WO2023231526A1 - 基于数字孪生DaaS平台的算法仓库管理方法及*** - Google Patents
基于数字孪生DaaS平台的算法仓库管理方法及*** Download PDFInfo
- Publication number
- WO2023231526A1 WO2023231526A1 PCT/CN2023/082910 CN2023082910W WO2023231526A1 WO 2023231526 A1 WO2023231526 A1 WO 2023231526A1 CN 2023082910 W CN2023082910 W CN 2023082910W WO 2023231526 A1 WO2023231526 A1 WO 2023231526A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- algorithm
- digital twin
- warehouse
- target
- platform
- Prior art date
Links
- 238000007726 management method Methods 0.000 title claims abstract description 107
- 238000000034 method Methods 0.000 claims abstract description 44
- 238000013473 artificial intelligence Methods 0.000 claims description 38
- 230000004048 modification Effects 0.000 claims description 26
- 238000012986 modification Methods 0.000 claims description 26
- 238000003860 storage Methods 0.000 claims description 17
- 238000012795 verification Methods 0.000 claims description 15
- 238000004140 cleaning Methods 0.000 claims description 10
- 230000006870 function Effects 0.000 claims description 10
- 230000008569 process Effects 0.000 abstract description 30
- 238000011161 development Methods 0.000 abstract description 15
- 238000004519 manufacturing process Methods 0.000 description 14
- 238000004891 communication Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- GNFTZDOKVXKIBK-UHFFFAOYSA-N 3-(2-methoxyethoxy)benzohydrazide Chemical compound COCCOC1=CC=CC(C(=O)NN)=C1 GNFTZDOKVXKIBK-UHFFFAOYSA-N 0.000 description 1
- FGUUSXIOTUKUDN-IBGZPJMESA-N C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 Chemical compound C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 FGUUSXIOTUKUDN-IBGZPJMESA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Definitions
- the present invention relates to the field of artificial intelligence technology, and in particular to an algorithmic warehouse management method and system based on the digital twin DaaS platform.
- the main purpose of the present invention is to propose an algorithm warehouse management method and system based on the digital twin DaaS platform, aiming to solve the problem of how to improve algorithm development efficiency and reduce costs.
- the algorithm warehouse management method based on the digital twin DaaS platform includes the following steps:
- the digital twin DaaS platform generates a target algorithm based on the algorithm set and the functional parameters, and sends the target algorithm to the target device.
- the steps of determining the algorithm type and functional parameters according to the algorithm generation instructions, and determining the algorithm set in the algorithm warehouse according to the algorithm type include:
- the algorithm type is compared with the type identification of each algorithm in the algorithm warehouse to determine an algorithm set based on the algorithms whose type identification is the same as the algorithm type.
- the step of generating a target algorithm based on the algorithm set and the functional parameters through the digital twin DaaS platform includes:
- the comparison result is that there is an algorithm in the algorithm set whose similarity is greater than the preset similarity threshold, then according to the The algorithm and the functional parameters generate the target algorithm;
- the artificial intelligence search engine will use the artificial intelligence search engine to search in the algorithm knowledge base connected to the digital twin DaaS platform based on the functional parameters. Search in to obtain an algorithm whose similarity between the functional parameter identifier and the functional parameter is greater than the preset similarity threshold, and generate a target algorithm based on the algorithm and the functional parameter.
- the method includes:
- a verification operation is performed on the modified target algorithm, and the modified target algorithm that passes the verification is sent to the target device.
- the method further includes:
- the algorithm in the algorithm warehouse is optimized through the digital twin DaaS platform, and the algorithm warehouse is optimized through the digital twin DaaS platform.
- the algorithm in performs cleaning operations.
- the steps of optimizing the algorithms in the algorithm warehouse through the digital twin DaaS platform include:
- the optimized algorithm is identified with algorithm type and functional parameters, and stored in the algorithm warehouse according to the algorithm type.
- the step of cleaning the algorithms in the algorithm warehouse through the digital twin DaaS platform includes:
- Each algorithm in the algorithm set with the same type identifier and the same functional parameter identifier is simulated and run, the performance index corresponding to each algorithm is determined, and the algorithm is cleaned according to the performance index.
- the present invention also provides an algorithm warehouse management device based on the digital twin DaaS platform.
- the algorithm warehouse management device based on the digital twin DaaS platform includes:
- a receiving module configured to determine the algorithm class based on the algorithm generation instruction when receiving the algorithm generation instruction sent by the target device.
- Type and function parameters and determine the algorithm set in the algorithm warehouse according to the algorithm type;
- a generation module configured to generate a target algorithm according to the algorithm set and the functional parameters through the digital twin DaaS platform, and send the target algorithm to the target device.
- the receiving module also includes a determining module, the determining module is used for:
- the algorithm type is compared with the type identification of each algorithm in the algorithm warehouse to determine an algorithm set based on the algorithms whose type identification is the same as the algorithm type.
- the generation module is also used to:
- the comparison result is that there is an algorithm in the algorithm set whose similarity is greater than the preset similarity threshold, then generate a target algorithm according to the algorithm and the functional parameters;
- the artificial intelligence search engine will use the artificial intelligence search engine to search in the algorithm knowledge base connected to the digital twin DaaS platform based on the functional parameters. Search in to obtain an algorithm whose similarity between the functional parameter identifier and the functional parameter is greater than the preset similarity threshold, and generate a target algorithm based on the algorithm and the functional parameter.
- the generation module also includes a modification verification module, which is used to:
- a verification operation is performed on the modified target algorithm, and the modified target algorithm that passes the verification is sent to the target device.
- the generation module also includes a management module, which is used to:
- the algorithm in the algorithm warehouse is optimized through the digital twin DaaS platform, and the algorithm warehouse is optimized through the digital twin DaaS platform.
- the algorithm in performs cleaning operations.
- management module is also used to:
- the algorithm stored in the algorithm warehouse is optimized through the digital twin DaaS platform to obtain the optimized algorithm and store it in the algorithm warehouse. ;
- the optimized algorithm is identified with algorithm type and functional parameters, and stored in the algorithm warehouse according to the algorithm type.
- management module is also used to:
- Each algorithm in the algorithm set with the same type identifier and the same functional parameter identifier is simulated and run, the performance index corresponding to each algorithm is determined, and the algorithm is cleaned according to the performance index.
- the present invention also provides an algorithm warehouse management system based on the digital twin DaaS platform.
- the algorithm warehouse management system based on the digital twin DaaS platform includes: a memory, a processor and an algorithm stored on the memory.
- An algorithm warehouse management program can be run on the processor. When the algorithm warehouse management program is executed by the processor, the steps of the algorithm warehouse management method based on the digital twin DaaS platform are implemented as described above.
- the present invention also provides a computer-readable storage medium.
- the readable storage medium is a computer-readable storage medium.
- An algorithm warehouse management program is stored on the readable storage medium. The algorithm When the warehouse management program is executed by the processor, the steps of the algorithmic warehouse management method based on the digital twin DaaS platform are implemented as described above.
- the algorithm warehouse management method based on the digital twin DaaS platform proposed by the present invention when receiving the algorithm generation instruction sent by the target device, determines the algorithm type and functional parameters according to the algorithm generation instruction, and determines them in the algorithm warehouse according to the algorithm type. Algorithm set; the digital twin DaaS platform generates a target algorithm according to the algorithm set and the functional parameters, and sends the target algorithm to the target device; the present invention selects the corresponding algorithm in the algorithm warehouse according to the algorithm type, And based on the algorithm and functional parameters, the target algorithm is generated without manual participation in the algorithm development process, thereby improving the efficiency of algorithm development and reducing costs.
- Figure 1 is a schematic diagram of the equipment structure of the hardware operating environment involved in the embodiment of the present invention.
- Figure 2 is a schematic flow chart of the first embodiment of the algorithmic warehouse management method based on the digital twin DaaS platform of the present invention.
- Figure 1 is a schematic diagram of the equipment structure of the hardware operating environment involved in the embodiment of the present invention.
- the device in this embodiment of the present invention may be a PC or a server device.
- the device may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002.
- the communication bus 1002 is used to realize connection communication between these components.
- the user interface 1003 may include a display screen (Display) and an input unit such as a keyboard (Keyboard).
- the optional user interface 1003 may also include a standard wired interface and a wireless interface.
- the network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).
- the memory 1005 can be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory.
- the memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.
- the device structure shown in Figure 1 does not constitute a limitation of the device, and may include more or fewer components than shown, or combine certain components, or arrange different components.
- memory 1005 which is a computer storage medium, may include an operating system, a network communication module, a user interface module, and an algorithm warehouse management program.
- the operating system is a program that manages and controls portable storage devices and software resources, and supports the operation of network communication modules, user interface modules, algorithm warehouse management programs, and other programs or software;
- the network communication module is used to manage and control the network interface 1002;
- the user interface module is used to manage and control the user interface 1003.
