CN109213482A - The graphical application platform of artificial intelligence and application method based on convolutional neural networks - Google Patents
The graphical application platform of artificial intelligence and application method based on convolutional neural networks Download PDFInfo
- Publication number
- CN109213482A CN109213482A CN201810689717.XA CN201810689717A CN109213482A CN 109213482 A CN109213482 A CN 109213482A CN 201810689717 A CN201810689717 A CN 201810689717A CN 109213482 A CN109213482 A CN 109213482A
- Authority
- CN
- China
- Prior art keywords
- model
- artificial intelligence
- function
- module
- unit
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/30—Creation or generation of source code
- G06F8/38—Creation or generation of source code for implementing user interfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
- Stored Programmes (AREA)
Abstract
The invention discloses a kind of graphical application platform of artificial intelligence based on convolutional neural networks and application methods, mainly include functional module, model module and graphic interface；The functional module includes computer vision function unit, natural language processing functional unit, machine learning functional unit, speech identifying function unit, enhancing learning functionality unit；The model module includes data selection unit, model construction unit, model training unit, model measurement unit, model release unit；The graphic interface selects specific function, calls each unit in the model module, the flow chart of dynamic generation artificial intelligence related algorithm by graphic method.The present invention reduces the threshold and cost that user uses artificial intelligence, can be with the various algorithms of the study artificial intelligence of systematization and realization by the use of platform.
Description
Technical field
The invention belongs to artificial intelligence application fields, especially relate to a kind of artificial intelligence based on convolutional neural networks
Graphical application platform and application method.
Background technique
With the rapid development of artificial intelligence technology, artificial intelligence is also more and more in the application of every field, however,
It is all very high using the threshold of artificial intelligent platform and cost at present, other than corresponding brain tactic, complexity in use
Environment configurations, deep learning frame selection, programming language selection the problems such as, and in model construction interminable standard code or
File structure all affects the concrete application of artificial intelligence technology.
Summary of the invention
In view of the above-mentioned problems, the present invention propose a kind of graphical application platform of the artificial intelligence based on convolutional neural networks and
Application method provides image conversion operating platform and operating method for artificial intelligence.
In order to achieve the above objectives, the technical scheme of the present invention is realized as follows:
The graphical application platform of artificial intelligence based on convolutional neural networks, mainly include functional module, model module and
Graphic interface；The functional module includes computer vision function, natural language processing function, machine learning function, voice
Identification function, enhancing learning functionality；The model module include data selection unit, model setting unit, model training unit,
Model measurement unit, model release unit；The graphic interface selects the function in the functional module, then passes through figure
The each unit in the model module, dynamic generation artificial intelligence application flow chart is arranged in change method；
The data selection unit selection is used for the data set of model training；The model setting unit passes through graphical side
Method calls the base components independent assortment building network of neural network；The model training unit is according to the data set of selection and fixed
Justice model be trained, and image conversion show training process with the number of iterations and increased accuracy situation of change；
To corresponding input data, the model treatment completed with training outputs test result the model measurement unit on interface, and
Compared with former input data；Trained model publication is applied in specific application environment by the model release unit.
Further, the graphic method is constructed using Object-oriented Technique, is run based on message-driven.
Further, the base components are one group of modules with input/output end port, and the input port of each module is used
In receiving data, for sending data, the built-in function of each module is then handled data output port；Each intermodule passes through
Line establishes port connection.
Further, the functional module further includes custom feature, and the custom feature passes through graphic method
The artificial intelligence application stream for the function and dynamic generation function of calling each unit dynamic definition in the model module new
Cheng Tu.
Further, the application platform is equipped with cloud platform and is equipped with mobile terminal and the desktop of functional interface with cloud platform
End；The cloud platform is used to receive the data uploaded, model and realizes backstage using cloud computing debugging model and Transfer Parameters
Model training and calculating, and operation result is returned, the desktop end and mobile terminal are provided using the graphic interface of human-computer interaction
Data, Project Realization and interface application, while realizing that cloud computing accesses.
Another aspect of the present invention additionally provides a kind of graphical application method of the artificial intelligence based on convolutional neural networks, packet
It includes:
(1) graphic interface, functional module, model module are constructed；
(2) the corresponding artificial intelligence function in selection function module as needed in graphic interface；
(3) data set of model training is used for according to the different selections of application；
(4) the base components independent assortment building network of neural network is called using graphic method；
(5) be trained according to the model of the data set of selection and definition, image conversion show training process with iteration
The situation of change of number and increased accuracy；
(6) to corresponding input data, the model treatment completed with training outputs test result on interface, and with original
Input data compares；
(7) trained model publication is applied in specific application environment.
