Table 13 Computational performance metrics of top 5 Models.

From: Comprehensive framework of machine learning and deep learning architectures with metaheuristic optimization for high-fidelity prediction of nanofluid specific heat capacity

Final Models

Training Time

Testing Time

Avg. Inference

Model Size (KB)

Gradient Boosting + LR

1.23520

0.00001

0.00001

1,123.798

CatBoost

0.52800

0.00099

0.00001

482.009

XGBoost

0.20593

0.00199

0.00001

638.161

MLP + LR

20.44244

0.00099

0.00001

482.428

GRU

1643.39918

0.62728

0.00247

317.670

Back to article page