Skip to main content

Table 3 Ablation study of different individual components of Weighted_Rank

From: A hybrid framework for improving uncertainty quantification in deep learning-based QSAR regression modeling

Combinations

Splitting strategies

IVIT

IVOT

OVOT

E + L + F

0.306

0.237

0.195

M + L + F

0.302

0.228

0.198

M + E + F

0.296

0.229

0.197

M + E + L

0.300

0.216

0.196

M + E + F + L

0.312

0.236

0.208

  1. SCC is used as the evaluation metric
  2. Each letter represents an individual method: M refers to MVE, E refers to ENS, L refers to LDIST and F refers to FDIST
  3. Best results are highlighted in boldface type for each splitting strategy
Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us