- the storage device calls the algorithm warehouse management program stored in the memory 1005 through the processor 1001, and performs the following operations in various embodiments of the algorithm warehouse management method based on the digital twin DaaS platform.
- Figure 2 is a schematic flow chart of the first embodiment of the algorithmic warehouse management method based on the digital twin DaaS platform of the present invention.
- the method includes:
- Step S10 when receiving the algorithm generation instruction sent by the target device, determine the algorithm type and functional parameters according to the algorithm generation instruction, and determine the algorithm set in the algorithm warehouse according to the algorithm type;
- Step S20 Generate a target algorithm based on the algorithm set and the functional parameters through the digital twin DaaS platform, and send the target algorithm to the target device.
- the algorithmic warehouse management method based on the digital twin DaaS platform is applied to the algorithmic warehouse management system based on the digital twin DaaS platform.
- the algorithmic warehouse management system based on the digital twin DaaS platform is implemented with the artificial intelligence Internet of Things platform and the digital twin DaaS platform.
- the algorithm warehouse management system based on the digital twin DaaS platform will be referred to as the management system below, and the management system will be used as an example to illustrate; when the management system receives the algorithm generation instruction sent by the target device, it obtains the algorithm generation user product needs in the directive, and based on the user product needs request to determine the algorithm type and functional parameters; the management system compares the algorithm type with the type identification of each algorithm in the algorithm warehouse to determine the algorithm set based on the algorithm with the same type identification and algorithm type; the management system obtains it through the digital twin DaaS platform The functional parameter identifier corresponding to each algorithm in the algorithm set, and calculate the similarity between the functional parameter identifier corresponding to each algorithm and the functional parameter; compare the similarity with the preset similarity threshold to obtain the comparison result; if the comparison result is If there is an algorithm with a similarity greater than the preset similarity threshold in the algorithm set, the target algorithm will be generated based on the algorithm and functional parameters; if the comparison result is that there is
- the digital twin DaaS platform is the AIOTDaaS digital twin platform, which includes an interconnected business middle platform and a data middle platform; the data middle platform is used to collect, calculate, store and process data collected through AIOTDaaS to form The standard data is stored on the one hand and transmitted to the business middle platform on the other hand; the business middle platform is used to combine the standard data transmitted based on the data middle platform with industry applications to form models and products for industry applications, so that users can use the business middle platform based on Quickly package business products.
- the artificial intelligence IoT platform is the AIOT PaaS IoT platform, which includes AI edge computing, digital chips, edge storage computing and module technology.
- the algorithm warehouse management method based on the digital twin DaaS platform in this embodiment when receiving the algorithm generation instruction sent by the target device, determines the algorithm type and functional parameters according to the algorithm generation instruction, and determines the algorithm set in the algorithm warehouse according to the algorithm type; Through the digital twin DaaS platform, the target algorithm is generated based on the algorithm set and functional parameters, and the target algorithm is sent to the target device; the present invention selects the corresponding algorithm in the algorithm warehouse based on the algorithm type, and generates the target algorithm based on the algorithm and functional parameters. There is no need for manual participation in the algorithm development process, which improves algorithm development efficiency and reduces costs.
- Step S10 when receiving the algorithm generation instruction sent by the target device, determine the algorithm type and functional parameters according to the algorithm generation instruction, and determine the algorithm set in the algorithm warehouse according to the algorithm type;
- the management system communicates with the artificial intelligence Internet of Things platform.
- the intelligent terminal equipment communicates with the artificial intelligence Internet of Things platform, and the intelligent terminal equipment is communicated with the artificial intelligence Internet of Things platform.
- the terminal device serves as the target device and sends algorithm generation instructions to the artificial intelligence IoT platform through the target device.
- the management system receives the algorithm generation instructions forwarded by the artificial intelligence IoT platform, determines the algorithm type and functional parameters according to the algorithm generation instructions, and determines the algorithm type according to the algorithm type.
- the algorithm type is the application field of the algorithm, such as face recognition, industrial manufacturing, medical diagnosis, autonomous driving, intelligent transportation, etc.
- functional parameters include business processes, technological processes, and application scenarios. etc.
- the algorithm warehouse is established in the management system in advance, and the algorithm warehouse contains different Algorithms of different types and functions, and each algorithm in the algorithm warehouse is identified with the algorithm type and functional parameters, which facilitates the management of algorithms in the algorithm warehouse.
- the steps of determining the algorithm type and functional parameters according to the algorithm generation instructions, and determining the algorithm set in the algorithm warehouse according to the algorithm type include:
- Step a Obtain the user product requirements in the algorithm generation instructions, and determine the algorithm type and functional parameters based on the user product requirements;
- relevant developers when relevant developers send algorithm generation instructions through the target device, they determine the algorithm type and functional parameters required by the target device according to their own needs, generate corresponding user product requirements based on the algorithm type and functional parameters, and generate algorithm generation Instructions are sent to the artificial intelligence IoT platform connected to the management system through the target device.
- the management system obtains the algorithm generation instructions in the artificial intelligence IoT platform and determines the relevant requirements based on the user product requirements in the algorithm generation instructions. Algorithm type and function parameters required by developers.
- Step b Compare the algorithm type with the type identifier of each algorithm in the algorithm warehouse to determine an algorithm set based on the algorithms whose type identifier is the same as the algorithm type.
- the management system compares the algorithm type with the type identification of each algorithm in the algorithm warehouse in order to obtain the comparison result.
- the obtained comparison result determines the algorithm set for algorithms with the same type identification and algorithm type. ; It is understandable that there are a large number of different types of algorithms stored in the algorithm warehouse.
- the management system first selects algorithms from all types of algorithms with the same type identifier as the algorithm type required by the relevant developers, which can improve the selection efficiency and thus help To improve the efficiency of algorithm development.
- Step S20 Generate a target algorithm based on the algorithm set and the functional parameters through the digital twin DaaS platform, and send the target algorithm to the target device.
- the management system after determining the algorithm set, obtains the functional parameter identifier corresponding to each algorithm in the algorithm set through the digital twin DaaS platform, and compares the functional parameter identifier corresponding to each algorithm with the functional parameter to obtain the comparison The results are combined with the comparison results and functional parameters to generate a target algorithm and send the target algorithm to the target device.
- step S20 includes:
- Step c Obtain the functional parameter identifier corresponding to each algorithm in the algorithm set through the digital twin DaaS platform, and calculate the similarity between the functional parameter identifier corresponding to each algorithm and the functional parameter;
- the management system obtains the functional parameter identifier corresponding to each algorithm in the algorithm set through the digital twin DaaS platform, and calculates the similarity between the functional parameter identifier corresponding to each algorithm and the functional parameters required by the relevant developers.
- the management system obtains a similarity of 100%, that is, the functional parameter identification existing in the algorithm warehouse is exactly the same as the functional parameter required by the relevant developer. At this time, the management system can directly store the functional parameters. Calculate method as the target algorithm and sent directly to the target device.
- Step d Compare the similarity with a preset similarity threshold to obtain a comparison result
- the management system compares the similarity between the functional parameter identifier and the functional parameter corresponding to each algorithm with the preset similarity threshold. Get comparison results.
- Step e if the comparison result is that there is an algorithm in the algorithm set whose similarity is greater than the preset similarity threshold, generate a target algorithm according to the algorithm and the functional parameters;
- the management system In this step, if the management system obtains a comparison result that there is an algorithm in the algorithm set with a similarity greater than the preset similarity threshold, it will generate the target algorithm based on the algorithm and functional parameters; it can be understood that the similarity is greater than the preset similarity threshold. There may be one or more algorithms.
- the management system combines all algorithms with similarities greater than the preset similarity threshold with functional parameters to generate the target algorithm.
- Step f if the comparison result is that there is no algorithm in the algorithm set whose similarity is greater than the preset similarity threshold, then the artificial intelligence search engine will use the artificial intelligence search engine to search for the algorithm connected to the digital twin DaaS platform based on the functional parameters. Search the algorithm knowledge base to obtain an algorithm whose similarity between the functional parameter identifier and the functional parameter is greater than the preset similarity threshold, and generate a target algorithm based on the algorithm and the functional parameter.
- the management system obtains a comparison result that there is no algorithm in the algorithm set with a similarity greater than the preset similarity threshold, it will call the artificial intelligence search engine in the digital twin DaaS platform and compare it with the digital twin DaaS based on the functional parameters.
- the algorithm knowledge base exists in On the Internet, when the management system cannot search for the corresponding algorithm in its own algorithm warehouse, it searches in the algorithm knowledge base on the Internet through the artificial intelligence search engine, and obtains that the similarity between the functional parameter identifier and the functional parameter is greater than the preset similarity. degree threshold algorithm, and generate the target algorithm based on the algorithm and functional parameters; it can be understood that the functional parameters include business processes, technological processes, application scenarios, etc., and the management system determines that the functional parameter identification of the algorithm is the function required by the relevant developers. The parameters are extremely similar. Therefore, the management system can quickly obtain the target algorithm by modifying and adjusting the functional parameters of the algorithm according to the functional parameters required by the relevant developers.
