Table 3 Hyperparameter Optimization Ranges

From: Using genomic context informed genotype data and within-model ancestry adjustment to classify type 2 diabetes

Optimization Task

Optimization Range

Number of Convolutional Blocks

1–3

Number of Convolutional Filters

10–100

Convolutional Filter Width

1–20

Pooling Width

1–20

Number of Dense Layers

1–3

Nodes Within Dense Layer

5–100

L2 regularization factor

0 – 1e-4

Dropout Rate

0–0.5

Gradient Reversal Weight

0.001–0.2

Learning Rate

1e-6–1e-3