R/mars_data.R

In parsnip: A Common API to Modeling and Analysis Functions

set_new_model("mars")

set_model_mode("mars", "classification")
set_model_mode("mars", "regression")

# ------------------------------------------------------------------------------

set_model_engine("mars", "classification", "earth")
set_model_engine("mars", "regression", "earth")
set_dependency("mars", "earth", "earth")

set_model_arg(
  model = "mars",
  eng = "earth",
  parsnip = "num_terms",
  original = "nprune",
  func = list(pkg = "dials", fun = "num_terms", range = c(2, 5)),
  has_submodel = TRUE
)
set_model_arg(
  model = "mars",
  eng = "earth",
  parsnip = "prod_degree",
  original = "degree",
  func = list(pkg = "dials", fun = "prod_degree"),
  has_submodel = FALSE
)
set_model_arg(
  model = "mars",
  eng = "earth",
  parsnip = "prune_method",
  original = "pmethod",
  func = list(pkg = "dials", fun = "prune_method"),
  has_submodel = FALSE
)

set_fit(
  model = "mars",
  eng = "earth",
  mode = "regression",
  value = list(
    interface = "formula",
    protect = c("formula", "data", "weights"),
    func = c(pkg = "earth", fun = "earth"),
    defaults = list(keepxy = TRUE)
  )
)

set_encoding(
  model = "mars",
  eng = "earth",
  mode = "regression",
  options = list(
    predictor_indicators = "none",
    compute_intercept = FALSE,
    remove_intercept = FALSE,
    allow_sparse_x = FALSE
  )
)

set_fit(
  model = "mars",
  eng = "earth",
  mode = "classification",
  value = list(
    interface = "formula",
    protect = c("formula", "data", "weights"),
    func = c(pkg = "earth", fun = "earth"),
    defaults = list(keepxy = TRUE)
  )
)

set_encoding(
  model = "mars",
  eng = "earth",
  mode = "classification",
  options = list(
    predictor_indicators = "none",
    compute_intercept = FALSE,
    remove_intercept = FALSE,
    allow_sparse_x = FALSE
  )
)

set_pred(
  model = "mars",
  eng = "earth",
  mode = "regression",
  type = "numeric",
  value = list(
    pre = NULL,
    post = maybe_multivariate,
    func = c(fun = "predict"),
    args =
      list(
        object = quote(object$fit),
        newdata = quote(new_data),
        type = "response"
      )
  )
)

set_pred(
  model = "mars",
  eng = "earth",
  mode = "regression",
  type = "raw",
  value = list(
    pre = NULL,
    post = NULL,
    func = c(fun = "predict"),
    args =
      list(object = quote(object$fit),
           newdata = quote(new_data))
  )
)

set_pred(
  model = "mars",
  eng = "earth",
  mode = "classification",
  type = "class",
  value = list(
    pre = NULL,
    post = function(x, object) {
      x <- ifelse(x[, 1] >= 0.5, object$lvl[2], object$lvl[1])
      x
    },
    func = c(fun = "predict"),
    args =
      list(
        object = quote(object$fit),
        newdata = quote(new_data),
        type = "response"
      )
  )
)

set_pred(
  model = "mars",
  eng = "earth",
  mode = "classification",
  type = "prob",
  value = list(
    pre = NULL,
    post = function(x, object) {
      x <- x[, 1]
      x <- tibble(v1 = 1 - x, v2 = x)
      colnames(x) <- object$lvl
      x
    },
    func = c(fun = "predict"),
    args =
      list(
        object = quote(object$fit),
        newdata = quote(new_data),
        type = "response"
      )
  )
)

set_pred(
  model = "mars",
  eng = "earth",
  mode = "classification",
  type = "raw",
  value = list(
    pre = NULL,
    post = NULL,
    func = c(fun = "predict"),
    args =
      list(object = quote(object$fit),
           newdata = quote(new_data))
  )
)