- the management system can be applied in industries such as industrial Internet, smart medical care, smart supply chain, smart finance, smart agriculture, smart communities, smart parks, smart transportation, etc.
- relevant developers of the industrial Internet need to When developing algorithms for a certain production equipment in the Industrial Internet, the relevant developers connect the production equipment to the artificial intelligence IoT platform in the management system, and send algorithm generation instructions to the artificial intelligence IoT platform through the production equipment.
- the management system obtains the process flow corresponding to the production equipment from the algorithm generation instructions, and based on the Process flow, search for the corresponding algorithm in the algorithm warehouse through the digital twin DaaS platform, and calculate the similarity between the process flow corresponding to the algorithm and the process flow of the production equipment.
- the algorithm is directly used as the algorithm of the production equipment, and the algorithm is sent to the artificial intelligence IoT platform through the digital twin DaaS platform, and then the artificial intelligence IoT platform sends it to the production equipment, and the production equipment Run according to the algorithm and execute the corresponding process flow to produce the corresponding product; when there is no process flow corresponding to the algorithm in the algorithm warehouse and the similarity between the process flow and the process flow of the production equipment is 100%, the algorithm with the highest similarity is selected, and Use the artificial intelligence search engine of the digital twin DaaS platform to search for other similar algorithms in the algorithm knowledge base, fuse and derive the algorithm set to obtain an algorithm that conforms to the process flow of the production equipment, and send the algorithm through the digital twin DaaS platform to the artificial intelligence IoT platform, and then sent to the production equipment by the artificial intelligence IoT platform.
- the production equipment runs according to the algorithm and executes the corresponding process to produce the corresponding products.
- the management system of this embodiment receives the algorithm generation instruction sent by the target device, it obtains the user product requirements in the algorithm generation instruction, and determines the algorithm type and functional parameters according to the user product requirements; the management system compares the algorithm type with the algorithm stored in the algorithm warehouse.
- the management system obtains the functional parameter identifier corresponding to each algorithm in the algorithm set through the digital twin DaaS platform, and calculates each algorithm The similarity between the corresponding functional parameter identifier and the functional parameter; compare the similarity with the preset similarity threshold to obtain the comparison result; if the comparison result is that there is an algorithm in the algorithm set with a similarity greater than the preset similarity threshold, the algorithm will and functional parameters to generate the target algorithm; if the comparison result is that there is no algorithm with a similarity greater than the preset similarity threshold in the algorithm set, the artificial intelligence search engine will use the artificial intelligence search engine to search the algorithm knowledge base connected to the digital twin DaaS platform based on the functional parameters.
- Search is performed to obtain an algorithm whose similarity between the functional parameter identifier and the functional parameter is greater than the preset similarity threshold, and the target algorithm is generated based on the algorithm and functional parameters, eliminating the need for manual participation in the algorithm development process, thereby improving algorithm development efficiency and reducing costs.
- the difference between the second embodiment of the algorithmic warehouse management method based on the digital twin DaaS platform and the first embodiment of the algorithmic warehouse management method based on the digital twin DaaS platform is that after step S20, the algorithmic warehouse management method based on the digital twin DaaS platform include:
- Step g When receiving the algorithm modification instruction sent by the target device, modify the target device according to the algorithm modification instruction.
- the algorithm is modified;
- Step h Perform a verification operation on the modified target algorithm, and send the modified target algorithm that passes the verification to the target device.
- the target device after the management system sends the target algorithm to the target device, the target device will simulate the target algorithm to verify whether the function and syntax of the target algorithm meet the requirements.
- the target algorithm can meet the target
- the target device determines that the target algorithm does not meet the requirements, it will generate an algorithm modification instruction based on the simulation operation results and send the algorithm modification instruction to the artificial intelligence Internet of Things platform that is connected to the management system.
- the algorithm modification instruction contains modification requirements.
- the modification requirements include the parts and reasons that the target device detects in the target algorithm that do not meet the requirements.
- the management system obtains the modification requirements in the algorithm modification instructions and selects the algorithm that meets the modification requirements in the algorithm warehouse.
- the algorithm is integrated with the target algorithm to modify the target algorithm corresponding to the target device, and the modified target algorithm is verified, and then the modified target algorithm that passes the verification is sent to the target device. It can be understood that in the process of integrating the algorithm that meets the modification requirements with the target algorithm, generally speaking, it is only necessary to replace the part of the algorithm that meets the modification requirements with the part of the target algorithm that does not meet the requirements of the target device. That’s it.
- the management system in this embodiment When the management system in this embodiment receives the algorithm modification instruction sent by the target device, it modifies the target algorithm according to the algorithm modification instruction; performs a verification operation on the modified target algorithm, and sends the modified target algorithm that passes the verification
- the algorithm is sent to the target device. Modify the target algorithm according to the algorithm modification instructions of the algorithm warehouse and the target device, so that the algorithm modification process does not require manual intervention, which can further improve the efficiency of algorithm development and reduce costs.
- Algorithmic warehouse management methods also include:
- Step i When the algorithm generation instruction sent by the target device is not received within the preset time period, the algorithm in the algorithm warehouse is optimized through the digital twin DaaS platform, and the algorithm in the algorithm warehouse is optimized through the digital twin DaaS platform. Perform cleaning operations on the algorithms in the algorithm warehouse.
- the management system when the management system does not receive the algorithm generation instructions sent by the target device within the preset time period, the management system performs optimization operations on the algorithms in the algorithm warehouse through the digital twin DaaS platform, and performs optimization operations on the algorithm warehouse through the digital twin DaaS platform.
- the algorithm in the algorithm is used for cleaning operations; it is understandable that when relevant developers need to develop algorithms for smart terminal devices, the smart terminal device is communicated and connected with the artificial intelligence Internet of Things platform, and the smart terminal device is used as the target device. And send algorithm generation instructions to the artificial intelligence IoT platform through the target device, and the management system will Assume that no algorithm generation instructions sent by the target device are received within the time period, that is, no target device is connected to the artificial intelligence IoT platform within the preset time period.
- the steps of optimizing the algorithms in the algorithm warehouse through the digital twin DaaS platform include:
- Step i1 Optimize the algorithm stored in the algorithm warehouse through the digital twin DaaS platform to obtain an optimized algorithm
- the management system when the management system does not receive the algorithm generation instruction sent by the target device within the preset time period, it optimizes the algorithm stored in the algorithm warehouse through the digital twin DaaS platform to obtain the optimized algorithm. It is understandable Yes, when the management system optimizes the algorithms stored in the algorithm warehouse through the digital twin DaaS platform, it can combine an algorithm in the algorithm warehouse with an application scenario to derive a new algorithm. It can combine certain algorithms in the algorithm warehouse. Algorithms are fused to derive new algorithms. You can also use artificial intelligence search engines to search in different algorithm libraries and third-party platforms that are connected to the digital twin DaaS platform. You can also search for algorithms that are not stored in the algorithm warehouse. You can also search for algorithms in the algorithm warehouse. Some of the algorithms in it are fused with algorithms obtained from artificial intelligence search engines to derive new algorithms.
- Step i2 Identify the algorithm type and functional parameters of the optimized algorithm, and store them in the algorithm warehouse according to the algorithm type.
- the management system identifies the algorithm type and functional parameters of the optimized algorithm, and stores them in the algorithm warehouse according to the algorithm type. For example, if the algorithm type of the optimized algorithm is identified as face recognition, it is summarized Stored in the algorithm collection of the face recognition category, the functional parameters of the optimized algorithms are identified, which allows the algorithm warehouse or digital twin DaaS platform to quickly determine whether the functional parameters of each algorithm are business processes, technological processes, or application scenarios.
- the steps of cleaning the algorithms in the algorithm warehouse through the digital twin DaaS platform include:
- Step i3 Obtain the type identifier and functional parameter identifier of each algorithm in the algorithm warehouse through the digital twin DaaS platform, compare the type identifier of each algorithm, and compare the functional parameter identifier of each algorithm, To obtain a set of algorithms with the same type identifier and the same function parameter identifier;
- Step i4 Simulate each algorithm in the algorithm set with the same type identifier and the same functional parameter identifier, determine the performance index corresponding to each algorithm, and clean the algorithm according to the performance index.
- steps i3 to i4 when the management system does not receive the algorithm generation instruction sent by the target device within the preset time period, it obtains the type identification and functional parameter identification of each algorithm in the algorithm warehouse through the digital twin DaaS platform, and Compare the type identification of each algorithm and compare the functional parameter identification of each algorithm to obtain the type identification.
- the algorithm in the algorithm warehouse is optimized through the digital twin DaaS platform, and the algorithm in the algorithm warehouse is optimized through the digital twin DaaS platform.
