Table 1 Parameters of DeEPsnap.

From: A deep ensemble framework for human essential gene prediction by integrating multi-omics data

  

Shape

Activation

Dropout

DNN Structure

Input

258

–

–

Hidden Layer 1

1024

ReLU

0.3

Hidden Layer 2

512

ReLU

0.3

Hidden Layer 3

256

ReLU

0.3

Output Layer

1

Sigmoid

–

 

Initial Learning Rate (\(lr_0\))

#Epochs per Cycle (C)

Snapshot ensemble

0.001

5

 

\(\#\)Epochs (T)

Optimizer

Batch size

Class weights

Training

50

Adam

32

1 : 4.5

Back to article page