Table 6 Performance of models on the test set for different feature sets.

From: Deep insight: an efficient hybrid model for oil well production forecasting using spatio-temporal convolutional networks and Kolmogorov–Arnold networks

Models

R2

RMSE

R2

RMSE

Models

R2

RMSE

R2

RMSE

Selected features

All features

Selected features

All features

RNN

0.9084

14.87

0.9150

13.95

KAN

0.9026

15.47

0.9105

14.35

LSTM

0.9177

14.42

0.9240

13.11

TCN-KAN

0.9307

13.55

0.9401

12.74

GRU

0.9115

14.72

0.9205

13.55

GA-TCN-KAN

0.9642

11.66

0.9700

10.95

TCN

0.8958

15.33

0.9052

14.27

PSO-TCN-KAN

0.9711

10.59

0.9765

9.88

CNN-LSTM

0.9140

16.04

0.9195

14.90

WOA-TCN-KAN

0.9815

9.93

0.9850

8.95