- Algorithm performs cleaning operations. Reducing unnecessary storage in the algorithm warehouse makes the algorithm warehouse concise, thereby enabling the management system to quickly determine the target algorithm based on the algorithms in the algorithm warehouse, which is beneficial to improving the efficiency of algorithm development.
- the present invention also provides an algorithm warehouse management device based on the digital twin DaaS platform.
- the algorithm warehouse management device based on the digital twin DaaS platform of the present invention includes:
- a receiving module configured to determine the algorithm type and functional parameters according to the algorithm generation instruction when receiving the algorithm generation instruction sent by the target device, and determine the algorithm set in the algorithm warehouse according to the algorithm type;
- a generation module configured to generate a target algorithm according to the algorithm set and the functional parameters through the digital twin DaaS platform, and send the target algorithm to the target device.
- the receiving module also includes a determining module, the determining module is used for:
- the algorithm type is compared with the type identification of each algorithm in the algorithm warehouse to determine an algorithm set based on the algorithms whose type identification is the same as the algorithm type.
- the generation module is also used to:
- the comparison result is that there is an algorithm in the algorithm set whose similarity is greater than the preset similarity threshold, then generate a target algorithm according to the algorithm and the functional parameters;
- the artificial intelligence search engine will use the artificial intelligence search engine to search in the algorithm knowledge base connected to the digital twin DaaS platform based on the functional parameters. Search in to obtain an algorithm whose similarity between the functional parameter identifier and the functional parameter is greater than the preset similarity threshold, and generate a target algorithm based on the algorithm and the functional parameter.
- the generation module also includes a modification verification module, which is used to:
- a verification operation is performed on the modified target algorithm, and the modified target algorithm that passes the verification is sent to the target device.
- the generation module also includes a management module, which is used to:
- the algorithm in the algorithm warehouse is optimized through the digital twin DaaS platform, and the algorithm warehouse is optimized through the digital twin DaaS platform.
- the algorithm in performs cleaning operations.
- management module is also used to:
- the optimized algorithm is identified with algorithm type and functional parameters, and stored in the algorithm warehouse according to the algorithm type.
- management module is also used to:
- Each algorithm in the algorithm set with the same type identifier and the same functional parameter identifier is simulated and run, the performance index corresponding to each algorithm is determined, and the algorithm is cleaned according to the performance index.
- the present invention also provides an algorithm warehouse management system based on the digital twin DaaS platform.
- the algorithm warehouse management system based on the digital twin DaaS platform includes: a memory, a processor, and an algorithm warehouse management program stored on the memory and executable on the processor.
- the algorithm warehouse management program is executed by the processor. The steps to implement the algorithmic warehouse management method based on the digital twin DaaS platform as mentioned above.
- the method implemented when the algorithm warehouse management program running on the processor is executed may refer to the various embodiments of the algorithm warehouse management method based on the digital twin DaaS platform of the present invention, and will not be described again here.
- the invention also provides a computer-readable storage medium.
- the computer-readable storage medium stores an algorithmic warehouse management program.
- the algorithmic warehouse management program is executed by the processor, the steps of the algorithmic warehouse management method based on the digital twin DaaS platform are implemented as described above.
- the method implemented when the algorithm warehouse management program running on the processor is executed may refer to the various embodiments of the algorithm warehouse management method based on the digital twin DaaS platform of the present invention, and will not be described again here.
