Table 1 Hyperparameter space searched
From: Predicting anticancer hyperfoods with graph convolutional networks
Hyperparameter | Space search |
|---|---|
Learning rate | 5.10−4,5.10−3 |
L2-regularization | 1.10−5,1.10−4,5.10−4 |
Number of convolutional layers | 1,2,3 |
Number of dropout layers | 1,2 |
Batch normalization | True,false |
Feature normalization | True,false |
n-hops for ChebNet | 2,4,6 |