Common interface to train a model. The learning method is provided in trainer
Usage
train(trainer, features, ...)
# S4 method for Trainer,ANY
train(
trainer,
x,
y,
weights = NULL,
offset = NULL,
resp.type,
observations = NULL,
features = NULL,
...
)
# S4 method for Trainer,missing
train(trainer, features, ...)
# S4 method for Trainer,list
train(
trainer,
x,
y,
weights = NULL,
offset = NULL,
resp.type,
observations = NULL,
features,
logger = Logger(verbose = FALSE),
...
)Arguments
- trainer
a Trainer object providing the learning method
- features
(optional) indices of predictors to keep
- ...
additional arguments to training method
- x
the input matrix, where rows are observations and columns are variables.
- y
the response variable. Its number of rows must match the number of rows of
x.- weights
priors of the observations
- offset
used only for GLM methods, it is an a priori known component to be included in the linear predictor during fitting
- resp.type
the response type
- observations
(optional) indices of observations to keep
Value
a Trained object or a TrainedList object if learning method runs multiple configurations (i.e. multiple sets of hyper-parameters).
Results are merged into one TrainedList object