Skip to main content

Table 7 Time and memory analysis of baseline models and DLM-DTI

From: DLM-DTI: a dual language model for the prediction of drug-target interaction with hint-based learning

 

Parameters (M)

Training time (S)\(^{1}\)

Memory capacity (GB)\(^{2}\)

AUPRC\(^{3}\)

MolTrans

62.8

75.33 ± 3.51

5.9

0.622 ± 0.007

Kang et al.

353.0

631.00 ± 17.06

47.4

0.623 ± 0.010

DLM-DTI

86.7

63.00 ± 1.00

7.7

0.643 ± 0.006

  1. M: millions, S: seconds, GB: Giga bytes
  2. \(^{1}\) Mean ± SD
  3. \(^{2}\) Batch size is matched to 32
  4. \(^{3}\) Results of Binding DB with single training setting. Best performance is highlighted in bold
Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us