Skip to main content

Table 3 Best hyperparameters found for SVM on PrADG/HEG validation dataset (PrADG=40;HEG=40) with Auto-Weka

From: Predicting probable Alzheimer’s disease using linguistic deficits and biomarkers

Algorithm

Seed

Training time

Optimisation method

Hyperparameters

SVM-top-1000-

2

3 hours

SMAC

-C 1.4786727172414378 -N 1 -K "RBFKernel -G

PrADG/HEG

   

0.0014243946679106075”

  1. seed = random integer for randomising the data during training; SMAC is a Bayesian optimisation method proposed as part of Auto-Weka
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us