Table 2 Hyperparameter setting for each of the used algorithms and models in this research.

From: A novel early stage drip irrigation system cost estimation model based on management and environmental variables

Model

1

2

3

Multivariate Linear Regression (MLR)

n_features = 39

regularization = 0.01

fit_intercept = True

Support Vector Regression (SVR)

kernel = “rbf”

C = 1.5

epsilon = 0.1

Artificial Neural Network (ANN)

hidden_layer_sizes = (100,)

activation = “relu”

learning_rate_init = 0.001

Gene Expression Programming (GEP)

population_size = 50

mutation_rate = 0.02

selection_method = “roulette”

Genetic Algorithm (GA)

crossover_prob = 0.8

mutation_prob = 0.1

elite_size = 5

Deep Learning (DL)

num_layers = 3

units_per_layer = [64, 32, 16]

dropout_rate = 0.25

Decision Tree (DT)

max_depth = 10

min_samples_split = 4

criterion = “mse”

Back to article page