table of shared
type of prediction used for predict() method. Currently, this vector has
four elements named "response", "link", "prob" and "class" and
each element represents compatible type of character for the model.
Methods
Public methods
-
-
-
-
-
-
-
-
-
-
-
-
-
Method new()
Usage
model.interface.default.class$new(...)
Method get.family()
Usage
model.interface.default.class$get.family(
x,
type = c("character", "family"),
envir
)
Method get.call()
Usage
model.interface.default.class$get.call(x)
Method get.data()
Usage
model.interface.default.class$get.data(x, envir, package = "", ...)
Method get.formula()
Usage
model.interface.default.class$get.formula(x, envir, package = "")
Method expand.formula()
Usage
model.interface.default.class$expand.formula(
f,
d,
specials = NULL,
package = NULL
)
Method predict()
Usage
model.interface.default.class$predict(object, newdata, type, random, ...)
Method get.link()
Usage
model.interface.default.class$get.link(x, envir)
Method get.linkinv()
Usage
model.interface.default.class$get.linkinv(x, envir)
Method get.model.type()
Usage
model.interface.default.class$get.model.type(x, envir, package = "", ...)
Method get.offset.names()
Usage
model.interface.default.class$get.offset.names(x, envir, package = "")
Method adjust.offset()
Usage
model.interface.default.class$adjust.offset(
x,
envir,
package,
pred,
newdata,
type,
divide.by.mean = TRUE
)
Method clone()
The objects of this class are cloneable with this method.
Usage
model.interface.default.class$clone(deep = FALSE)
Arguments
deepWhether to make a deep clone.
Marchen/model.adapter documentation built on July 3, 2023, 11:07 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
| model.interface-class | R Documentation |
(Internal) An interface for model.adapter and statistical/machine learning models.
Description
This class provides unified interface for calls/objects of
statistical/machine learning models.
model.adapter-class{model.adapter} internally using object
of this class to get information from model calls/objects.
Methods
get.family(x, type = c("character", "family"), envir)
Get family from the model.
xcall or model object.
type = c("character", "family")-
a character literal specifying the type of data returned. If "character", this returns character vector of family name. If "family", this returns
familyobject. For some family specific for certain model (e.g., categorical family of MCMCglmm), their family object is not implimented. For such family, this method raise stop error. envirenvironment where x is evaluated.
get.call(x)
This method returns call by which the object is made. If call is not available, this returns NULL. To distinguish the returned value of NULL is intended action or not, inherited classes are encouraged to inherit this method to explicitly return NULL if x does not have call.
xcall or model object.
get.data(x, envir, package = "", ...)
Get a data.frame containing the data used for modeling. If data is not available this method returns NULL.
x-
a model object/call from which data is extracted.
enviran environment in which call is evaluated.
get.formula(x, envir, package = "")
Extract formula from model object/call.
If couldn't retrieve formula from x, this method returns NULL.
x-
a model object/call from which formula is extracted.
envir-
an environment in which call in x is evaluated.
package-
name of the package having the modeling function.
expand.formula(f, d, specials = NULL, package = NULL)
Expand . in formula.
fa formula to expand.
da data.frame used to expand . in formula.
specials = NULL-
special characterss passed to
terms.formula. package = NULL-
a character literal of package name having the model function.
predict(object, newdata, type, random, ...)
Calculate predictions.
objecta model object used for prediction.
newdata-
a data.frame containing data used for prediction.
type-
the type of prediciton. This should be a type specific for each modeling functions.
random = ~0-
the random effect to use. Tentatively, ~0 means don't use random effects.
...other variables passed to predict methods.
get.link(x, envir)
Get link function. If the model does not have link function, this
function returns identity function.
x-
an object of statistical model or a call of model function.
envir-
an environment where call in
xis evaluated.
get.linkinv(x, envir)
Get inverse function of link function. If the model does not have
link function, this function returns identity
function.
x-
an object of statistical model or a call of model function.
envir-
an environment where call in
xis evaluated.
get.model.type(x, envir, package = "", ...)
Return a character vector specifying model type (regression or classification). If the model is regression model, it returns 'regression'. If the model is classification model, it returns 'classification'.
x-
an object of statistical model or a call of model function.
envir-
an environment where call in
xis evaluated.
has.offset.argument(x, envir, package = "")
Check whether the model has offset argument.
x-
an object of statistical model or a call of model function.
envir-
an environment where call in
xis evaluated. package = NULL-
a character literal of package name having the model function.
has.offset.in.formula(x, envir, package = "")
Check whether the model has offset term(s) in formula.
x-
an object of statistical model or a call of model function.
envir-
an environment where formula and data in
xis evaluated. package = NULL-
a character literal of package name having the model function.
Active bindings
predict.typesa named character vector representing the conversion table of shared type of prediction used for predict() method. Currently, this vector has four elements named "response", "link", "prob" and "class" and each element represents compatible type of character for the model.
Methods
Public methods
Method new()
Usage
model.interface.default.class$new(...)
Method get.family()
Usage
model.interface.default.class$get.family(
x,
type = c("character", "family"),
envir
)Method get.call()
Usage
model.interface.default.class$get.call(x)
Method get.data()
Usage
model.interface.default.class$get.data(x, envir, package = "", ...)
Method get.formula()
Usage
model.interface.default.class$get.formula(x, envir, package = "")
Method expand.formula()
Usage
model.interface.default.class$expand.formula( f, d, specials = NULL, package = NULL )
Method predict()
Usage
model.interface.default.class$predict(object, newdata, type, random, ...)
Method get.link()
Usage
model.interface.default.class$get.link(x, envir)
Method get.linkinv()
Usage
model.interface.default.class$get.linkinv(x, envir)
Method get.model.type()
Usage
model.interface.default.class$get.model.type(x, envir, package = "", ...)
Method get.offset.names()
Usage
model.interface.default.class$get.offset.names(x, envir, package = "")
Method adjust.offset()
Usage
model.interface.default.class$adjust.offset( x, envir, package, pred, newdata, type, divide.by.mean = TRUE )
Method clone()
The objects of this class are cloneable with this method.
Usage
model.interface.default.class$clone(deep = FALSE)
Arguments
deepWhether to make a deep clone.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.