precompute_feature_info | R Documentation |
Creates data assignment and subsequently extracts feature information such as normalisation and clustering parameters.
precompute_feature_info(
formula = NULL,
data = NULL,
experiment_data = NULL,
cl = NULL,
experimental_design = "fs+mb",
verbose = TRUE,
...
)
formula |
An R formula. The formula can only contain feature names and
dot ( Use of the formula interface is optional. |
data |
A
All data is expected to be in wide format, and ideally has a sample
identifier (see In case paths are provided, the data should be stored as |
experiment_data |
Experimental data may provided in the form of |
cl |
Cluster created using the This parameter has no effect if the |
experimental_design |
(required) Defines what the experiment looks
like, e.g.
The different components are linked using Different subsampling methods can be used in conjunction with the basic workflow components:
As shown in the example above, sampling algorithms can be nested. Though neither variable importance is determined nor models are learned
within The simplest valid experimental design is This argument is ignored if the |
verbose |
Indicates verbosity of the results. Default is TRUE, and all messages and warnings are returned. |
... |
Arguments passed on to
|
This is a thin wrapper around summon_familiar
, and functions like
it, but automatically skips computation of variable importance, learning
and subsequent evaluation steps.
The function returns an experimentData
object, which can be used to
warm-start other experiments by providing it to the experiment_data
argument.
An experimentData
object.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.