Skip to main content

Table 1 Range of hyperparameter searched in the CNN classifier

From: Capturing the differences between humoral immunity in the normal and tumor environments from repertoire-seq of B-cell receptors using supervised machine learning

Name of hyperparameter Lower limit Upper limit
Initial learning rate 1e-7 1e-3
Dropout rate 0.4 1.0
# of convolutional layers 1 2
# of convolutional kernels 80 300
The width of convolution kernels 2 3
The width of pooling filters 2 3
# of fully connected layers 1 3
# of units in a fully connected layer 100 300