English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep learning and Convolutional Neural Network (CNN) have becomingincreasingly more popular and important in both academic and industrial areasin recent years cause they are able to provide better accuracy and result inclassification, detection and recognition areas, compared to traditionalapproaches. Currently, there are many popular frameworks in the market for deeplearning development, such as Caffe, TensorFlow, Pytorch, and most offrameworks natively support CPU and consider GPU as the mainline accelerator bydefault. FPGA device, viewed as a potential heterogeneous platform, stillcannot provide a comprehensive support for CNN development in popularframeworks, in particular to the training phase. In this paper, we firstlypropose the FeCaffe, i.e. FPGA-enabled Caffe, a hierarchical software andhardware design methodology based on the Caffe to enable FPGA to supportmainline deep learning development features, e.g. training and inference withCaffe. Furthermore, we provide some benchmarks with FeCaffe by taking someclassical CNN networks as examples, and further analysis of kernel executiontime in details accordingly. Finally, some optimization directions includingFPGA kernel design, system pipeline, network architecture, user caseapplication and heterogeneous platform levels, have been proposed gradually toimprove FeCaffe performance and efficiency. The result demonstrates theproposed FeCaffe is capable of supporting almost full features during CNNnetwork training and inference respectively with high degree of designflexibility, expansibility and reusability for deep learning development.Compared to prior studies, our architecture can support more network andtraining settings, and current configuration can achieve 6.4x and 8.4x averageexecution time improvement for forward and backward respectively for LeNet.

FeCaffe: FPGA-enabled Caffe with OpenCL for Deep Learning Training and Inference on Intel Stratix 10