Unlock instant, AI-driven research and patent intelligence for your innovation.

Data batch standardization method, computing device and computer readable storage medium

A data and batch technology, applied in the field of deep neural network computing, can solve problems such as low cache hit rate

Active Publication Date: 2021-06-08
SHANGHAI BIREN TECH CO LTD
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, during the operation of the BN operator, the hit rate of the cache is low, and data needs to be imported from the memory, which further leads to additional overhead

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
  • Data batch standardization method, computing device and computer readable storage medium
  • Data batch standardization method, computing device and computer readable storage medium
  • Data batch standardization method, computing device and computer readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0018] Preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although preferred embodiments of the invention are shown in the drawings, it should be understood that the invention may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0019] As used herein, the term "comprise" and its variants mean open inclusion, ie "including but not limited to". The term "or" means "and / or" unless otherwise stated. The term "based on" means "based at least in part on". The terms "one embodiment" and "some embodiments" mean "at least one example embodiment." The term "another embodiment" means "at least one further embodiment". The terms "first", "second", etc. may refer to different or the same object...

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 invention provides a data batch standardization method for a deep neural network, a computing device and a computer readable storage medium. The method comprises the steps that batch input data is received, the batch input data has multiple dimensions, and the multiple dimensions comprise channel dimensions used for indicating the number of channels of the batch input data; converting the channel dimension of the batch input data into a combination of a group dimension and a sub-channel dimension; transposing the group dimension and the sub-channel dimension to obtain transposed input data of the batch input data; removing the group of dimensions from the transposed input data to obtain dimension-reduced input data of the batch input data; and determining a mean value and a variance of each sub-channel indicated by the sub-channel dimension based on the dimension reduction input data.

Description

technical field [0001] The present invention generally relates to the field of deep neural network computing, and more specifically, relates to a data batch normalization method, a computing device, and a computer-readable storage medium. Background technique [0002] A deep neural network (DNN) is a neural network that contains multiple hidden layers (intermediate layers). At present, deep neural networks have been widely used in speech recognition, image recognition and other fields. In these fields, deep neural networks can be trained using pre-acquired voice samples, image samples, etc., to obtain corresponding trained neural network models. The trained neural network model can be used to recognize new voice data or image data. [0003] The training of a deep neural network is a complex process. Small changes in the front layer of the network will be accumulated and amplified to the subsequent layers, so that the update of the training parameters of the previous layer ...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/08G06F18/211G06F18/214Y02D10/00
Inventor 不公告发明人
Owner SHANGHAI BIREN TECH CO LTD