Research on NVIDIA Deep Learning Accelerator
Gaofeng Zhou, Jianyang Zhou, Haijun Lin · 2018
This paper introduces the NVIDIA deep learning accelerator (NVDLA), including its hardware architecture specification and software environment. At the same time, the basic function on the inference side of convolutional neural networks (CNN) of NVDLA is verified by using the LeNet network model of Caffe framework in the virtual platform provided by NVIDIA. Experiment shows that NVDLA can analyze the forward-designed LeNet network model information and use the network training data to implement the inference function to complete the handwriting digit recognition.