CN110008121A - A kind of personalization test macro and its test method - Google Patents
A kind of personalization test macro and its test method Download PDFInfo
- Publication number
- CN110008121A CN110008121A CN201910209593.5A CN201910209593A CN110008121A CN 110008121 A CN110008121 A CN 110008121A CN 201910209593 A CN201910209593 A CN 201910209593A CN 110008121 A CN110008121 A CN 110008121A
- Authority
- CN
- China
- Prior art keywords
- parameter
- module
- model
- machine learning
- output
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3668—Software testing
- G06F11/3672—Test management
- G06F11/3684—Test management for test design, e.g. generating new test cases
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a kind of personalized test macro and its test methods, including a model creating unit, the model creating unit includes a custom parameter module and a model uploading module, the custom parameter module communicatedly connects the model uploading module, the model uploading module creates an at least machine learning model, the mapping relations of machine learning model described in the custom parameter module editor；An and model measurement unit, the model measurement unit includes a personalized display module and a selecting module, the personality module connects the custom parameter module communicatedly to complete the mapping of machine learning model displaying, and the selecting module connects the personalized display module communicatedly to select one of mapping relations of one of them machine learning model.Since preset parameter can be carried out mapping displaying by the model measurement unit, enables tester to choose desired machine learning model and test.
Description
Technical field
The present invention relates to the communications field, in particular to a kind of personalized test macro and its survey based on machine learning model
Method for testing.
Background technique
Information age arrives, and many industries are risen therewith, can all generate huge data information daily.These data are more
And it is mixed and disorderly, but each is all of crucial importance, valuable information is contained in the inside.Conventional data analysis passes through to prearrange
Good method analyzes data, therefrom excavates valuable information, and big data analysis would not be limited by this, it is direct
From substantial amounts, useful information is analyzed in complicated data, thus make the maximum value of data performance, but this
Process is considerably complicated, is difficult to be quickly obtained information, it is therefore necessary to complete by machine learning model.The target of big data technology
Realize that the development with machine learning is inevitable inseparable.
Therefore, for the development of social every aspect, the design of machine mould is essential.Machine learning is artificial intelligence
Core, be the fundamental way for making computer that there is intelligence, application spreads the every field of artificial intelligence, it is mainly using returning
It receives, integrate rather than deduce.That is, user can be according to some specific when a machine learning model is established
The acquisition and analysis of data, and then obtain a predictable result.
After a machine learning model is developed, especially when multiple developers have shared demand.Different models
Parameter is input into the machine learning model, and wherein the form of expression of each machine learning model is also different, it may be possible to
Picture, video or audio etc..Therefore, when tester tests, a test data, obtained test result are inputted
It is different, even big difference, test process is caused to become extremely complex.Also, the parameter person of different developers
Parameter request it is different, as tester, can not learn the developer of the machine learning model, therefore in that case, survey
Examination person's error-prone.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of personalized test macros, for solving when multiple developers'
The personalized question of measuring stability and test in machine learning model.
In order to solve the above-mentioned technical problem, the technical solution of the present invention is as follows:
A kind of personalization test macro, comprising:
One model creating unit, the model creating unit include a custom parameter module and a model uploading module,
Wherein the custom parameter module communicatedly connects the model uploading module, wherein model uploading module creation is at least
One machine learning model, the mapping relations of machine learning model described in the custom parameter module editor；And
One model measurement unit, the model measurement unit include a personalized display module and a selecting module, wherein
The personality module connects the custom parameter module communicatedly to complete the mapping of machine learning model displaying, institute
It states selecting module and connects the personalized display module communicatedly to select wherein the one of one of them machine learning model
A mapping relations.
Preferably, the custom parameter module includes an at least input element and an at least output element, the input
Element and the output element are connected to the machine learning module to form mapping relations.
Preferably, the custom parameter module further includes editor's element, connects institute to editor's element editable
Input element and the output element are stated to edit the parameter information output and input.
