Skip to main content

Table 2 Hyperparameters optimization for MPN and SAMPN

From: A self-attention based message passing neural network for predicting molecular lipophilicity and aqueous solubility

MPN and SAMPN

Hyperparameters

Range (interval)

 

Activation function

Tanh, ELU, LeakyReLU ReLU, PReLU, SELU

 

Steps of message passing

2–6 (1)

 

Graph embedding size

32–512 (32)

 

Dropout rate

0.0–0.4 (0.05)

 

Layers of fully connected network

1–3 (1)

Back to article page