gemseo / mlearning / regression

random_forest module

Random forest regression

The random forest regression uses averaging methods on an ensemble of decision trees.


The regression model relies on the RandomForestRegressor class of the scikit-learn library.

class gemseo.mlearning.regression.random_forest.RandomForestRegressor(data, transformer=None, input_names=None, output_names=None, n_estimators=100, **parameters)[source]

Bases: gemseo.mlearning.regression.regression.MLRegressionAlgo

Random forest regression


  • data (Dataset) – learning dataset.

  • transformer (dict(str)) – transformation strategy for data groups. If None, do not transform data. Default: None.

  • input_names (list(str)) – names of the input variables.

  • output_names (list(str)) – names of the output variables.

  • n_estimators (int) – number of trees in the forest.

  • parameters – other keyword arguments for the sklearn algo.

ABBR = 'RandomForestRegressor'
LIBRARY = 'scikit-learn'