Preferably, the personalized display module further includes a parameter preset element and a show element, wherein described pre-
The custom parameter module is connected to setting parameter element communication, is generated at least according to the mapping relations of the machine learning model
One default output parameter, the show element communicatedly connect the parameter preset element, to show the default output parameter
Shown information.
Preferably, the selecting module receives the machine learning model of selected correspondence mappings relationship after a selected signal
With the input element and the output element, the input element and the output element become test input element at this time
With test output element with etc. it is to be tested.
Preferably, the model measurement unit further includes a parameter prompt module, and the parameter prompt module communicatedly connects
The selecting module is connect to prompt the parameter request of the selected mapping relations, wherein the personalization test macro further includes
One log unit, it is described to record that the log unit communicatedly connects the model measurement unit and the model creating unit
The operating process of personalized test macro.
The present invention also provides a kind of system detection methods, comprising the following steps:
(a) the parameter information feature of input and output is set；
(b) it receives a model file and creates an at least machine learning model, while being connected to the machine learning with mapping
Model and at least an input element and at least an output element；
(c) it shows in at least displaying of the parameter information of an output parameter mapped out comprising preset input parameter
Hold, after one of displaying content is chosen, selectes the input element and the output element of correspondence mappings relationship；With
And
(d) at this time the input element and the output element becomes current test input element and test output member
Part, to wait for testing.
Preferably, wherein in step (b), machine learning model mapping ground connects multiple input elements and multiple outputs
Element or machine learning module mapping ground one input element of connection and an output element.
Preferably, wherein step (c) further includes following steps:
(c1) it after starting the system testing unit, exports preset input parameter and maps corresponding default output ginseng
Number；And
(c2) one of them described default output parameter is being selected, the selected corresponding machine learning model is accordingly reflected
Penetrate the input element and the output element corresponding to relationship.
Preferably, wherein step (d) is further comprising the steps of:
(d1) parameter request required for prompt selected the test input element and the test output element.
By adopting the above technical scheme, since preset parameter can be carried out mapping displaying by the model measurement unit, make
Tester can choose desired machine learning model and test, and can prompt needed for tester during the test
The parameter request wanted.
Detailed description of the invention
Fig. 1 is the system schematic of personalized test macro of the present invention；
Fig. 2 is the process schematic of the model creating unit of personalized test macro of the present invention；
Fig. 3 is a portion process signal of the model measurement unit of personalized test macro of the present invention
Figure；
Fig. 4 is another part mistake of the model measurement unit of the personalized test macro of the present invention of hookup 3
Journey schematic diagram；
Fig. 5 A is one of mapping relations of the machine learning model of personalized test macro of the present invention
Schematic diagram.
Fig. 5 B is showing for another mapping relations of the machine learning model of personalized test macro of the present invention
It is intended to.
Specific embodiment
Specific embodiments of the present invention will be further explained with reference to the accompanying drawing.It should be noted that for
The explanation of these embodiments is used to help understand the present invention, but and does not constitute a limitation of the invention.In addition, disclosed below
The each embodiment of the present invention involved in technical characteristic can be combined with each other as long as they do not conflict with each other.
As shown in Figure 1, the personalization test macro includes a model creating unit 10 and a model measurement unit 20,
Described in model creating unit 10 for creating at least one machine learning model 100, and the model measurement unit 20 is then used
In the test machine learning model 100.Specifically, the model measurement unit 20 of the personalized test macro includes one
Personalized display module 21, the model measurement unit 20 further include a selecting module 22 and a parameter prompt module 23, described
Selecting module 22 is used to select information shown by the default output parameter that the show element 212 is shown.It is wherein described
Personalized test macro further includes a log unit 30, and the log unit 30 communicatedly connects 20 He of model measurement unit
The model creating unit 10 is to record the operating process of the personalized test macro.
