Skip to main content

Table 3 Parameters tuned during the training phase using grid search

From: KekuleScope: prediction of cancer cell line sensitivity and compound potency using convolutional neural networks trained on compound images

Parameter Values evaluated
Learning rate (Lr) {0.1, 0.01, 0.001, 0.005, 0.001, 0.0001}
Decay rate {0.1, 0.6}
Annealing rate step {10, 25}
Data augmentation (augmentation) {Yes: 1, No: 0}
Batch size (batch) {4, 16, 32}
  1. The names in parentheses indicate the parameter name abbreviation used in the main text and figures