Skip to main content

Table 6 System parameters to train 3D U-Net network

From: A state-of-the-art technique to perform cloud-based semantic segmentation using deep learning 3D U-Net architecture

Initial learning rate

5e-4

Maxepochs

100

Learning rate schedule

Piecewise

Learning rate drop period

5

Learning rate drop factor

0.95

Validation frequency

400

Verbose

False

Mini-batch size

8

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us