As shown in Fig. 2, the model creating unit 10 includes a custom parameter module 11, the custom parameter module
The parameter name that the model from User Exploitation includes is received in 11, and the custom parameter module 11 is according to the parameter name
A parameter name, a parameter display name and a parameter display type can be exported.And when multiple developers are in the custom parameter
After inputting its parameter in module 11, the parameter display name and the parameter display type of final output are difference.In addition,
The personalization display module 21 communicatedly connects the custom parameter module 11, so that passing through the custom parameter module
11 customized can export the model for meeting user's idea.
Therefore, the model creating unit 10 further includes a model uploading module 12, and the model uploading module 12 receives
The model file uploaded from developer, the model uploading module 12 communicatedly connect the custom parameter module
11.The model file includes that a sample database, a sample database categorization module, a characteristic extracting module and a model form module,
Specifically, a preliminary machine learning model 100 can be created that by the model file.
As shown in Figure 3 and Figure 4, the custom parameter module 11 includes an at least input element 111 and at least one output
Element 112, wherein the input element 111 receives input data from the user i.e. parameter name, parameter display name and ginseng
Number display type, the output element 112 output data generated i.e. the parameter name, the parameter display name and institute
Stating parameter display type can be by setting.Specifically, the custom parameter module 11 includes editor's element 113, described
Editor's element 113 communicatedly connects the input element 111 and the output element 112,113 editable of editor's element
The output element 112 is connected, the output data is edited.
It is understood that the input data in the custom parameter module 11 can be in above system
According to parameter name required for the model file that developer oneself is uploaded.And the output data then can be by developer
The information oneself edited, that is to say, that can be parameter name identical with input data, display name and type, be also possible to
Different parameter name, display name and types.
When the model file uploads to the model creating unit 10, the personalization test macro can be surveyed
Examination.That is, the personalization display module 21 is connected to the machine learning model 100.Meanwhile the personalized displaying mould
Block 21 communicatedly connects the custom parameter module 11, so that the personalization display module 21 is shown through the output member
The output data that part 112 generates.
In this course, it is not yet tested.But developer's setting is shown by the personalized display module 21
The machine learning model 100 output data.In test process, the selected one of output data of tester is described
Personalized test macro can improve input data requirement required for the corresponding output data to tester.Therefore, it tests
Person may know that the required test data requirement in current test process.
Specifically, in the first scenario, editor's element 113 communicatedly connects the model uploading module 12, makes
The mapping relations of the machine learning model 100 can be obtained by obtaining editor's element 113, complete 111 He of input element
Mapping relations between the output element 112.
In the latter case, it is set in the custom parameter module 11 by editor's element 113 described defeated
Enter the parameter display name and type of element 111 and the output element 112.In this case, 112 institute of output element is defeated
Parameter display name and type out is defined.
Meanwhile after the model uploading module 12 receives the model file of upload, the model uploading module
12 generate a machine learning model 100 according to the model file.The default input parameter of input one in the input element 111,
And according to the default input parameter, the personalization display module 21 obtains the output parameter, and then forms an individual character
Change and shows information.
The model measurement unit 20 further includes a selecting module 22, and the selecting module 22 is for selecting one of them
Propertyization shows information.After the personalized displaying information is selected, a selected signal of the generation of selecting module 22.Cause
This, the selecting module 22 communicatedly connects the custom parameter module 11, and selectes corresponding to the default output parameter
The machine learning model 100 mapping relations, that is, select the input element 111 and the output element 112.
According to above system content, the personalization test macro includes a setting up procedure, a creation process and a test
Process.Wherein in the setting up procedure, developer is by editor's element 113 to the input element 111 and described defeated
Its parameter information is set separately in element 112 out, and the parameter information includes parameter name, parameter display name and parameter type.Setting
After the completion, parameter request required for the input element 111 and the output element 112 is defined.