- the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation.
- the technical solution of the present invention can be embodied in the form of a software product that is essentially or contributes to the existing technology.
- the computer software product is stored in a storage medium (such as ROM/RAM) as mentioned above. , magnetic disk, optical disk), including several instructions to cause a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the method described in various embodiments of the present invention.
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Development Economics (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
本发明公开了一种基于数字孪生DaaS平台的算法仓库管理方法及***，该方法包括：在接收目标设备发送的算法生成指令时，根据所述算法生成指令确定算法类型和功能参数，并根据所述算法类型在算法仓库中确定算法集合；通过数字孪生DaaS平台根据所述算法集合和所述功能参数，生成目标算法，并将所述目标算法发送到所述目标设备；本发明根据算法类型在算法仓库中选择对应的算法，并根据算法和功能参数，生成目标算法，无需人工参与算法的开发过程，进而提高了算法开发效率。
Description
本发明涉及人工智能技术领域，尤其涉及基于数字孪生DaaS平台的算法仓库管理方法及***。
目前的各个行业中，都需要用到各种智能设备，各种智能设备都需要不同的算法处理对应的事务，但是目前智能设备的算法都是通过人工进行开发的，这导致算法开发效率较低，人力物力成本高。因此，如何提高算法开发效率，降低成本是急需解决的问题。
发明内容
本发明的主要目的在于提出一种基于数字孪生DaaS平台的算法仓库管理方法及***，旨在解决如何提高算法开发效率，降低成本的问题。
为实现上述目的，本发明提供一种基于数字孪生DaaS平台的算法仓库管理方法，所述基于数字孪生DaaS平台的算法仓库管理方法包括如下步骤：
在接收目标设备发送的算法生成指令时，根据所述算法生成指令确定算法类型和功能参数，并根据所述算法类型在算法仓库中确定算法集合；
通过数字孪生DaaS平台根据所述算法集合和所述功能参数，生成目标算法，并将所述目标算法发送到所述目标设备。
可选地，根据所述算法生成指令确定算法类型和功能参数，并根据所述算法类型在算法仓库中确定算法集合的步骤包括：
获取所述算法生成指令中的用户产品需求，并根据所述用户产品需求确定算法类型和功能参数；
将所述算法类型与算法仓库中的每个算法的类型标识进行对比，以根据所述类型标识与所述算法类型相同的算法确定算法集合。
可选地，通过数字孪生DaaS平台根据所述算法集合和所述功能参数，生成目标算法的步骤包括：
通过数字孪生DaaS平台获取所述算法集合中的每个算法对应的功能参数标识，并计算每个算法对应的功能参数标识与所述功能参数的相似度；
将所述相似度与预设相似度阈值进行对比，得到对比结果；
若所述对比结果为所述算法集合中存在所述相似度大于预设相似度阈值的算法，则根据所述
算法和所述功能参数生成目标算法；
若所述对比结果为所述算法集合中不存在所述相似度大于预设相似度阈值的算法，则通过人工智能搜索引擎根据所述功能参数在与所述数字孪生DaaS平台连接的算法知识库中进行搜索，得到功能参数标识与所述功能参数的相似度大于预设相似度阈值的算法，并根据所述算法和所述功能参数生成目标算法。
可选地，将所述目标算法发送到所述目标设备的步骤之后，包括：
在接收到所述目标设备发送的算法修改指令时，根据所述算法修改指令对所述目标算法进行修改；
对修改后的目标算法进行验证操作，并将通过验证的修改后的目标算法发送到所述目标设备。
可选地，将所述目标算法发送到所述目标设备的步骤之后，还包括：
在预设时间段内未接收到目标设备发送的算法生成指令时，通过所述数字孪生DaaS平台对所述算法仓库中的算法进行优化操作，并通过所述数字孪生DaaS平台对所述算法仓库中的算法进行清理操作。
可选地，通过数字孪生DaaS平台对所述算法仓库中的算法进行优化操作的步骤包括：
通过所述数字孪生DaaS平台对所述算法仓库中储存的算法进行优化，以得到优化后的算法；
对优化后的算法进行算法类型和功能参数的标识操作，并根据算法类型存储在所述算法仓库中。
可选地，通过所述数字孪生DaaS平台对所述算法仓库中的算法进行清理操作的步骤包括：
通过所述数字孪生DaaS平台获取所述算法仓库中的每个算法的类型标识和功能参数标识，将每个算法的类型标识进行对比，并将每个算法的功能参数标识进行对比，以得到类型标识相同且功能参数标识也相同的算法集合；
将类型标识相同且功能参数标识也相同的算法集合中的每个算法进行模拟运行，确定每个算法对应的性能指标，并根据所述性能指标对算法进行清理操作。
此外，为实现上述目的，本发明还提供一种基于数字孪生DaaS平台的算法仓库管理装置，所述基于数字孪生DaaS平台的算法仓库管理装置包括：
接收模块，用于在接收目标设备发送的算法生成指令时，根据所述算法生成指令确定算法类
型和功能参数，并根据所述算法类型在算法仓库中确定算法集合；
生成模块，用于通过数字孪生DaaS平台根据所述算法集合和所述功能参数，生成目标算法，并将所述目标算法发送到所述目标设备。
进一步地，所述接收模块还包括确定模块，所述确定模块用于：
获取所述算法生成指令中的用户产品需求，并根据所述用户产品需求，确定算法类型和功能参数；
将所述算法类型与算法仓库中的每个算法的类型标识进行对比，以根据所述类型标识与所述算法类型相同的算法确定算法集合。
进一步地，所述生成模块还用于：
通过数字孪生DaaS平台获取所述算法集合中的每个算法对应的功能参数标识，并计算每个算法对应的功能参数标识与所述功能参数的相似度；
将所述相似度与预设相似度阈值进行对比，得到对比结果；
若所述对比结果为所述算法集合中存在所述相似度大于预设相似度阈值的算法，则根据所述算法和所述功能参数生成目标算法；
若所述对比结果为所述算法集合中不存在所述相似度大于预设相似度阈值的算法，则通过人工智能搜索引擎根据所述功能参数在与所述数字孪生DaaS平台连接的算法知识库中进行搜索，得到功能参数标识与所述功能参数的相似度大于预设相似度阈值的算法，并根据所述算法和所述功能参数生成目标算法。
进一步地，所述生成模块还包括修改验证模块，所述修改验证模块用于：
在接收到所述目标设备发送的算法修改指令时，根据所述算法修改指令对所述目标算法进行修改；
对修改后的目标算法进行验证操作，并将通过验证的修改后的目标算法发送到所述目标设备。
进一步地，所述生成模块还包括管理模块，所述管理模块用于：
在预设时间段内未接收到目标设备发送的算法生成指令时，通过所述数字孪生DaaS平台对所述算法仓库中的算法进行优化操作，并通过所述数字孪生DaaS平台对所述算法仓库中的算法进行清理操作。
进一步地，所述管理模块还用于：
通过所述数字孪生DaaS平台对所述算法仓库中储存的算法进行优化，以得到优化后的算法存储在算法仓库中。；
对优化后的算法进行算法类型和功能参数的标识操作，并根据算法类型存储在所述算法仓库中。
进一步地，所述管理模块还用于：
通过所述数字孪生DaaS平台获取所述算法仓库中的每个算法的类型标识和功能参数标识，将每个算法的类型标识进行对比，并将每个算法的功能参数标识进行对比，以得到类型标识相同且功能参数标识也相同的算法集合；
将类型标识相同且功能参数标识也相同的算法集合中的每个算法进行模拟运行，确定每个算法对应的性能指标，并根据所述性能指标对算法进行清理操作。
此外，为实现上述目的，本发明还提供一种基于数字孪生DaaS平台的算法仓库管理***，所述基于数字孪生DaaS平台的算法仓库管理***包括：存储器、处理器及储存在所述存储器上并可在所述处理器上运行的算法仓库管理程序，所述算法仓库管理程序被所述处理器执行时实现如上所述的基于数字孪生DaaS平台的算法仓库管理方法的步骤。
此外，为实现上述目的，本发明还提供一种计算机可读存储介质，所述可读储存介质为计算机计算机可读存储介质，所述可读储存介质上储存有算法仓库管理程序，所述算法仓库管理程序被处理器执行时实现如上所述的基于数字孪生DaaS平台的算法仓库管理方法的步骤。
本发明提出的基于数字孪生DaaS平台的算法仓库管理方法，在接收目标设备发送的算法生成指令时，根据所述算法生成指令确定算法类型和功能参数，并根据所述算法类型在算法仓库中确定算法集合；通过数字孪生DaaS平台根据所述算法集合和所述功能参数，生成目标算法，并将所述目标算法发送到所述目标设备；本发明根据算法类型在算法仓库中选择对应的算法，并根据算法和功能参数，生成目标算法，无需人工参与算法的开发过程，进而提高了算法开发效率，降低成本。
图1是本发明实施例方案涉及的硬件运行环境的设备结构示意图；
图2为本发明基于数字孪生DaaS平台的算法仓库管理方法第一实施例的流程示意图。
本发明目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
应当理解，此处所描述的具体实施例仅仅用以解释本发明，并不用于限定本发明。
如图1所示，图1是本发明实施例方案涉及的硬件运行环境的设备结构示意图。
本发明实施例设备可以是PC机或服务器设备。
如图1所示，该设备可以包括：处理器1001，例如CPU，网络接口1004，用户接口1003，存储器1005，通信总线1002。其中，通信总线1002用于实现这些组件之间的连接通信。用户接口1003可以包括显示屏(Display)、输入单元比如键盘(Keyboard)，可选用户接口1003还可以包括标准的有线接口、无线接口。网络接口1004可选的可以包括标准的有线接口、无线接口(如WI-FI接口)。存储器1005可以是高速RAM存储器，也可以是稳定的存储器(non-volatile memory)，例如磁盘存储器。存储器1005可选的还可以是独立于前述处理器1001的储存装置。
