Speaker: Wei Wang
SINGA is an Apache incubator project started since March 2015. 3 versions have been released with focus on distributed training of deep learning models. In this talk, we will share the architecture, programming model and examples of V1.0, which improves the usability and efficiency over V0.3. In particular, users can write C++ and Python code to train deep learning models. With the support of OpenCL, SINGA is able to run on Nvidia GPU and other hardware devices (e.g., AMD GPU and FPGA). Optimisations in terms of speed and memory management would also be discussed.
Produced by Engineers.SG
Help us caption & translate this video!