Neural network model transfer method and system, electronic device, program and medium
A neural network model and network layer technology, applied to biological neural network models, neural architecture, character and pattern recognition, etc., can solve the problems of large storage and computing resources, complex calculations, and large volume, and achieve good performance , storage and computing resource savings
Active Publication Date: 2021-05-04
BEIJING SENSETIME TECH DEV CO LTD
View PDF7 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Although the larger and more complex the network model, the more accurate and reliable the recognition results are, but due to the large volume and complex calculations, it will occupy a large amount of storage and computing resources in practical applications.
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
[0068]Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement, digital expression, and numerical value of the components and steps set forth in these embodiments are not intended to limit the scope of the present invention unless otherwise specified.
[0069]At the same time, it should be understood that in order to facilitate the description, the dimensions of the respective portions shown in the figures are not drawn in accordance with the actual ratio relationship.
[0070]The description of at least one exemplary embodiment is merely illustrative, and it is not necessary to use any limitation of the invention and its application or use.
[0071]For technical, methods, and equipment known to those of ordinary skill in the art may not be discussed in detail, in appropriate, the techniques, methods, and equipment should be considered part of the specification.
[0072]It...
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 embodiment of the present invention discloses a neural network model migration method and system, electronic equipment, programs and media, wherein the method includes: respectively transferring to at least two trained first neural network models and a second neural network to be trained The model inputs the same object to be processed; wherein, the size of the second neural network model is smaller than the size of each of the first neural network models; based on the at least two first neural network models for the object to be processed Processing the result, jointly training the second neural network model until the preset training completion condition is satisfied. The embodiment of the present invention realizes the joint training of the small neural network model based on multiple large neural network models, and can transfer the performance of multiple large neural network models to the small neural network model.
Description
Technical field[0001]The present invention relates to artificial intelligence technologies, especially a neural network model migration method and system, electronic equipment, procedures, and media.Background technique[0002]A general consensus in today's image recognition task is that the larger the network model used under the sample data, the training method is appropriate, the network model design is scientifically, the more complicated, and the more accurate recognition results are often obtained. Although the larger the network model, the more complex, the more accurately reliable, but due to the large amount of volume, complex problems, thereby makes it takes up a lot of storage and computing resources in practical applications.Inventive content[0003]Embodiments of the present invention provide a technical solution for performing neural network model migration.[0004]According to one aspect of the embodiments of the present invention, there is provided a method of migration of...
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 Patents(China)
IPC IPC(8): G06N3/04G06K9/62
CPCG06N3/045G06F18/24
Inventor 邵婧闫俊杰
Owner BEIJING SENSETIME TECH DEV CO LTD