本领域技术人员可以理解，图1中示出的设备结构并不构成对设备的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
如图1所示，作为一种计算机储存介质的存储器1005中可以包括操作***、网络通信模块、用户接口模块以及算法仓库管理程序。
其中，操作***是管理和控制便携储存设备与软件资源的程序，支持网络通信模块、用户接口模块、算法仓库管理程序以及其他程序或软件的运行；网络通信模块用于管理和控制网络接口1002；用户接口模块用于管理和控制用户接口1003。
在图1所示的储存设备中，所述储存设备通过处理器1001调用存储器1005中储存的算法仓库管理程序，并执行下述基于数字孪生DaaS平台的算法仓库管理方法各个实施例中的操作。
基于上述硬件结构，提出本发明基于数字孪生DaaS平台的算法仓库管理方法实施例。
参照图2，图2为本发明基于数字孪生DaaS平台的算法仓库管理方法第一实施例的流程示意图，所述方法包括：
步骤S10，在接收目标设备发送的算法生成指令时，根据所述算法生成指令确定算法类型和功能参数，并根据所述算法类型在算法仓库中确定算法集合；
步骤S20，通过数字孪生DaaS平台根据所述算法集合和所述功能参数，生成目标算法，并将所述目标算法发送到所述目标设备。
本实施例基于数字孪生DaaS平台的算法仓库管理方法运用于基于数字孪生DaaS平台的算法仓库管理***中，该基于数字孪生DaaS平台的算法仓库管理***与人工智能物联网平台和数字孪生DaaS平台进行通信连接；为了方便描述，以下将基于数字孪生DaaS平台的算法仓库管理***简称为管理***，并以管理***为例进行说明；管理***在接收到目标设备发送的算法生成指令时，获取算法生成指令中的用户产品需求，并根据用户产品需
求，确定算法类型和功能参数；管理***将算法类型与算法仓库中的每个算法的类型标识进行对比，以根据类型标识与算法类型相同的算法确定算法集合；管理***通过数字孪生DaaS平台获取算法集合中的每个算法对应的功能参数标识，并计算每个算法对应的功能参数标识与功能参数的相似度；将相似度与预设相似度阈值进行对比，得到对比结果；若对比结果为算法集合中存在相似度大于预设相似度阈值的算法，则根据算法和功能参数生成目标算法；若对比结果为算法集合中不存在相似度大于预设相似度阈值的算法，则通过人工智能搜索引擎根据所述功能参数在与数字孪生DaaS平台连接的算法知识库中进行搜索，得到功能参数标识与功能参数的相似度大于预设相似度阈值的算法，并根据算法和功能参数生成目标算法。需要说明的是，数字孪生DaaS平台为AIOTDaaS数字孪生平台，包括相互连接的业务中台和数据中台；数据中台用于对通过AIOTDaaS方式收集的数据进行采集、计算、存储、加工，形成的标准数据一方面进行储存一方面传送至业务中台；业务中台用于将基于数据中台传送的标准数据并结合行业应用，形成针对行业应用的模型及产品，以使用户能够基于业务中台快速封装出业务产品。人工智能物联网平台为AIOT PaaS物联网平台，其中包括AI边缘计算，数字芯片，边缘存储计算和模组技术。
本实施例中的基于数字孪生DaaS平台的算法仓库管理方法，在接收目标设备发送的算法生成指令时，根据算法生成指令确定算法类型和功能参数，并根据算法类型在算法仓库中确定算法集合；通过数字孪生DaaS平台根据算法集合和功能参数，生成目标算法，并将目标算法发送到目标设备；本发明根据算法类型在算法仓库中选择对应的算法，并根据算法和功能参数，生成目标算法，无需人工参与算法的开发过程，进而提高了算法开发效率，降低成本。
以下将对各个步骤进行详细说明：
步骤S10，在接收目标设备发送的算法生成指令时，根据所述算法生成指令确定算法类型和功能参数，并根据所述算法类型在算法仓库中确定算法集合；
在本实施例中，管理***与人工智能物联网平台进行通信连接，当相关开发人员需要对智能终端设备进行算法开发时，将该智能终端设备与人工智能物联网平台进行通信连接，以该智能终端设备作为目标设备，并通过目标设备向人工智能物联网平台发送算法生成指令，管理***接收人工智能物联网平台转发的算法生成指令，根据算法生成指令确定算法类型和功能参数，并根据算法类型在算法仓库中确定算法集合；需要说明的是，算法类型即算法的应用领域，例如人脸识别、工业制造、医疗诊断，自动驾驶，智能交通等；功能参数包括业务流程、工艺流程、应用场景等；算法仓库是提前建立在管理***中的，算法仓库中包含了不同
类型和不同功能的算法，并且算法仓库中的每个算法都进行了算法类型和功能参数的标识，便于对算法仓库的算法进行管理。
具体地，根据所述算法生成指令确定算法类型和功能参数，并根据所述算法类型在算法仓库中确定算法集合的步骤包括：
步骤a，获取所述算法生成指令中的用户产品需求，并根据所述用户产品需求，确定算法类型和功能参数；
在该步骤中，相关开发人员在通过目标设备发送算法生成指令时，根据自身需求，确定目标设备需要的算法类型和功能参数，根据算法类型和功能参数生成对应的用户产品需求，并生成算法生成指令，通过目标设备将算法生成指令发送到与管理***连接的人工智能物联网平台中，管理***获取人工智能物联网平台中的算法生成指令，并根据算法生成指令中的用户产品需求，确定相关开发人员需要的算法类型和功能参数。
步骤b，将所述算法类型与算法仓库中的每个算法的类型标识进行对比，以根据所述类型标识与所述算法类型相同的算法确定算法集合。
在该步骤中，管理***在确定算法类型后，将算法类型依次与算法仓库中的每个算法的类型标识进行对比，得到对比结果，获取对比结果为类型标识与算法类型相同的算法确定算法集合；可以理解的是，算法仓库中储存有大量的不同类型的算法，管理***先从所有类型的算法中选择类型标识与相关开发人员要求的算法类型相同的算法，能够提高选择效率，进而有助于提高算法开发效率。
步骤S20，通过数字孪生DaaS平台根据所述算法集合和所述功能参数，生成目标算法，并将所述目标算法发送到所述目标设备。
在本实施例中，管理***在确定算法集合后，通过数字孪生DaaS平台获取算法集合中的每个算法对应的功能参数标识，并将每个算法对应的功能标识与功能参数进行对比，得到对比结果，并结合对比结果和功能参数，生成目标算法，将目标算法发送到目标设备中。
具体地，步骤S20包括：
步骤c，通过数字孪生DaaS平台获取所述算法集合中的每个算法对应的功能参数标识，并计算每个算法对应的功能参数标识与所述功能参数的相似度；
在该步骤中，管理***通过数字孪生DaaS平台获取算法集合中的每个算法对应的功能参数标识，并计算每个算法对应的功能参数标识与相关开发人员要求的功能参数的相似度。
进一步地，在一种可能的实施例中，管理***得到相似度为100％，即算法仓库中存在与功能参数标识与相关开发人员要求的功能参数完全相同，此时，管理***可直接将该算
法作为目标算法，并直接发送到目标设备中。
步骤d，将所述相似度与预设相似度阈值进行对比，得到对比结果；
在该步骤中，管理***在确定每个算法对应的功能参数标识与功能参数的相似度后，将每个算法对应的功能参数标识与功能参数的相似度分别与预设相似度阈值进行对比，得到对比结果。
步骤e，若所述对比结果为所述算法集合中存在所述相似度大于预设相似度阈值的算法，则根据所述算法和所述功能参数生成目标算法；
在该步骤中，管理***若得到对比结果为算法集合中存在相似度大于预设相似度阈值的算法，则根据算法和功能参数生成目标算法；可以理解的是，相似度大于预设相似度阈值的算法，可能存在一个或多个，管理***将所有相似度大于预设相似度阈值的算法结合功能参数，生成目标算法。
步骤f，若所述对比结果为所述算法集合中不存在所述相似度大于预设相似度阈值的算法，则通过人工智能搜索引擎根据所述功能参数在与所述数字孪生DaaS平台连接的算法知识库中进行搜索，得到功能参数标识与所述功能参数的相似度大于预设相似度阈值的算法，并根据所述算法和所述功能参数生成目标算法。
在该步骤中，管理***若得到对比结果为算法集合中不存在相似度大于预设相似度阈值的算法，则通过数字孪生DaaS平台调用其中的人工智能搜索引擎，根据功能参数在与数字孪生DaaS平台连接的算法知识库中进行搜索，得到功能参数标识与功能参数的相似度大于预设相似度阈值的算法，并根据算法和功能参数生成目标算法；需要说明的是，算法知识库是存在于互联网上的，管理***在自身的算法仓库中无法搜索到对应的算法时，通过人工智能搜索引擎在互联网上的算法知识库中进行搜索，得到功能参数标识与功能参数的相似度大于预设相似度阈值的算法，并根据算法和功能参数生成目标算法；可以理解的是，功能参数中包括业务流程、工艺流程、应用场景等，管理***确定算法的功能参数标识是与相关开发人员要求的功能参数相似度极高的，因此，管理***根据相关开发人员要求的功能参数，对算法的功能参数进行修改调整，便能快速地得到目标算法。
具体地，管理***可以应用在工业互联网，智慧医疗，智慧供应链，智慧金融，智慧农业，智慧社区，智慧园区，智慧交通等行业，例如应用在工业互联网时，工业互联网的相关开发人员需要对工业互联网中的某个生产设备进行算法的开发时，相关开发人员将该生产设备连接到管理***中的人工智能物联网平台中，并通过该生产设备发送算法生成指令到人工智能物联网平台，管理***从算法生成指令中获取该生产设备对应的工艺流程，根据该
工艺流程，通过数字孪生DaaS平台在算法仓库中搜索对应的算法，并计算算法对应的工艺流程与该生产设备的工艺流程的相似度，当算法对应的工艺流程与该生产设备的工艺流程的相似度为100％时，直接使用该算法作为该生产设备的算法，通过数字孪生DaaS平台将该算法发送到人工智能物联网平台，再由人工智能物联网平台发送到该生产设备中，该生产设备根据该算法运行，执行相应的工艺流程生产对应的产品；当算法仓库中不存在算法对应的工艺流程与该生产设备的工艺流程的相似度为100％时，则选择相似度最高的算法，并通过数字孪生DaaS平台的人工智能搜索引擎在算法知识库中进行搜索其他相似的算法，将算法集合进行融合派生，得到符合该生产设备的工艺流程的算法，并通过数字孪生DaaS平台将该算法发送到人工智能物联网平台，再由人工智能物联网平台发送到该生产设备中，该生产设备根据该算法运行，执行相应的工艺流程生产对应的产品。
又例如：应用在智慧交通时，相关开发人员需要对智慧交通中的某个自动驾驶汽车进行算法的开发时，同样通过上述管理***生成对应的自动驾驶算法，并发送到自动驾驶汽车上，使得自动驾驶汽车能够根据生成的自动驾驶算法进行自动驾驶。
