Extending Caffe for Machine Learning of Large Neural Networks Distributed on GPUs


KIPS Transactions on Computer and Communication Systems, Vol. 7, No. 4, pp. 99-102, Apr. 2018
10.3745/KTCCS.2018.7.4.99,   PDF Download:
Keywords: Parallel Programming, GPU Computing, Machine Learning
Abstract

Caffe is a neural net learning software which is widely used in academic researches. The GPU memory capacity is one of the most important aspects of designing neural net architectures. For example, many object detection systems require to use less than 12GB to fit a single GPU. In this paper, we extended Caffe to allow to use more than 12GB GPU memory. To verify the effectiveness of the extended software, we executed some training experiments to determine the learning efficiency of the object detection neural net software using a PC with three GPUs.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
O. Jong-soo and L. Dongho, "Extending Caffe for Machine Learning of Large Neural Networks Distributed on GPUs," KIPS Transactions on Computer and Communication Systems, vol. 7, no. 4, pp. 99-102, 2018. DOI: 10.3745/KTCCS.2018.7.4.99.

[ACM Style]
Oh Jong-soo and Lee Dongho. 2018. Extending Caffe for Machine Learning of Large Neural Networks Distributed on GPUs. KIPS Transactions on Computer and Communication Systems, 7, 4, (2018), 99-102. DOI: 10.3745/KTCCS.2018.7.4.99.