Skip to main content

Table 4 ROC-AUC score per feature reduction technique for TabNet

From: Analysis of RNA-Seq data using self-supervised learning for vital status prediction of colorectal cancer patients

Model

Feature reduction

ROC-AUC

TabNet without pretraining (supervised learning)

Lasso

0.712 ± 0.032

PCA

0.642 ± 0.054

T-test

0.742 ± 0.042

TabNet with pretraining (self supervised learning)

Lasso

0.729 ± 0.031

PCA

0.73 ± 0.047

T-test

0.842 ± 0.022

  1. The RoC-AUC of the best performing model is indicated in bold
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us