Skip to main content

Table 2 The parameter setting for our Curv-SIAGN model

From: CurvAGN: Curvature-based Adaptive Graph Neural Networks for Predicting Protein-Ligand Binding Affinity

Learning rate

Batch size

Hyper-parameter

Cutoff for graph

Dim of Node_emb

0.001

32

\(\lambda =1.75\)

\(\theta _d=5\)

128

Heads

Dim of Edge_emb

Cutoff for matrix

Filtrations

Angle domains

4

128

\(\theta _{\rho } = 12\)

50

6

Dropout

Blocks

Dim of curvature

  

0.2

4

128

  
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us