During creation, the model uploading module 12 receives the model file, so be created that one it is preliminary
Machine learning model 100, the model uploading module 12 communicatedly connect the input element 111 and the output element 112,
So that the machine learning model 100 is connected to the input element 111 and the output element 112.Therefore, when a parameter is believed
Parameter information when breath input, according to the parsing of the machine learning model 100, after one study of output.
It should be noted that the surface i.e. display name and type of parameter information are by the custom parameter module
11 are limited, and the mapping relations of the parameter information are then limited by the machine learning model 100.
During the test, be the personalized displaying stage first, in this stage in, it is described personalization display module 21 wrap
Include a Prediction Parameters element 211, the default input parameter equipped with developer's setting in the Prediction Parameters element 211.Simultaneously
The personalization display module 21 further includes a show element 212, and the show element 212 is for showing current preset input ginseng
The output parameter in number situation.
Specifically, the parameter preset element 211 communicatedly connects the model uploading module 12, when the model uploads
After the machine learning model 100 is completed in the creation of module 12, the default input parameter of the input of parameter preset element 211
It is input to the machine learning model 100, the output element 112 exports a default output parameter at this time.The output element
112 communicatedly connect the show element 212, and the show element 212 can show that the default output parameter described at this time is shown
The information shown.
As shown in Figure 5A, it is to be understood that when multiple developers develop multiple models, the show element 212
Automatically queue shows information shown by the default output parameter.Each default output parameter corresponds to a machine
The mapping relations of device learning model 100.The selecting module 22 communicatedly connects the custom parameter module 11, is worked as with selected
Mapping relations corresponding to preceding default output parameter, that is, the mapping relations of the machine learning model 100.
As shown in Figure 5 B, in another case, when developer sets multiple outputs in a machine learning model 100
When mapping, that is to say, that set the mapping of multiple input elements 111 and output element 112 in editor's element 113.
The automatically queue of show element 212 shows information shown by the default output parameter.Each default output parameter
Correspond to one of mapping relations of the machine learning model 100.The selecting module 22 is made by oneself described in communicatedly connecting
Adopted parameter module 11, to select mapping relations corresponding to the default output parameter, that is, the machine learning module 100
One of mapping relations.
Meanwhile the parameter prompt module 23 prompts tester to need to improve corresponding parameter request.
According to above system content, the present invention provides a system testing processes, including following below scheme:
Step 1: setting input/output argument information
Editor's element 113 receives parameter information set by developer, so that the input element 111 and described defeated
The surface of the parameter information between element 112 is defined out.
Step 2: creation machine learning model
The model uploading module 12 receives at least model element that developer uploads, and creates a machine according to model element
Device learning model.The model uploading module 12 communicatedly connects the custom parameter module 11, the machine learning simultaneously
The mapping of model 100 ground at least one input element 111 of connection and at least one output element 112, to complete reflecting for data
It penetrates.
Step 3: personalized show
After starting the model measurement unit 20, the parameter preset element 211 communicatedly connects the custom parameter
Module 11, so that default input parameter built-in in the parameter preset element 211 is according to the corresponding machine learning model
100 map one default output parameter of output.
The show element 212 obtains at least one default output parameter, and automatically shows the default output ginseng
Several parameter informations.
Step 4: the type tested needed for selected
The selecting module 22 selectes the parameter information of one of them default output parameter, at this time the parameter prompt module
23 prompt testers input corresponding input parameter request in the input element 111.
Specifically, the selecting module 22 receives a selected signal, communicatedly connects the custom parameter module 11,
And select the mapping relations of the machine learning model 100 corresponding to the default output parameter.It is currently reflected at this point, having selected
Penetrate the input element 111 and the output element 112 of relationship.Meanwhile the parameter prompt module 23 prompts tester to need
It carries out testing required parameter request.
It is noted that in second step, the input element 111 and the output element 112 and the engineering
The mapping relations for practising model 100 are that a machine learning model 100 corresponds to multiple input elements 111 and multiple output elements
112.And in another embodiment, the input element 111 and the output element 112 and the machine learning model
100 mapping relations are the corresponding input element 111 of a machine learning module 100 and an output element 112.
