Table 3 Performance comparison of different models in pre data prediction.

From: High-resolution climate prediction in mountainous terrain using a ConvLSTM-XGBoost hybrid model with dynamic bayesian weighting

Indicator

ConvLSTM Model

XGBoost Model

ConvLSTM-XGBoost Hybrid Model

Accuracy (Error ≤ 3%)

72.92%

55.56%

79.17%

MAE(Mean Absolute Error)

0.0090

0.0128

0.0089

MSE(Mean Squared Error)

0.0001

0.0003

0.0001

RMSE(Root Mean Squared Error)

0.0114

0.0177

0.0117

MAPE(Mean Absolute Percentage Error)

2.08%

3.84%

1.97%

R²(Coefficient of Determination)

0.9759

0.9557

0.9747

  1. *Compared to XGBoost; F1-score for extreme precipitation events improved by 20%.
Back to article page