Further, the graphic method described in is constructed using Object-oriented Technique, and message-driven is based on
Operation.
Further, functional module described in step (2) includes computer vision function, natural language processing function, machine
Device learning functionality, speech identifying function, enhancing learning functionality；It further include custom feature, the custom feature passes through figure
The artificial intelligence of the new function and dynamic generation function of each unit dynamic definition in model module described in shape method call
It can applicating flow chart.
Further, base components described in step (4) are one group of modules with input/output end port, each module
For receiving data, for sending data, the built-in function of each module is then handled data output port input port；Respectively
Intermodule establishes port connection by line.
Further, the application method is equipped with mobile terminal and the desktop of functional interface by cloud platform and with cloud platform
It realizes at end；The cloud platform is used to receive the data uploaded, model, using cloud computing debugging model and Transfer Parameters, after realization
The model training of platform and calculating, and return to operation result, the desktop end and mobile terminal use the graphic interface of human-computer interaction,
Data, Project Realization and interface application are provided, while realizing that cloud computing accesses.
Compared with prior art, the present invention have it is following the utility model has the advantages that
(1) present invention reduces the threshold and cost that user uses artificial intelligence.It can be with systematization by the use of platform
Learn various algorithms and the realization of artificial intelligence；
(2) present invention can allow user to realize that collection selection, model construction, parameter are selected by graphic interface interaction
It selects, model visualization, model realization, model training, model measurement, export a series of artificial intelligence such as demonstration, model application
Habit and applying step, to make the user do not need to consider complicated video card selection, environment configurations, the selection of deep learning frame, programming
The problems such as speech selection, the building and application of direct implementation model；
(3) present invention realizes pull-alongs to neural network basic module and generates, and corresponding assembly is selected directly to pull combination,
The image conversion that model can be generated indicates.Model and view and different behaviour can be configured in any combination in this way
Make, very flexibly.And the interface operation of user can be converted to the attribute change of model by this mode, thus for figure
The logic of shape file sets up one and does not see still rapidly and accurately immanent structure.
Detailed description of the invention
Fig. 1 is use schematic diagram of the invention.
Specific embodiment
It should be noted that in the absence of conflict, the feature in embodiment and embodiment in the present invention can phase
Mutually combination.
The present invention provides a user the method and system that artificial intelligence is graphically applied, and it is flat using artificial intelligence to reduce user
The threshold and cost of platform.It can be with the various algorithms of the study artificial intelligence of systematization and realization by the use of platform.
The present invention provides the input data set of artificial intelligence study needs, user can be allowed to pass through graphic interface interaction
Realize that collection selection, model construction, parameter selection, model visualization, model realization, model training, model measurement, output are drilled
Show, a series of study of artificial intelligence and the applying step such as model application.To make the user do not need to consider complicated video card selection, ring
The problems such as border configuration, the selection of deep learning frame, programming language selection, the building and application of direct implementation model.
The easy to operate human-computer interaction of image conversion keeps institute functional directly visible by graphically indicating guidance user's operation.
The functional modules such as " model construction ", " model training ", " model measurement " can choose by interface first.
In model construction process, interface can show the infrastructure component of each neural network, such as data layers,
Convolution layers, pooling layers, connection layers of full etc., user can draw and equally freely carry out group to component
It closes, without writing interminable standard code or file structure manually, by the building for drawing network structure implementation model.This image
Changing operating platform realizes pull-alongs generation to these neural network basic modules, selects corresponding assembly directly to pull combination, just
The image conversion that model can be generated indicates.Model and view and different operation can be configured in any combination in this way,
Very flexibly.And the interface operation of user can be converted to the attribute change of model by this mode, to be figure
The logic for changing file sets up one and does not see still rapidly and accurately immanent structure.
Equally, the data input of model training, model measurement etc., parameter assignment, accuracy selection etc. are also by interface phase
The image conversion module answered is realized, goes in corresponding code document to modify without user.It simultaneously can be with real-time display when model training
The number of iterations, accuracy, learning rate, and draw corresponding curvilinear function.
