An Improved Stacking Model for Equipment Spare Parts Demand Forecasting Based on Scenario Analysis

Figure 11

Performance of Different Models. (a) cv_mean of different models. (b) score of different models. (c) test_score of different models. (d) The error of different models. (cv_mean is the cross_val_score of models on the validation set, which is R2_Score; train_score is the cross_val_score of models on the training set, which is R2_Score; test_score returns the coefficient of determination R^2 of the prediction; mean_absolute_error is the expected value of the absolute error loss; mean_squared_error is the expected value of the squared (quadratic) error or loss.).

(a)

(b)

(c)

(d)