Theano-based Large-Scale Visual Recognition with Multiple GPUs

By Weiguang Ding, Ruoyan Wang, Fei Mao, Graham W. Taylor. 2014

46 citations recorded by the source corpus.

In this report, we describe a Theano-based AlexNet (Krizhevsky et al., 2012) implementation and its naive data parallelism on multiple GPUs. Our performance on 2 GPUs is comparable with the state-of-art Caffe library (Jia et al., 2014) run on 1 GPU. To the best of our knowledge, this is the first open-source Python-based AlexNet implementation to-date.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Compression of fully-connected layer in neural network by Kronecker product (2015, methodology): “We use the AlexNet [1] as the baseline network and use the implementation in [25].”
Multi-Way, Multilingual Neural Machine Translation with a Shared Attention Mechanism (2016, methodology): “The implementation was based on the work in (Ding et al., 2014).”
RPAN: An End-to-End Recurrent Pose-Attention Network for Action Recognition in Videos (2017, methodology): “We implement our RPAN by Theano [35], with multi-GPU of BPTT [8].”
Comparative Study of Deep Learning Software Frameworks (2015, methodology): “We also noticed from our experiments that the LMDB database used in Caﬀe has signiﬁcantly better performance than the other data access layers used in Neon, Theano and Torch frameworks 10 as it supports concurrent reads.”