本实施例的管理***在接收到目标设备发送的算法生成指令时，获取算法生成指令中的用户产品需求，并根据用户产品需求，确定算法类型和功能参数；管理***将算法类型与算法仓库中的每个算法的类型标识进行对比，以根据类型标识与算法类型相同的算法确定算法集合；管理***通过数字孪生DaaS平台获取算法集合中的每个算法对应的功能参数标识，并计算每个算法对应的功能参数标识与功能参数的相似度；将相似度与预设相似度阈值进行对比，得到对比结果；若对比结果为算法集合中存在相似度大于预设相似度阈值的算法，则根据算法和功能参数生成目标算法；若对比结果为算法集合中不存在相似度大于预设相似度阈值的算法，则通过人工智能搜索引擎根据所述功能参数在与数字孪生DaaS平台连接的算法知识库中进行搜索，得到功能参数标识与功能参数的相似度大于预设相似度阈值的算法，并根据算法和功能参数生成目标算法，无需人工参与算法的开发过程，进而提高了算法开发效率，降低成本。
进一步地，基于本发明基于数字孪生DaaS平台的算法仓库管理方法第一实施例，提出本发明基于数字孪生DaaS平台的算法仓库管理方法第二实施例。
基于数字孪生DaaS平台的算法仓库管理方法的第二实施例与基于数字孪生DaaS平台的算法仓库管理方法的第一实施例的区别在于，在步骤S20之后，基于数字孪生DaaS平台的算法仓库管理方法包括：
步骤g，在接收到所述目标设备发送的算法修改指令时，根据所述算法修改指令对所述目标
算法进行修改；
步骤h，对修改后的目标算法进行验证操作，并将通过验证的修改后的目标算法发送到所述目标设备。
在本实施例中，管理***将目标算法发送至目标设备后，目标设备会对目标算法进行模拟运行，以验证目标算法的功能和语法是否符合要求，一般情况下，目标算法都是能符合目标设备的要求的，在一些特殊情况下，目标设备确定目标算法不符合要求，则根据模拟运行结果，生成算法修改指令，并将算法修改指令发送给与管理***通信连接的人工智能物联网平台，算法修改指令中包含修改需求，修改需求中包括目标设备检测出目标算法中不符合要求的部分以及原因，管理***获取算法修改指令中的修改需求，并在算法仓库中选择满足修改需求的算法，将该算法与目标算法融合，以对目标设备对应的目标算法进行修改，并对修改后的目标算法进行验证操作，再将通过验证的修改后的目标算法发送到目标设备。可以理解的是，在满足修改需求的算法与目标算法融合的过程中，一般来说只需要将满足修改需求的算法中满足修改需求的部分替换到目标算法中不符合目标设备的要求的部分中便可。
本实施例中的管理***在接收到目标设备发送的算法修改指令时，根据算法修改指令对所述目标算法进行修改；对修改后的目标算法进行验证操作，并将通过验证的修改后的目标算法发送到目标设备。根据算法仓库和目标设备的算法修改指令对目标算法进行修改，使得算法修改过程也不需要人工干预，能够进一步提高算法开发的效率，降低成本。
进一步地，基于本发明基于数字孪生DaaS平台的算法仓库管理方法第一实施例和第二实施例，提出本发明基于数字孪生DaaS平台的算法仓库管理方法第三实施例。
基于数字孪生DaaS平台的算法仓库管理方法的第三实施例与基于数字孪生DaaS平台的算法仓库管理方法的第一实施例和第二实施例的区别在于，在步骤S20之后，基于数字孪生DaaS平台的算法仓库管理方法还包括：
步骤i，在预设时间段内未接收到目标设备发送的算法生成指令时，通过所述数字孪生DaaS平台对所述算法仓库中的算法进行优化操作，并通过所述数字孪生DaaS平台对所述算法仓库中的算法进行清理操作。
在本实施例中，管理***在预设时间段内未接收到目标设备发送的算法生成指令时，通过数字孪生DaaS平台对算法仓库中的算法进行优化操作，并通过数字孪生DaaS平台对算法仓库中的算法进行清理操作；可以理解的是，当相关开发人员需要对智能终端设备进行算法开发时，将该智能终端设备与人工智能物联网平台进行通信连接，以该智能终端设备作为目标设备，并通过目标设备向人工智能物联网平台发送算法生成指令，管理***在预
设时间段内未接收到目标设备发送的算法生成指令，即在预设时间段内都没有目标设备连接到人工智能物联网平台上。
具体地，通过数字孪生DaaS平台对所述算法仓库中的算法进行优化操作的步骤包括：
步骤i1，通过所述数字孪生DaaS平台对所述算法仓库中储存的算法进行优化，以得到优化后的算法；
在该步骤中，管理***在预设时间段内未接收到目标设备发送的算法生成指令时，通过数字孪生DaaS平台对算法仓库中储存的算法进行优化，以得到优化后的算法，可以理解的是，管理***通过数字孪生DaaS平台对算法仓库中储存的算法进行优化时，可将算法仓库中的某个算法结合某个应用场景，派生出新的算法，可将算法仓库中的某几个算法进行融合，派生出新的算法，还可通过人工智能搜索引擎在与数字孪生DaaS平台对接的不同算法库、第三方平台等进行搜索，搜索算法仓库中未存储的算法，还可将算法仓库中的某几个算法与人工智能搜索引擎搜索得到的算法进行融合，派生出新的算法。
步骤i2，对优化后的算法进行算法类型和功能参数的标识操作，并根据算法类型存储在所述算法仓库中。
在该步骤中，管理***对优化后的算法进行算法类型和功能参数的标识操作，并根据算法类型存储在算法仓库中，如：优化后的算法的算法类型标识为人脸识别的，将其归纳储存在人脸识别类别的算法集合中，对优化后的算法进行功能参数标识，可以使得算法仓库或数字孪生DaaS平台能够快速确定每个算法的功能参数是业务流程、工艺流程或应用场景等。
具体地，通过所述数字孪生DaaS平台对所述算法仓库中的算法进行清理操作的步骤包括：
步骤i3，通过所述数字孪生DaaS平台获取所述算法仓库中的每个算法的类型标识和功能参数标识，将每个算法的类型标识进行对比，并将每个算法的功能参数标识进行对比，以得到类型标识相同且功能参数标识也相同的算法集合；
步骤i4，将类型标识相同且功能参数标识也相同的算法集合中的每个算法进行模拟运行，确定每个算法对应的性能指标，并根据所述性能指标对算法进行清理操作。
在步骤i3至步骤i4中，管理***在预设时间段内未接收到目标设备发送的算法生成指令时，通过数字孪生DaaS平台获取算法仓库中的每个算法的类型标识和功能参数标识，将每个算法的类型标识进行对比，并将每个算法的功能参数标识进行对比，以得到类型标识
相同且功能参数标识也相同的算法集合；管理***将类型标识相同且功能参数标识也相同的算法集合中的每个算法进行模拟运行，确定每个算法对应的性能指标，并根据性能指标，将性能指标最优的算法保留，并将其他算法进行清理操作，以保证算法仓库简洁，使得管理***能快速根据算法仓库中的算法确定目标算法，提高目标算法的开发效率。
本实施例的管理***在预设时间段内未接收到目标设备发送的算法生成指令时，通过数字孪生DaaS平台对算法仓库中的算法进行优化操作，并通过数字孪生DaaS平台对算法仓库中的算法进行清理操作。减少算法仓库中不必要的存储，使得算法仓库简洁，进而使得管理***能快速根据算法仓库中的算法确定目标算法，有利于提高算法的开发效率。
本发明还提供一种基于数字孪生DaaS平台的算法仓库管理装置。本发明基于数字孪生DaaS平台的算法仓库管理装置包括：
接收模块，用于在接收目标设备发送的算法生成指令时，根据所述算法生成指令确定算法类型和功能参数，并根据所述算法类型在算法仓库中确定算法集合；
生成模块，用于通过数字孪生DaaS平台根据所述算法集合和所述功能参数，生成目标算法，并将所述目标算法发送到所述目标设备。
进一步地，所述接收模块还包括确定模块，所述确定模块用于：
获取所述算法生成指令中的用户产品需求，并根据所述用户产品需求，确定算法类型和功能参数；
将所述算法类型与算法仓库中的每个算法的类型标识进行对比，以根据所述类型标识与所述算法类型相同的算法确定算法集合。
进一步地，所述生成模块还用于：
通过数字孪生DaaS平台获取所述算法集合中的每个算法对应的功能参数标识，并计算每个算法对应的功能参数标识与所述功能参数的相似度；
将所述相似度与预设相似度阈值进行对比，得到对比结果；
若所述对比结果为所述算法集合中存在所述相似度大于预设相似度阈值的算法，则根据所述算法和所述功能参数生成目标算法；
若所述对比结果为所述算法集合中不存在所述相似度大于预设相似度阈值的算法，则通过人工智能搜索引擎根据所述功能参数在与所述数字孪生DaaS平台连接的算法知识库中进行搜索，得到功能参数标识与所述功能参数的相似度大于预设相似度阈值的算法，并根据所述算法和所述功能参数生成目标算法。
进一步地，所述生成模块还包括修改验证模块，所述修改验证模块用于：
在接收到所述目标设备发送的算法修改指令时，根据所述算法修改指令对所述目标算法进行修改；
对修改后的目标算法进行验证操作，并将通过验证的修改后的目标算法发送到所述目标设备。
进一步地，所述生成模块还包括管理模块，所述管理模块用于：
在预设时间段内未接收到目标设备发送的算法生成指令时，通过所述数字孪生DaaS平台对所述算法仓库中的算法进行优化操作，并通过所述数字孪生DaaS平台对所述算法仓库中的算法进行清理操作。
进一步地，所述管理模块还用于：
通过所述数字孪生DaaS平台对所述算法仓库中储存的算法进行优化，以得到优化后的算法；
对优化后的算法进行算法类型和功能参数的标识操作，并根据算法类型存储在所述算法仓库中。
进一步地，所述管理模块还用于：
通过所述数字孪生DaaS平台获取所述算法仓库中的每个算法的类型标识和功能参数标识，将每个算法的类型标识进行对比，并将每个算法的功能参数标识进行对比，以得到类型标识相同且功能参数标识也相同的算法集合；
将类型标识相同且功能参数标识也相同的算法集合中的每个算法进行模拟运行，确定每个算法对应的性能指标，并根据所述性能指标对算法进行清理操作。
本发明还提供一种基于数字孪生DaaS平台的算法仓库管理***。
基于数字孪生DaaS平台的算法仓库管理***包括：存储器、处理器及储存在所述存储器上并可在所述处理器上运行的算法仓库管理程序，所述算法仓库管理程序被所述处理器执行时实现如上所述的基于数字孪生DaaS平台的算法仓库管理方法的步骤。
其中，在所述处理器上运行的算法仓库管理程序被执行时所实现的方法可参照本发明基于数字孪生DaaS平台的算法仓库管理方法各个实施例，此处不再赘述。
本发明还提供一种计算机可读存储介质。
该计算机可读存储介质上储存有算法仓库管理程序，所述算法仓库管理程序被处理器执行时实现如上所述的基于数字孪生DaaS平台的算法仓库管理方法的步骤。
其中，在所述处理器上运行的算法仓库管理程序被执行时所实现的方法可参照本发明基于数字孪生DaaS平台的算法仓库管理方法各个实施例，此处不再赘述。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者***不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者***所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者***中还存在另外的相同要素。
