公开(公告)号：US20180218518A1

公开(公告)日：2018-08-02

申请号：US15422359

申请日：2017-02-01

Applicant: NVIDIA Corporation

Inventor： Zhou Yan ,  Franciscus Wilhelmus Sijstermans ,  Yuanzhi Hua ,  Xiaojun Wang ,  Jeffrey Michael Pool ,  William J. Dally ,  Liang Chen

IPC: G06T9/00 ,  G06T1/20 ,  G06T1/60

CPC classification number: G06T9/002 ,  G06T1/20 ,  G06T1/60

Abstract: A method, computer program product, and system for sparse convolutional neural networks that improves efficiency is described. Multi-bit data for input to a processing element is received at a compaction engine. The multi-bit data is determined to equal zero and a single bit signal is transmitted from the memory interface to the processing element in lieu of the multi-bit data, where the single bit signal indicates that the multi-bit data equals zero. A compacted data sequence for input to a processing element is received by a memory interface. The compacted data sequence is transmitted from the memory interface to an expansion engine. Non-zero values are extracted from the compacted data sequence and zeros are inserted between the non-zero values by the expansion engine to generate an expanded data sequence that is output to the processing element.

公开(公告)号：US10096134B2

公开(公告)日：2018-10-09

申请号：US15422359

申请日：2017-02-01

Applicant: NVIDIA Corporation

Inventor： Zhou Yan ,  Franciscus Wilhelmus Sijstermans ,  Yuanzhi Hua ,  Xiaojun Wang ,  Jeffrey Michael Pool ,  William J. Dally ,  Liang Chen

IPC: H03M7/00 ,  G06T9/00 ,  G06T1/20 ,  G06T1/60

Abstract: A method, computer program product, and system for sparse convolutional neural networks that improves efficiency is described. Multi-bit data for input to a processing element is received at a compaction engine. The multi-bit data is determined to equal zero and a single bit signal is transmitted from the memory interface to the processing element in lieu of the multi-bit data, where the single bit signal indicates that the multi-bit data equals zero. A compacted data sequence for input to a processing element is received by a memory interface. The compacted data sequence is transmitted from the memory interface to an expansion engine. Non-zero values are extracted from the compacted data sequence and zeros are inserted between the non-zero values by the expansion engine to generate an expanded data sequence that is output to the processing element.