Table 6 Performance comparison of different models in tmin 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%)

88.19%

74.31%

81.25%

MAE(Mean Absolute Error)

0.0069

0.0082

0.0079

MSE(Mean Squared Error)

0.0001

0.0001

0.0001

RMSE(Root Mean Squared Error)

0.0094

0.0120

0.0107

MAPE(Mean Absolute Percentage Error)

1.58%

2.60%

1.78%

R²(Coefficient of Determination)

0.9353

0.9076

0.9160

  1. Extreme low-temperature capture rate improved to 85.6% (from 68.5%).
Back to article page