According to above system content and system flow, invention further provides a kind of system detection methods, including
Following steps:
(a) the parameter information feature of input and output is set；
(b) it receives a model file and creates an at least machine learning model, while being connected to the machine learning with mapping
Model and at least an input element and at least an output element；
(c) it shows in at least displaying of the parameter information of an output parameter mapped out comprising preset input parameter
Hold, after one of displaying content is chosen, selectes the input element and the output element of correspondence mappings relationship；With
And
(d) at this time the input element and the output element becomes current test input element and test output member
Part, to wait for testing.
According to above system test method, wherein in step (a), the parameter information feature includes parameter name, and parameter is aobvious
Show name and parameter type.
According to above system test method, wherein in step (b), machine learning model mapping ground connects multiple inputs
Element and multiple output elements.
According to above system test method, wherein in step (b), the machine learning module mapping one input of ground connection
Element and an output element.
According to above system test method, wherein step (c) further includes following steps:
(c1) it after starting the system testing unit, exports preset input parameter and maps corresponding default output ginseng
Number；
(c2) one of them described default output parameter is being selected, the selected corresponding machine learning model is accordingly reflected
Penetrate the input element and the output element corresponding to relationship；
According to above system test method, wherein step (d) is further comprising the steps of:
(d1) parameter request required for prompt selected the test input element and the test output element.
In conjunction with attached drawing, the embodiments of the present invention are described in detail above, but the present invention is not limited to described implementations
Mode.For a person skilled in the art, in the case where not departing from the principle of the invention and spirit, to these embodiments
A variety of change, modification, replacement and modification are carried out, are still fallen in protection scope of the present invention.
Claims (10)
1. a kind of personalization test macro characterized by comprising
One model creating unit, the model creating unit include a custom parameter module and a model uploading module, wherein
The custom parameter module communicates to connect the model uploading module, wherein the model uploading module creates an at least machine
Learning model, the mapping relations of machine learning model described in the custom parameter module editor；And
One model measurement unit, the model measurement unit includes a personalized display module and a selecting module, wherein described
Personality module communicates to connect the custom parameter module to complete the mapping of machine learning model displaying, the selection
Module communicates to connect the personalized display module to select one of mapping of one of them machine learning model to close
System.
2. personalization test macro according to claim 1, which is characterized in that the custom parameter module includes at least
One input element and at least an output element, the input element and the output element are connected to the machine learning module
To form mapping relations.
3. personalization test macro according to claim 2, which is characterized in that the custom parameter module further includes one
Element is edited, connects the input element and the output element to edit and output and input to editor's element editable
Parameter information.
4. personalization test macro according to claim 3, which is characterized in that the personalization display module further includes one
Parameter preset element and a show element, wherein the parameter preset element communication connects the custom parameter module, according to
The mapping relations of the machine learning model generate at least one default output parameter, and the show element communication connection is described default
Parametric device, to show information shown by the default output parameter.
5. personalization test macro according to claim 4, which is characterized in that the selecting module receives a selected signal
The machine learning model of selected correspondence mappings relationship and the input element and the output element afterwards, the at this time input
Element and the output element become test input element and test output element with etc. it is to be tested.
6. personalization test macro according to claim 1-5, which is characterized in that the model measurement unit is also
Including a parameter prompt module, the parameter prompt module communicates to connect the selecting module to prompt the selected mapping to close
The parameter request of system, wherein the personalization test macro further includes a log unit, described in the log unit communication connection
Model measurement unit and the model creating unit are to record the operating process of the personalized test macro.
7. a kind of system detection method, which comprises the following steps:
(a) the parameter information feature of input and output is set；
(b) it receives a model file and creates an at least machine learning model, while being connected to the machine learning model with mapping
An at least input element and at least an output element；
(c) the displaying content for showing at least parameter information of an output parameter mapped out comprising preset input parameter, when
After one of displaying content is chosen, the input element and the output element of correspondence mappings relationship are selected；And
(d) at this time the input element and the output element becomes current test input element and test output element,
To wait for testing.
