Data processing method and device, processor and computer readable storage medium
A data processing device and data processing technology, applied in the field of neural networks, can solve problems such as low efficiency, large circuit area, and large fitting interval, and achieve the effects of increasing processing speed, improving approximation ability, and reducing overhead
Active Publication Date: 2020-02-14
中昊芯英(杭州)科技有限公司
View PDF8 Cites 4 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Generally speaking, the function value of the nonlinear function fitting in the prior art is generally calculated by a general-purpose processor, which is inefficient
Even if a dedicated fitting circuit is used, it requires a large fitting interval and involves a large number of fitting parameters, resulting in high storage resource overhead and large circuit area, and its data processing speed is difficult to meet the needs of neural networks.
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
[0047] The terms "first", "second" and the like in the description and claims of the present invention and the drawings are used to distinguish different objects, rather than to describe a specific order. The term "at least one" means one or more than one, and the term "plurality" means two or more, unless specifically defined otherwise. Furthermore, the terms "include" and "have", as well as any variations thereof, are intended to cover a non-exclusive inclus...
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
Login to View More Abstract
The invention discloses a data processing method and device, a processor and a computer readable storage medium. According to the invention, a function value mapping circuit and a fitting operation circuit are matched to quickly and efficiently process the data so as to meet the requirement of a neural network for the nonlinearity, thereby improving the approximation capability of the neural network for a complex function. In the prior art, only a fitting operation circuit is used to carry out full-interval fitting operation so as to fit a function value of a specific function. According to the invention, the hardware resource overhead can be reduced by reducing the use of the fitting parameters, and the data processing speed of the data processing device can be increased by reducing the time for matching the fitting parameters by the fitting operation circuit.
Description
technical field [0001] The present invention relates to the field of neural networks, in particular to a data processing method, device, processor and computer-readable storage medium. Background technique [0002] In the deep learning application scenario of neural network, we use neural network to express the complex relationship between input and output. In the language of mathematics, it is to use artificial neural networks to realize complex functions. The neural network is composed of multiple layers of neurons, and each neuron has an activation function. The neuron will receive multiple inputs at the same time and unify them, and then use the activation function to process and output, so that the input and output There is a certain mapping relationship. If the activation function in each neuron is a linear function, the mapping relationship between the input and the output is linear. Then the neural network only outputs multiple inputs through linear combination, tha...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/063
CPCG06N3/063
Inventor 郑瀚寻杨龚轶凡闯小明曾昭睿周远航
Owner 中昊芯英(杭州)科技有限公司