Linear Regression Constraint

Module for inserting ordinary Scikit-Learn regression models into a PySCIPOpt Model.

The following linear models are tested and should work:

• sklearn.linear_model.LinearRegression

• sklearn.linear_model.Ridge

• sklearn.linear_model.Lasso

• sklearn.linear_model.ElasticNet

pyscipopt_ml.sklearn.add_linear_regression_constr(scip_model, linear_regression, input_vars, output_vars=None, unique_naming_prefix='', **kwargs)

Formulate linear_regression as a constraint in a PySCIPOpt Model.

The formulation predicts the values of output_vars using input_vars according to linear_regression.

Parameters:

• scip_model (PySCIPOpt Model) – The scip model where the predictor should be inserted.

• linear_regression (sklearn.linear_model.LinearRegression)

• types (The linear regression to insert. It can be any of the following) –

  • sklearn.linear_model.LinearRegression

  • sklearn.linear_model.Ridge

  • sklearn.linear_model.Lasso

  • sklearn.linear_model.ElasticNet

• input_vars (list or dict) – Decision variables used as input for the regression model

• output_vars (list or dict) – Decision variables used as output for the regression model

• unique_naming_prefix (str, optional) – A unique naming prefix that is used before all variable and constraint names. This parameter is important if the SCIP model is later printed to file and many predictors are added to the same SCIP model.

Returns:

Object containing information about what was added to scip to formulate linear_regression.

Return type:

LinearRegressionConstr

Note

See add_predictor_constr for acceptable values for input_vars and output_vars

class pyscipopt_ml.sklearn.linear_regression.LinearRegressionConstr(scip_model, predictor, input_vars, output_vars=None, unique_naming_prefix='', **kwargs)

Class to model trained sklearn.linear_model.LinearRegression with SCIP

Stores the changes to the SCIP Model for representing an instance into it. Inherits from AbstractPredictorConstr..