LAMP: Large Deep Nets with Automated Model Parallelism for Image Segmentation

Wentao Zhu
Wentao Zhu
Can Zhao
Can Zhao
Wenqi Li
Wenqi Li

medical image computing and computer assisted intervention, pp. 374-384, 2020.

Cited by: 0|Views31

Abstract:

Deep Learning (DL) models are becoming larger, because the increase in model size might offer significant accuracy gain. To enable the training of large deep networks, data parallelism and model parallelism are two well-known approaches for parallel training. However, data parallelism does not help reduce memory footprint per device. In t...More

Code:

Data:

Get fulltext within 24h
Bibtex
Your rating :
0

 

Tags
Comments