A Novel Stacking Heterogeneous Ensemble Model with Hybrid Wrapper-Based Feature Selection for Reservoir Productivity Predictions

Algorithm 1.

Algorithm 1: Proposed reservoir productivity forecast model.

(i)	Require:
(ii)	Training dataset1 ,  = 1, …, , where are the initial features and is the size of the training set;
(iii)	Training dataset2 ,  = 1, …, , where are the main features and is the size of the training set;
(iv)	Base-regressors , , …, , where is the number of base-regressors;
(v)	Ensure:
(vi)	Stacking heterogeneous ensemble regressor ;
(1)	for do
(2)	Train base-regressor on ;
(3)	end for
(4)	for each do
(5)	Construct new dataset of predictions, where ; Train a meta-regressor on .
(6)	end for
(7)	for eachdo
(8)	Train base-regressor on ;
(9)	end for
(10)	for eachdo
(11)	Construct new dataset of predictions, where ; Train a meta-regressor on .
(12)	end for
(13)	for each do
(14)	Construct new dataset of predictions, where ; Train a meta-regressor on .
(15)	end for
(16)	return;