R6 class for prediction models

Provides standardized estimation and prediction methods

Author

Klaus Kähler Holst

Public fields

info

Optional information/name of the model

formals

List with formal arguments of estimation and prediction functions

formula

Formula specifying response and design matrix

args

additional arguments specified during initialization

Active bindings

fit

Active binding returning estimated model object

Methods

Public methods

• ml_model$new()

• ml_model$estimate()

• ml_model$predict()

• ml_model$update()

• ml_model$print()

• ml_model$response()

• ml_model$design()

• ml_model$opt()

• ml_model$clone()

---

Method new()

Create a new prediction model object

Usage

ml_model$new(
  formula = NULL,
  estimate,
  predict = stats::predict,
  predict.args = NULL,
  info = NULL,
  specials = c(),
  response.arg = "y",
  x.arg = "x",
  ...
)

Arguments

formula

formula specifying outcome and design matrix

estimate

function for fitting the model (must be a function response, 'y', and design matrix, 'x'. Alternatively, a function with a single 'formula' argument)

predict

prediction function (must be a function of model object, 'object', and new design matrix, 'newdata')

predict.args

optional arguments to prediction function

info

optional description of the model

specials

optional additional terms (weights, offset, id, subset, ...) passed to 'estimate'

response.arg

name of response argument

x.arg

name of design matrix argument

...

optional arguments to fitting function

---

Method estimate()

Estimation method

Usage

ml_model$estimate(data, ..., store = TRUE)

Arguments

data

data.frame

...

Additional arguments to estimation method

store

Logical determining if estimated model should be stored inside the class.

---

Method predict()

Prediction method

Usage

ml_model$predict(newdata, ..., object = NULL)

Arguments

newdata

data.frame

...

Additional arguments to prediction method

object

Optional model fit object

---

Method update()

Update formula

Usage

ml_model$update(formula, ...)

Arguments

formula

formula or character which defines the new response

...

Additional arguments to lower level functions

---

Method print()

Print method

Usage

ml_model$print(...)

Arguments

...

Additional arguments to lower level functions

---

Method response()

Extract response from data

Usage

ml_model$response(data, ...)

Arguments

data

data.frame

...

additional arguments to 'design'

---

Method design()

Extract design matrix (features) from data

Usage

ml_model$design(data, ...)

Arguments

data

data.frame

...

additional arguments to 'design'

---

Method opt()

Get options

Usage

ml_model$opt(arg, ...)

Arguments

arg

name of option to get value of

...

additional arguments to lower level functions

---

Method clone()

The objects of this class are cloneable with this method.

Usage

ml_model$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

data(iris)
rf <- function(formula, ...)
ml_model$new(formula, info="grf::probability_forest",
  estimate=function(x,y, ...) grf::probability_forest(X=x, Y=y, ...),
  predict=function(object, newdata)
             predict(object, newdata)$predictions, ...)

args <- expand.list(num.trees=c(100,200), mtry=1:3,
          formula=c(Species ~ ., Species ~ Sepal.Length + Sepal.Width))
models <- lapply(args, function(par) do.call(rf, par))

x <- models[[1]]$clone()
x$estimate(iris)
predict(x, newdata=head(iris))
#>         setosa  versicolor   virginica
#> [1,] 0.9869394 0.007651515 0.005409091
#> [2,] 0.9643561 0.033068182 0.002575758
#> [3,] 0.9870833 0.011250000 0.001666667
#> [4,] 0.9804167 0.017916667 0.001666667
#> [5,] 0.9927727 0.001818182 0.005409091
#> [6,] 0.9184302 0.059244006 0.022325758

 # Reduce Ex. timing
a <- targeted::cv(models, data=iris)
cbind(coef(a), attr(args, "table"))
#>              brier -logscore num.trees mtry
#> model1  0.09876431 0.2162048       100    1
#> model2  0.09676479 0.2133566       200    1
#> model3  0.08640535 0.1858628       100    2
#> model4  0.08757708 0.1854903       200    2
#> model5  0.08523466 0.1714144       100    3
#> model6  0.08144620 0.1673086       200    3
#> model7  0.33669049 0.5520703       100    1
#> model8  0.34025907 0.5600141       200    1
#> model9  0.33974766 0.5507261       100    2
#> model10 0.33571456 0.5493773       200    2
#> model11 0.33823288 0.5501077       100    3
#> model12 0.33389234 0.5471020       200    3
#>                                      formula
#> model1                           Species ~ .
#> model2                           Species ~ .
#> model3                           Species ~ .
#> model4                           Species ~ .
#> model5                           Species ~ .
#> model6                           Species ~ .
#> model7  Species ~ Sepal.Length + Sepal.Width
#> model8  Species ~ Sepal.Length + Sepal.Width
#> model9  Species ~ Sepal.Length + Sepal.Width
#> model10 Species ~ Sepal.Length + Sepal.Width
#> model11 Species ~ Sepal.Length + Sepal.Width
#> model12 Species ~ Sepal.Length + Sepal.Width


ff <- ml_model$new(estimate=function(y,x) lm.fit(x=x, y=y),
        predict=function(object, newdata) newdata%*%object$coefficients)
## tmp <- ff$estimate(y, x=x)
## ff$predict(x)