8. system detection method according to claim 7, which is characterized in that wherein in step (b), the machine learning mould
Type mapping ground connects multiple input elements and multiple output elements or machine learning module mapping ground connects an input member
Part and an output element.
9. system detection method according to claim 8, which is characterized in that wherein step (c) further includes following
Step:
(c1) it after starting the system testing unit, exports preset input parameter and maps corresponding default output parameter；
And
(c2) one of them described default output parameter is being selected, the corresponding mapping for selecting the corresponding machine learning model is closed
The system corresponding input element and the output element.
10. system detection method according to claim 9, which is characterized in that wherein step (d) is further comprising the steps of:
(d1) parameter request required for prompt selected the test input element and the test output element.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910209593.5A CN110008121B (en) | 2019-03-19 | 2019-03-19 | Personalized test system and test method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910209593.5A CN110008121B (en) | 2019-03-19 | 2019-03-19 | Personalized test system and test method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110008121A true CN110008121A (en) | 2019-07-12 |
| CN110008121B CN110008121B (en) | 2022-07-12 |
Family
ID=67167413
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910209593.5A Active CN110008121B (en) | 2019-03-19 | 2019-03-19 | Personalized test system and test method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110008121B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104899141A (en) * | 2015-06-05 | 2015-09-09 | 北京航空航天大学 | Test case selecting and expanding method facing network application system |
| US20160232457A1 (en) * | 2015-02-11 | 2016-08-11 | Skytree, Inc. | User Interface for Unified Data Science Platform Including Management of Models, Experiments, Data Sets, Projects, Actions and Features |
| CN106250986A (en) * | 2015-06-04 | 2016-12-21 | 波音公司 | Advanced analysis base frame for machine learning |
| WO2018005489A1 (en) * | 2016-06-27 | 2018-01-04 | Purepredictive, Inc. | Data quality detection and compensation for machine learning |
| CN108009643A (en) * | 2017-12-15 | 2018-05-08 | 清华大学 | A kind of machine learning algorithm automatic selecting method and system |
| CN108713205A (en) * | 2016-08-22 | 2018-10-26 | 甲骨文国际公司 | System and method for the data type that automatic mapping and data stream environment are used together |
| US20180349256A1 (en) * | 2017-06-01 | 2018-12-06 | Royal Bank Of Canada | System and method for test generation |
| US20190028557A1 (en) * | 2015-08-28 | 2019-01-24 | Ankur MODI | Predictive human behavioral analysis of psychometric features on a computer network |
| CN109272119A (en) * | 2018-09-11 | 2019-01-25 | 杭州祁睿信息技术有限公司 | A kind of user oriented type artificial intelligence system platform based on machine learning model |
| US10191721B1 (en) * | 2017-07-27 | 2019-01-29 | Huma.Ai | Systems and methods for generating functional application designs |
| CN109409533A (en) * | 2018-09-28 | 2019-03-01 | 深圳乐信软件技术有限公司 | A kind of generation method of machine learning model, device, equipment and storage medium |
-
2019
- 2019-03-19 CN CN201910209593.5A patent/CN110008121B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160232457A1 (en) * | 2015-02-11 | 2016-08-11 | Skytree, Inc. | User Interface for Unified Data Science Platform Including Management of Models, Experiments, Data Sets, Projects, Actions and Features |
| CN106250986A (en) * | 2015-06-04 | 2016-12-21 | 波音公司 | Advanced analysis base frame for machine learning |
| CN104899141A (en) * | 2015-06-05 | 2015-09-09 | 北京航空航天大学 | Test case selecting and expanding method facing network application system |
| US20190028557A1 (en) * | 2015-08-28 | 2019-01-24 | Ankur MODI | Predictive human behavioral analysis of psychometric features on a computer network |
