Skip to main content

Table 4 The detailed parameters of interaction prediction head

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

Block

Layers

Input dimensions

Output dimensions

Dropout rate

Input

Concatenation

512\(^{1}\), 512\(^{2}\)

1024

 

Block 1

FC layer

1024

2048

 

GeLU & Dropout

2048

2048

0.1

Block 2

FC layer

2048

1024

 

GeLU & Dropout

1024

1024

0.1

Block 3

FC layer

1024

512

 

GeLU & Dropout

512

512

0.1

Output

FC layer

512

1

 
  1. \(^{1}\) Feature of drug sequence, \(z_{\textrm{drug}}\)
  2. \(^{2}\) Feature of target sequence, \(z_{\textrm{target}}\)
Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us