Image conversion operation interface provides the functions such as modules dragged, link block function, assignment, while can timely respond to every
Single stepping bring interior change, if the sequential organization of task changes, the parent-child structure of task changes.
As shown in Figure 1, be functional module area on the left of image conversion operation interface of the present invention, main functional selection, data
The functions such as selection, model construction, model training, model measurement, model publication；Interface center be main viewing area, display model,
Data, training process etc.；Right side is the regulatory region of each functional module, carries out parameter setting for each function.Function selection can
To select the application field of artificial intelligence；Data selection is exactly the data set that model training is used for according to the different selections of application；Mould
Type building can pass through the base components of dragging neural network --- and input layer, convolutional layer, pond layer etc. construct network；Model instruction
White silk can be trained according to the data set of selection and the model of definition, image conversion show training process with the number of iterations and
The situation of change of increased accuracy；Test is exactly the model treatment completed with training, on interface to corresponding input data
Output test result, and compared with former input data；Model publication is exactly that trained model publication is applied to some specifically
Application environment in.
Major function of the invention is provided by function modules ground, including data selection, Construction of A Model, model training, model
The functions such as test.By graphic method call the function in this module library can dynamic product process figure, boundary is realized by PyQt
The design of face GUI.The flow chart of generation saves as txt or json document, and the parameter of corresponding function is resolved to by file io
Value.Function call is realized using python language in backstage, assigns parameter for respective function.After compiler compiles successful connection, by .py
The executable binary code of file generated, it is final to realize graphical corresponding task.
Graphical drag operation of the invention is constructed with Object-oriented Technique, and operation algorithm is based on message-driven
's.In graphic programming platform, element (conv layers, pooling layers, fc layers) is the function with one group of input/output end port
It can module.For receiving data, for sending data, built-in function is then handled data output port input port, even
Line is the middleware for establishing port connection, and the data transmission between element sends message by object and carries out.Element has independent
Property, reusability, the connection of element be by user editor when determine, connected object can be revoked and change at any time
Become, object can realize Dynamic link library at runtime, and this connection is separable.When transmitting data, element is in order
It is automatically activated, can initiatively send data again after execution and activate the execution of subsequent element.
Cv algorithm, nlp algorithm, enhancing learning algorithm, conventional machines learning algorithm are all used as domain model is integrated to be defined on
On platform of the invention, when the functional module of gui interface is selected when the user clicks, interface jumps and is sent to the control on backstage
The module I D of one response feedback message of processing procedure sequence, i.e. user selection.It controls the ID in program parsing message and enters corresponding
Algorithm function module library.When user inputs specific number at interface or pulls control, the concrete function of some algorithm is realized in customization
When, control program is given by messaging list of the interface transmitting comprising relevant parameter, calls respective algorithms after program parsing messaging list
The packaged python function of function modules ground, by compiler compiling file and executes.The graphical operation at interface and each algorithm
Function mutually to call all realized by unified control program.
The functional module of the computer vision model setting includes image classification module, module of target detection, semantic point
Cut module；The functional module of the conventional machines learning model setting includes SVM module, decision tree module；The natural language
The functional module for handling model setting includes coding module, decoder module, attention mechanism module；The enhancing learning model setting
Functional module include AlphaGo module.
It can choose the module of domain model, such as CV module, NLP module, machine into platform interface
Learning module, Reinforcement Learning module.Into after corresponding module, corresponding function mould can choose
Block, such as CV module have image classification function, target detection function, semantic segmentation module.It can be selected under into corresponding module
It selects and can choose SSD algorithm, faster rcnn algorithm, mask rcnn algorithm under corresponding algorithm, such as module of target detection.
Under this view, it can be customized by the user algorithm, can choose data set, building network model, setting model parameter, training
Model, Visualization Model, accuracy of test model etc..Such as the SSD algorithm of the target detection functional module of CV module, user
It can choose input PASCAL VOC data set, learning rate is arranged in customized dragging VGG model on interface, gradient decline is calculated
The parameters such as method, mini-batch size, the number of iterations, selection training later, so that it may see on interface with training pattern
Accuracy variation.In test phase, a test set picture is selected, then the good network of application training on this picture,
Model can detecte out include in picture in classification list higher than certain threshold value object and section out, shown in main interface
Detect the picture completed.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (10)
1. the graphical application platform of artificial intelligence based on convolutional neural networks, which is characterized in that mainly include functional module, mould
Pattern block and graphic interface；The functional module includes computer vision function, natural language processing function, machine learning function
Energy, speech identifying function, enhancing learning functionality；The model module includes data selection unit, model setting unit, model instruction
Practice unit, model measurement unit, model release unit；The graphic interface selects the function in the functional module, then
By graphic method, each unit in the model module, dynamic generation artificial intelligence application flow chart are set；
The data selection unit selection is used for the data set of model training；The model setting unit passes through graphic method tune
With the base components independent assortment building network of neural network；Data set and definition of the model training unit according to selection
Model is trained, and image conversion show training process with the number of iterations and increased accuracy situation of change；It is described
To corresponding input data, the model treatment completed with training outputs test result model measurement unit on interface, and with original
Input data compares；Trained model publication is applied in specific application environment by the model release unit.
