MachineShop (version 1.1.0)

predict: Model Prediction

Description

Predict outcomes with a fitted model.

Usage

# S3 method for MLModelFit
predict(object, newdata = NULL,
  type = c("response", "prob"), cutoff = 0.5, times = numeric(), ...)

Arguments

object

MLModelFit object from a model fit.

newdata

optional data.frame with which to obtain predictions. If not specified, the training data will be used by default.

type

specifies prediction on the original outcome scale ("response") or on a probability distribution scale ("prob").

cutoff

threshold above which binary factor probabilities are classified as events, below which survival probabilities are classified, and at which expected values are rounded for integer outcomes.

times

numeric vector of follow-up times at which to predict survival events.

...

arguments passed to model-specific prediction functions.

See Also

fit, confusion, performance

Examples

# NOT RUN {
## Survival response example
library(survival)
library(MASS)

gbmfit <- fit(Surv(time, status != 2) ~ sex + age + year + thickness + ulcer,
              data = Melanoma, model = GBMModel)
predict(gbmfit, newdata = Melanoma, times = 365 * c(2, 5, 10), type = "prob")

# }