Skip to main content

Table 7 Comparison of joint VAE and Regression training strategies

From: Generative model based on junction tree variational autoencoder for HOMO value prediction and molecular optimization

Order

Train

Test

HOMO Acc

VAE Acc (%)

Enc, Dec\(\xrightarrow []{}\)FFNN

MIX

QM9

0.32

84

Enc, Dec\(\xrightarrow []{}\)Enc, FFNN

MIX

QM9

0.09

0

Enc, Dec\(\xrightarrow []{}\)Enc, FFNN\(\xrightarrow []{}\)Dec

MIX

QM9

0.09

81

Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us