2. the graphical application platform of artificial intelligence according to claim 1, which is characterized in that the graphic method is to make
It is constructed with Object-oriented Technique, is run based on message-driven.
3. the graphical application platform of artificial intelligence according to claim 1, which is characterized in that the base components are one group
Module with input/output end port, for receiving data, output port is for sending data, respectively for the input port of each module
The built-in function of module is then handled data；Each intermodule establishes port connection by line.
4. the graphical application platform of artificial intelligence according to claim 1, which is characterized in that the functional module further includes
Custom feature, the custom feature pass through graphic method and call each unit dynamic definition in the model module new
The function and dynamic generation function artificial intelligence application flow chart.
5. the graphical application platform of artificial intelligence according to claim 1, which is characterized in that the application platform is equipped with cloud
Platform and mobile terminal and the desktop end that functional interface is equipped with cloud platform；The cloud platform is used to receive the data uploaded, mould
Type realizes model training and the calculating on backstage using cloud computing debugging model and Transfer Parameters, and returns to operation result, described
Desktop end and mobile terminal use the graphic interface of human-computer interaction, provide data, Project Realization and interface application, while realizing cloud
Calculate access.
6. the graphical application method of artificial intelligence based on convolutional neural networks characterized by comprising
(1) graphic interface, functional module, model module are constructed；
(2) the corresponding artificial intelligence function in selection function module as needed in graphic interface；
(3) data set of model training is used for according to the different selections of application；
(4) the base components independent assortment building network of neural network is called using graphic method；
(5) be trained according to the model of the data set of selection and definition, image conversion show training process with the number of iterations
With the situation of change of increased accuracy；
(6) to corresponding input data, the model treatment completed with training outputs test result on interface, and inputs with original
Data compare；
(7) trained model publication is applied in specific application environment.
7. the graphical application method of artificial intelligence according to claim 6, which is characterized in that described described in step (1)
Graphic method is constructed using Object-oriented Technique, is run based on message-driven.
8. the graphical application method of artificial intelligence according to claim 6, which is characterized in that function described in step (2)
Module includes computer vision function, natural language processing function, machine learning function, speech identifying function, enhancing study function
Energy；It further include custom feature, the custom feature is to pass through graphic method to call each unit in the model module
The artificial intelligence application flow chart of the new function of dynamic definition and the dynamic generation function.
9. the graphical application method of artificial intelligence according to claim 6, which is characterized in that basis described in step (4)
Element is one group of module with input/output end port, and for receiving data, output port is for sending out for the input port of each module
Data are sent, the built-in function of each module is then handled data；Each intermodule establishes port connection by line.
