Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method and apparatus for processing data sequences

A technology for processing data and data sequences, applied in physical implementation, neural architecture, biological neural network model, etc., can solve problems such as inability to process data, small first-level cache capacity, inability to store RNN model weight matrix at the same time, etc., to improve efficiency Effect

Inactive Publication Date: 2017-03-22
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF3 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, the capacity of the first-level cache is generally small, and it cannot store all the weight matrices in the RNN model at the same time, and because of the feedback, the RNN model cannot process each data in the data sequence at the same time
Therefore, when using the RNN model to process each data in the data sequence, it is necessary to re-copy the weight matrix to the first-level cache of the CPU or GPU, resulting in low efficiency of the RNN model to process the data sequence

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and apparatus for processing data sequences
  • Method and apparatus for processing data sequences
  • Method and apparatus for processing data sequences

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] The application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain related inventions, rather than to limit the invention. It should also be noted that, for the convenience of description, only the parts related to the related invention are shown in the drawings.

[0026] It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and embodiments.

[0027] figure 1 An exemplary system architecture 100 is shown to which embodiments of the method for processing a data sequence or the embodiment of the apparatus for processing a data sequence of the present application can be applied.

[0028] like figure 1 As shown, the...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The application discloses a method and apparatus for processing data sequences. The method comprises: an inputted to-be-processed data sequence is received; a weight matrix of a recurrent neural network model is copied to an embedded block random access memory (RAM) of a field programmable gate array (FPGA); all to-be-processed data are processed in the to-be-processed data sequence by using an activation function in the recurrent neural network model and the weight matrix stored into the embedded block random access memory; and a processed data sequence corresponding to the to-be-processed data sequence is outputted. Therefore, the efficiency of processing the data sequence by the recurrent neural network model is improved.

Description

technical field [0001] The present application relates to the field of computer technology, specifically to the field of data processing technology, and in particular to methods and devices for processing data sequences. Background technique [0002] RNN (Recurrent neural Network) is an artificial neural network in which nodes are directional connected into a ring. The internal state of the network can exhibit dynamic timing behavior. Different from the feedforward neural network, RNN can use its internal memory to process any sequence of input sequences, which makes it easier to handle such as non-segmented handwriting recognition, speech recognition, etc. At present, the RNN model has been widely used in video processing, speech recognition, semantic understanding and other services. [0003] When using the RNN model to process data sequences, it is necessary to first copy the weight matrix of the RNN model to the first-level cache of the CPU (Central Processing Unit, ce...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06N3/06
CPCG06N3/063G06N3/044
Inventor 王勇欧阳剑漆维李思仲
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com