Table 2 Hyper-parameters defined for the models.

From: Application of multimodal machine learning-based analysis for the biomethane yields of NaOH-pretreated biomass

Model

Hyper-parameter

Specifications

ANN

Hidden layers

(10, 10)

Activation function

(ReLU)

Training algorithm

Levenberg Marquardt (LM)

Solver

Adam

Maximum iteration

500

SVM

Kernel

RBF

Regularization parameter (C)

100

Gamma

0.1

Epsilon

0.1

Decision Tree

Maximum Depth

5

Random Forest

Number of estimators

100

Maximum Depth

5

Back to article page