10. the graphical application method of artificial intelligence according to claim 6, which is characterized in that the application method passes through
Cloud platform and mobile terminal and the desktop end realization that functional interface is equipped with cloud platform；The cloud platform is used to receive the number uploaded
Model training and the calculating on backstage are realized using cloud computing debugging model and Transfer Parameters according to, model, and return to operation result,
The desktop end and mobile terminal use the graphic interface of human-computer interaction, provide data, Project Realization and interface application, while real
Existing cloud computing access.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810689717.XA CN109213482A (en) | 2018-06-28 | 2018-06-28 | The graphical application platform of artificial intelligence and application method based on convolutional neural networks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810689717.XA CN109213482A (en) | 2018-06-28 | 2018-06-28 | The graphical application platform of artificial intelligence and application method based on convolutional neural networks |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109213482A true CN109213482A (en) | 2019-01-15 |
Family
ID=64989857
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810689717.XA Pending CN109213482A (en) | 2018-06-28 | 2018-06-28 | The graphical application platform of artificial intelligence and application method based on convolutional neural networks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109213482A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109901824A (en) * | 2019-02-27 | 2019-06-18 | 愚夫智学教育科技(深圳)有限公司 | A kind of data processing method and system based on deep learning |
| CN109948804A (en) * | 2019-03-15 | 2019-06-28 | 北京清瞳时代科技有限公司 | Cross-platform towed deep learning modeling and training method and device |
| CN110334809A (en) * | 2019-07-03 | 2019-10-15 | 成都淞幸科技有限责任公司 | A kind of Component encapsulating method and system of intelligent algorithm |
| CN110363280A (en) * | 2019-09-02 | 2019-10-22 | 国家气象信息中心 | Algorithm model training analysis system |
| CN110531975A (en) * | 2019-08-30 | 2019-12-03 | 陕西思科锐迪网络安全技术有限责任公司 | A kind of deep learning model training method of graphic programming |
| CN110533170A (en) * | 2019-08-30 | 2019-12-03 | 陕西思科锐迪网络安全技术有限责任公司 | A kind of deep learning neural network building method of graphic programming |
| CN110716716A (en) * | 2019-10-12 | 2020-01-21 | 易鸿雁 | Mobile terminal visual AI programming platform |
| CN111027608A (en) * | 2019-11-29 | 2020-04-17 | 蔡尔谦 | Deep learning-based method and device for classification and identification of galaxies |
| CN111339375A (en) * | 2020-03-19 | 2020-06-26 | 中国海洋石油集团有限公司 | Universal big data model configuration and analysis method |
| CN111369011A (en) * | 2020-04-16 | 2020-07-03 | 光际科技(上海)有限公司 | Method and device for applying machine learning model, computer equipment and storage medium |
| CN111667044A (en) * | 2019-03-08 | 2020-09-15 | 北京火星人视野科技有限公司 | Artificial intelligence teaching system based on graphic programming |
| CN112130827A (en) * | 2019-06-25 | 2020-12-25 | 北京启瞳智能科技有限公司 | Model development method and platform based on cloud modularization technology and intelligent terminal |
| CN114461184A (en) * | 2022-04-12 | 2022-05-10 | 飞诺门阵(北京)科技有限公司 | AI application generation method, electronic device, and storage medium |
| CN116578294A (en) * | 2023-04-19 | 2023-08-11 | 哈尔滨工业大学 | Python-based artificial intelligence algorithm tool kit design method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105183445A (en) * | 2015-07-10 | 2015-12-23 | 珠海金山网络游戏科技有限公司 | Visual design system of artificial intelligence of game on the basis of XML (Extensive Markup Language) |
| CN106959851A (en) * | 2017-03-03 | 2017-07-18 | 同济大学 | A kind of Modular programmable distributed interactive system towards artificial intelligence study |
| CN107169575A (en) * | 2017-06-27 | 2017-09-15 | 北京天机数测数据科技有限公司 | A kind of modeling and method for visualizing machine learning training pattern |
| CN107704917A (en) * | 2017-08-24 | 2018-02-16 | 北京理工大学 | A kind of method of effectively training depth convolutional neural networks |
-
2018
- 2018-06-28 CN CN201810689717.XA patent/CN109213482A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105183445A (en) * | 2015-07-10 | 2015-12-23 | 珠海金山网络游戏科技有限公司 | Visual design system of artificial intelligence of game on the basis of XML (Extensive Markup Language) |
| CN106959851A (en) * | 2017-03-03 | 2017-07-18 | 同济大学 | A kind of Modular programmable distributed interactive system towards artificial intelligence study |
| CN107169575A (en) * | 2017-06-27 | 2017-09-15 | 北京天机数测数据科技有限公司 | A kind of modeling and method for visualizing machine learning training pattern |
