catboost.virtual_ensembles_predict {catboost}R Documentation

Apply the model with several virtual ensembles

Description

Apply the model to the given dataset using several independent truncated models - virtual ensembles. Each tree in ensemble predicts its own value for each document from pool.

Peculiarities: Return value varies on prediction_type: array for 'VirtEnsembles' and matrix for 'TotalUncertainty'

Usage

catboost.virtual_ensembles_predict(
  model,
  pool,
  verbose = FALSE,
  prediction_type = "VirtEnsembles",
  ntree_end = 0L,
  virtual_ensembles_count = 10,
  thread_count = -1
)

Arguments

model

The model obtained as the result of training.

Default value: Required argument

pool

The input dataset.

Default value: Required argument

verbose

Verbose output to stdout.

Default value: FALSE (not used)

prediction_type

The format for displaying approximated values in output data (see https://catboost.ai/docs/concepts/python-reference_virtual_ensembles_predict.html#python-reference_catboostclassifier_predict__output-format).

Possible values:

  • 'VirtEnsembles'

  • 'TotalUncertainty'

Default value: 'VirtEnsembles'

ntree_end

Index of the first tree not to be used when applying the model or calculating the metrics (zero-based indexing).

Default value: 0 (the index of the last tree to use equals to the number of trees in the model minus one)

virtual_ensembles_count

Number of tree ensembles to use. Each virtual ensemble can be considered as truncated model.

Default value: 10

thread_count

The number of threads to use when applying the model. If -1, then the number of threads is set to the number of CPU cores.

Allows you to optimize the speed of execution. This parameter doesn't affect results.

Default value: -1

Value

Matrix or Array of predictions (for 'TotalUncertainty' and 'VirtEnsembles' prediction_type correspondingly)

See Also

https://catboost.ai/docs/concepts/python-reference_virtual_ensembles_predict.html?lang=en


[Package catboost version 1.0.4 Index]