Method and device for generating sample data, method and device for training model
A sample data and model technology, applied in the field of data processing, can solve problems such as data imbalance, poor model training accuracy, difficulty in learning small category information, etc., to achieve the effect of avoiding data imbalance and high precision
Active Publication Date: 2020-12-22
GUOXIN YOUE DATA CO LTD
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] However, in related technologies, due to the aimless image segmentation for data acquisition, the problem of data imbalance is serious, which leads to small category information (probably useful information) being overwhelmed by large category information at the sample structure and feature dimensions. Information masking makes it difficult to learn small category information in subsequent semantic segmentation, resulting in poor accuracy of model training
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
[0051] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only It is a part of the embodiments of this application, not all of them. The components of the embodiments of the application generally described and illustrated in the figures herein may be arranged and designed in a variety of different configurations. Accordingly, the following detailed description of the embodiments of the application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without...
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 present application provides a method and device for generating sample data, and a method and device for training a model, wherein the method for generating sample data includes: obtaining a sample picture, which contains multiple target categories; determining that the proportion of distribution in the sample picture is less than The first target category with the first preset ratio, and / or the second target category with a distribution ratio greater than the second preset ratio; traverse the sample image according to the preset window size, and generate slices to be analyzed; according to the preset The filter condition determines the sample data from the slice to be analyzed so that the obtained sample data meets the following conditions: for the case where the first target category is used as the screening basis, the proportion of sample data containing the first target category in the sample data increases; for the second target category When the target category is the screening basis, the proportion of sample data including the second target category in the sample data is reduced. This application avoids the problem of data imbalance, and improves the accuracy of model training by constructing balanced data.
Description
technical field [0001] The present application relates to the technical field of data processing, in particular, to a method and device for generating sample data, and a method and device for training a model. Background technique [0002] For machine learning, especially deep learning, the operation of most algorithms needs to be based on a large amount of sample data. The richness and accuracy of sample data are very important for machine learning. [0003] For example, semantic segmentation based on deep learning needs to use a large amount of sample data to train the neural network model, so that the trained neural network model can obtain better semantic segmentation results. Wherein, the above sample data may include: a large number of sample pictures and pictures obtained by precisely semantically segmenting objects in the sample pictures according to object categories. [0004] Although the amount of data in the above-mentioned sample images is particularly large, ...
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): G06K9/62G06T7/11
CPCG06T7/11G06T2207/20081G06F18/2155
Inventor 刘萌夏珺峥李长升孙源良
Owner GUOXIN YOUE DATA CO LTD