| CN107704917A (en) * | 2017-08-24 | 2018-02-16 | 北京理工大学 | A kind of method of effectively training depth convolutional neural networks |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109901824A (en) * | 2019-02-27 | 2019-06-18 | 愚夫智学教育科技(深圳)有限公司 | A kind of data processing method and system based on deep learning |
| CN111667044A (en) * | 2019-03-08 | 2020-09-15 | 北京火星人视野科技有限公司 | Artificial intelligence teaching system based on graphic programming |
| CN109948804A (en) * | 2019-03-15 | 2019-06-28 | 北京清瞳时代科技有限公司 | Cross-platform towed deep learning modeling and training method and device |
| CN112130827A (en) * | 2019-06-25 | 2020-12-25 | 北京启瞳智能科技有限公司 | Model development method and platform based on cloud modularization technology and intelligent terminal |
| CN110334809A (en) * | 2019-07-03 | 2019-10-15 | 成都淞幸科技有限责任公司 | A kind of Component encapsulating method and system of intelligent algorithm |
| CN110531975A (en) * | 2019-08-30 | 2019-12-03 | 陕西思科锐迪网络安全技术有限责任公司 | A kind of deep learning model training method of graphic programming |
| CN110533170A (en) * | 2019-08-30 | 2019-12-03 | 陕西思科锐迪网络安全技术有限责任公司 | A kind of deep learning neural network building method of graphic programming |
| CN110363280A (en) * | 2019-09-02 | 2019-10-22 | 国家气象信息中心 | Algorithm model training analysis system |
| CN110716716A (en) * | 2019-10-12 | 2020-01-21 | 易鸿雁 | Mobile terminal visual AI programming platform |
| CN110716716B (en) * | 2019-10-12 | 2023-10-27 | 成都阿加犀智能科技有限公司 | Mobile terminal visual AI programming platform |
| CN111027608A (en) * | 2019-11-29 | 2020-04-17 | 蔡尔谦 | Deep learning-based method and device for classification and identification of galaxies |
| CN111339375A (en) * | 2020-03-19 | 2020-06-26 | 中国海洋石油集团有限公司 | Universal big data model configuration and analysis method |
| CN111369011A (en) * | 2020-04-16 | 2020-07-03 | 光际科技(上海)有限公司 | Method and device for applying machine learning model, computer equipment and storage medium |
| CN114461184A (en) * | 2022-04-12 | 2022-05-10 | 飞诺门阵(北京)科技有限公司 | AI application generation method, electronic device, and storage medium |
| CN114461184B (en) * | 2022-04-12 | 2022-07-01 | 飞诺门阵(北京)科技有限公司 | AI application generation method, electronic device, and storage medium |
| CN116578294A (en) * | 2023-04-19 | 2023-08-11 | 哈尔滨工业大学 | Python-based artificial intelligence algorithm tool kit design method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109213482A (en) | The graphical application platform of artificial intelligence and application method based on convolutional neural networks | |
| CN110322010B (en) | Pulse neural network operation system and method for brain-like intelligence and cognitive computation | |
| US11783227B2 (en) | Method, apparatus, device and readable medium for transfer learning in machine learning | |
| CN108809694B (en) | Service arrangement method, system, device and computer readable storage medium | |
| CN110198237A (en) | A kind of wireless configuration method for being intended to driving network | |
| CN109840111B (en) | Graphical service processing system and method | |
| CA3050159C (en) | Artificial intelligence (ai) based automatic rule generation | |
| CN106021816B (en) | A kind of implementation method of the distributed system behavior simulation analysis tool of Behavior-based control tree | |
| US11645108B2 (en) | Automated semantic tagging | |
| CN113449858A (en) | Processing method of neural network model and related equipment | |
| CN108509185A (en) | System and method for the modeling of arbitrary software logic | |
| CN110750298B (en) | AI model compiling method, equipment and storage medium | |
| CN109816114A (en) | A kind of generation method of machine learning model, device | |
| TWI751387B (en) | Software defined driven ict service provider system based on end to end orchestration | |
| Rkhami et al. | On the use of graph neural networks for virtual network embedding | |
| Pujol-Perich et al. | Ignnition: Bridging the gap between graph neural networks and networking systems | |
| CN109754072A (en) | Processing method, artificial intelligence process device and the Related product of internet off-line model | |
| CN116127899A (en) | Chip design system, method, electronic device, and storage medium | |
| CN106446341A (en) | Process algebra-based real-time protocol analysis and verification system | |
| CN106096159B (en) | A kind of implementation method of distributed system behavior simulation analysis system under cloud platform | |
| CN104978170B (en) | A kind of multi-agent system generation method based on graph-based | |
| Zhang et al. | Interactive AI with retrieval-augmented generation for next generation networking | |
| Lee et al. | Latent L-systems: Transformer-based tree generator | |
| CN106776275B (en) | A kind of testing process automatic generation method based on Packet Multiplexing | |
| US11621988B1 (en) | Adaptable internet-of-things (IoT) computing systems and methods for improved declarative control of streaming data |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190115 |