Memory testing method based on naive Bayes classifier
A Bayesian classifier and memory testing technology, which is applied in faulty hardware testing methods, instruments, and faulty computer hardware detection, can solve problems such as predicting the yield rate of mass production on the production line, and the high yield rate of the production line. , achieving significant progress, reliable design principles, and outstanding substantive features
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0038] like figure 1 As shown, the present invention provides a kind of memory testing method based on Naive Bayesian classifier, comprises the steps:
[0039] S1. Select the memory introduction test item; the memory introduction test item includes RMT test, utilization rate test, performance test and golden finger thickness test;
[0040] S2. Generate training data according to the production line yield corresponding to the test results of the memory-introduced test items;
[0041] S3. Use the training data to train and classify the Naive Bayesian classifier;
[0042] S4. Use the trained naive Bayesian classifier to predict the production line yield of the memory to be introduced;
[0043] S5. Determine whether to import the memory to be imported according to the predicted yield rate of the production line.
Embodiment 2
[0045] like figure 2 As shown, the present invention provides a kind of memory testing method based on Naive Bayesian classifier, comprises the steps:
[0046] S1. Select the memory import test item;
[0047] S2. Generate training data according to the production line yield corresponding to the test results of the memory-introduced test items; the specific steps are as follows:
[0048] S21. Obtain the test result of each memory introduction test item;
[0049] S22. Parameterize the test results of introducing each memory into the test item; the specific steps are as follows:
[0050] S221. Obtain the test standard of each memory introduction test item;
[0051] S222. Take the difference between the test result of each memory introduction test item and its corresponding test standard;
[0052] S23. Parameterize the yield rate of the production line; there are two parameters for the yield rate of the production line, which are less than the standard yield threshold F and g...
Embodiment 3
[0062] The present invention provides a kind of memory test method based on Naive Bayesian classifier, comprises the following steps:
[0063] S1. Select the memory introduction test item; the memory introduction test item includes RMT test, utilization rate test, performance test and golden finger thickness test. The reason for these four test items is that the test results of these four test items can be standardized. Can ignore the influence of mutation factor, can improve the accuracy of the present invention;
[0064] S2. Generate training data according to the production line yield corresponding to the test results of the memory-introduced test items; the specific steps are as follows:
[0065] S21. Obtain the test result of each memory introduction test item;
[0066] S22. Parameterize the test results of introducing each memory into the test item; the specific steps are as follows:
[0067] S221. Obtain the test standard of each memory introduction test item;
[006...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
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