| WO2018005489A1 (en) * | 2016-06-27 | 2018-01-04 | Purepredictive, Inc. | Data quality detection and compensation for machine learning |
| CN108713205A (en) * | 2016-08-22 | 2018-10-26 | 甲骨文国际公司 | System and method for the data type that automatic mapping and data stream environment are used together |
| US20180349256A1 (en) * | 2017-06-01 | 2018-12-06 | Royal Bank Of Canada | System and method for test generation |
| US10191721B1 (en) * | 2017-07-27 | 2019-01-29 | Huma.Ai | Systems and methods for generating functional application designs |
| CN108009643A (en) * | 2017-12-15 | 2018-05-08 | 清华大学 | A kind of machine learning algorithm automatic selecting method and system |
| CN109272119A (en) * | 2018-09-11 | 2019-01-25 | 杭州祁睿信息技术有限公司 | A kind of user oriented type artificial intelligence system platform based on machine learning model |
| CN109409533A (en) * | 2018-09-28 | 2019-03-01 | 深圳乐信软件技术有限公司 | A kind of generation method of machine learning model, device, equipment and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| RAMESH MEDAR 等: "Impact of Training and Testing Data Splits on Accuracy of Time Series Forecasting in Machine Learning", 《2017 INTERNATIONAL CONFERENCE ON COMPUTING, COMMUNICATION, CONTROL AND AUTOMATION》 * |
| ZBPJLC: "【干货】Python机器学习项目实战2—模型选择，超参数调整和评估", 《HTTP://WWW.360DOC.COM/CONTENT/18/0531/23/224841_758634862.SHTML》 * |
| 乐朝晖: "基于隐马尔柯夫模型的Web应用防火墙的设计与实现", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110008121B (en) | 2022-07-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101719278B1 (en) | Deep learnig framework and image recognition method for content-based visual image recognition | |
| CN104679658B (en) | The test method and system of WEB page | |
| CN106446412B (en) | Model-based test method for avionics system | |
| US20070061641A1 (en) | Apparatus and method for generating test driver | |
| CN112444731B (en) | Chip testing method and device, processor chip and server | |
| US20080004855A1 (en) | Design support apparatus, design support method, and design support program | |
| CN103885341B (en) | Performance analysis system based on automotive performance simulator and method | |
| CN107451064A (en) | Automatic test approach, device, computer equipment and readable storage medium storing program for executing | |
| US7272753B2 (en) | Methods and apparatus for translating application tests for execution with simulation software tools | |
| CN115061772B (en) | Multi-field simulation model integration method and system | |
| CN112527676A (en) | Model automation test method, device and storage medium | |
| CN108776723B (en) | Test system self-checking adapter connection line generation method, device, equipment and storage medium | |
| CN108776632A (en) | Signal-oriented self-test automatic process generating method, apparatus, equipment and readable storage medium storing program for executing | |
| CN116261091B (en) | Bluetooth testing system and method capable of customizing testing flow | |
| CN110008121A (en) | A kind of personalization test macro and its test method | |
| CN117076316A (en) | Vehicle-mounted application testing method, system, electronic equipment and storage medium | |
| CN111859539A (en) | Finite element automatic attribute and material parameter assigning method based on Tcl or Tk secondary development | |
| CN110221955A (en) | Test report generating means | |
| CN107135402A (en) | A kind of method and device for recognizing TV station's icon | |
| CN109522213A (en) | A kind of test method and device based on customized test script | |
| CN115062571A (en) | Dynamic selection method, system, equipment and computer readable storage medium for data area of integrated circuit device | |
| CN114218078A (en) | UI page testing method, device and equipment and readable storage medium | |
| CN114090459A (en) | Test system and test method based on microkernel | |
| CN106338747A (en) | Lock-losing recapturing test method based on multi-mode multi-band OEM board card | |
| CN114238613B (en) | Method and device for determining mastery degree of knowledge points and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |