Table 2 Parameter setup

From: An AI-driven multi-omics framework identifies lactylation-mediated therapeutic targets to overcome drug resistance in ovarian cancer

Hyperparameters

Typical values

Hidden units per dense layer

64, 128, 256

Epochs

50, 100

Dropout rate

0.2, 0.3, 0.5

Optimizer

Adam, RMSProp

Batch size

32, 64, 128

Learning rate

0.001, 0.0001

Number of LSTM layers

1, 2, 3

Activation function

ReLU, tanh, sigmoid

Number of MLP hidden layers

1, 2, 3

Number of filters (if CNN used)

32, 64

Back to article page