上述本发明实施例序号仅仅为了描述，不代表实施例的优劣。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品储存在如上所述的一个储存介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，计算机，服务器，或者网络设备等)执行本发明各个实施例所述的方法。
以上仅为本发明的优选实施例，并非因此限制本发明的专利范围，凡是利用本发明说明书与附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本发明的专利保护范围内。
Claims (10)
- 一种基于数字孪生DaaS平台的算法仓库管理方法，其特征在于，所述基于数字孪生DaaS平台的算法仓库管理方法包括如下步骤：在接收目标设备发送的算法生成指令时，根据所述算法生成指令确定算法类型和功能参数，并根据所述算法类型在算法仓库中确定算法集合；通过数字孪生DaaS平台根据所述算法集合和所述功能参数，生成目标算法，并将所述目标算法发送到所述目标设备。
- 如权利要求1所述的基于数字孪生DaaS平台的算法仓库管理方法，其特征在于，所述根据所述算法生成指令确定算法类型和功能参数，并根据所述算法类型在算法仓库中确定算法集合的步骤包括：获取所述算法生成指令中的用户产品需求，并根据所述用户产品需求，确定算法类型和功能参数；将所述算法类型与算法仓库中的每个算法的类型标识进行对比，以根据所述类型标识与所述算法类型相同的算法确定算法集合。
- 如权利要求1所述的基于数字孪生DaaS平台的算法仓库管理方法，其特征在于，所述通过数字孪生DaaS平台根据所述算法集合和所述功能参数，生成目标算法的步骤包括：通过数字孪生DaaS平台获取所述算法集合中的每个算法对应的功能参数标识，并计算每个算法对应的功能参数标识与所述功能参数的相似度；将所述相似度与预设相似度阈值进行对比，得到对比结果；若所述对比结果为所述算法集合中存在所述相似度大于预设相似度阈值的算法，则根据所述算法和所述功能参数生成目标算法；若所述对比结果为所述算法集合中不存在所述相似度大于预设相似度阈值的算法，则通过人工智能搜索引擎根据所述功能参数在与所述数字孪生DaaS平台连接的算法知识库中进行搜索，得到功能参数标识与所述功能参数的相似度大于预设相似度阈值的算法，并根据所述算法和所述功能参数生成目标算法。
- 如权利要求1所述的基于数字孪生DaaS平台的算法仓库管理方法，其特征在于，所述将所述目标算法发送到所述目标设备的步骤之后，包括：在接收到所述目标设备发送的算法修改指令时，根据所述算法修改指令对所述目标算法进行修改；对修改后的目标算法进行验证操作，并将通过验证的修改后的目标算法发送到所述目标设备。
- 如权利要求1中所述的基于数字孪生DaaS平台的算法仓库管理方法，其特征在于，所述将所述目标算法发送到所述目标设备的步骤之后，还包括：在预设时间段内未接收到目标设备发送的算法生成指令时，通过所述数字孪生DaaS平台对所述算法仓库中的算法进行优化操作，并通过所述数字孪生DaaS平台对所述算法仓库中的算法进行清理操作。
- 如权利要求5所述的基于数字孪生DaaS平台的算法仓库管理方法，其特征在于，所述通过数字孪生DaaS平台对所述算法仓库中的算法进行优化操作的步骤包括：通过所述数字孪生DaaS平台对所述算法仓库中储存的算法进行优化，以得到优化后的算法；对优化后的算法进行算法类型和功能参数的标识操作，并根据算法类型存储在所述算法仓库中。
- 如权利要求5所述的基于数字孪生DaaS平台的算法仓库管理方法，其特征在于，所述通过所述数字孪生DaaS平台对所述算法仓库中的算法进行清理操作的步骤包括：通过所述数字孪生DaaS平台获取所述算法仓库中的每个算法的类型标识和功能参数标识，将每个算法的类型标识进行对比，并将每个算法的功能参数标识进行对比，以得到类型标识相同且功能参数标识也相同的算法集合；将类型标识相同且功能参数标识也相同的算法集合中的每个算法进行模拟运行，确定每个算法对应的性能指标，并根据所述性能指标对算法进行清理操作。
- 一种基于数字孪生DaaS平台的算法仓库管理装置，其特征在于，所述基于数字孪生DaaS平台的算法仓库管理装置包括：接收模块，用于在接收目标设备发送的算法生成指令时，根据所述算法生成指令确定算法类型和功能参数，并根据所述算法类型在算法仓库中确定算法集合；生成模块，用于通过数字孪生DaaS平台根据所述算法集合和所述功能参数，生成目标算法，并将所述目标算法发送到所述目标设备。
- 一种基于数字孪生DaaS平台的算法仓库管理***，其特征在于，所述基于数字孪生DaaS平台的算法仓库管理***包括：存储器、处理器及储存在所述存储器上并可在所述处理器上运行的算法仓库管理程序，所述算法仓库管理程序被所述处理器执行时实现如权利要求1至7中任一项所述的基于数字孪生DaaS平台的算法仓库管理方法的步骤。
- 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质上储存有算法仓库管理程序，所述算法仓库管理程序被处理器执行时实现如权利要求1至7中任一项所述的基于 数字孪生DaaS平台的算法仓库管理方法的步骤。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210596620.0 | 2022-05-30 | ||
| CN202210596620.0A CN114693220B (zh) | 2022-05-30 | 2022-05-30 | 基于数字孪生DaaS平台的算法仓库管理方法及*** |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023231526A1 true WO2023231526A1 (zh) | 2023-12-07 |
Family
ID=82144606
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2023/082910 WO2023231526A1 (zh) | 2022-05-30 | 2023-03-21 | 基于数字孪生DaaS平台的算法仓库管理方法及*** |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114693220B (zh) |
| WO (1) | WO2023231526A1 (zh) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114693220B (zh) * | 2022-05-30 | 2022-09-20 | 深圳市爱云信息科技有限公司 | 基于数字孪生DaaS平台的算法仓库管理方法及*** |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140101635A1 (en) * | 2012-10-09 | 2014-04-10 | Martin Hoffmann | Automated generation of two-tier mobile applications |
| CN112130849A (zh) * | 2020-09-27 | 2020-12-25 | 建信金融科技有限责任公司 | 代码自动生成方法及装置 |
| CN112416338A (zh) * | 2020-11-26 | 2021-02-26 | 上海睿成软件有限公司 | 一种基于标签的代码仓库*** |
| CN112883116A (zh) * | 2021-02-24 | 2021-06-01 | 深圳市爱云信息科技有限公司 | 基于区块链的供应链金融AI DaaS算法仓库平台 |
| CN113946366A (zh) * | 2020-07-17 | 2022-01-18 | 华为技术有限公司 | 代码分析的方法、***及计算设备 |
| CN114153482A (zh) * | 2022-02-09 | 2022-03-08 | 深圳市爱云信息科技有限公司 | 基于数字孪生DaaS平台的深度学习编程方法及*** |
| CN114693220A (zh) * | 2022-05-30 | 2022-07-01 | 深圳市爱云信息科技有限公司 | 基于数字孪生DaaS平台的算法仓库管理方法及*** |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020172865A1 (zh) * | 2019-02-28 | 2020-09-03 | 西门子（中国）有限公司 | 数字双胞胎建模仿真方法、装置和*** |
| CN110096263A (zh) * | 2019-04-28 | 2019-08-06 | 贝壳技术有限公司 | Web前端代码自动生成方法及装置 |
| CN112767058A (zh) * | 2021-02-05 | 2021-05-07 | 深圳市爱云信息科技有限公司 | AIOT DaaS数字孪生云平台 |
| CN114327607A (zh) * | 2021-04-28 | 2022-04-12 | 四川书写人生科技有限公司 | 一种bs代码自动生成方法 |
| CN113986245A (zh) * | 2021-10-28 | 2022-01-28 | 平安银行股份有限公司 | 基于halo平台的目标代码生成方法、装置、设备及介质 |
| CN114115895A (zh) * | 2021-11-23 | 2022-03-01 | 深圳Tcl新技术有限公司 | 一种代码查询方法、装置、电子设备和存储介质 |
-
2022
- 2022-05-30 CN CN202210596620.0A patent/CN114693220B/zh active Active
-
2023
- 2023-03-21 WO PCT/CN2023/082910 patent/WO2023231526A1/zh unknown
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140101635A1 (en) * | 2012-10-09 | 2014-04-10 | Martin Hoffmann | Automated generation of two-tier mobile applications |
| CN113946366A (zh) * | 2020-07-17 | 2022-01-18 | 华为技术有限公司 | 代码分析的方法、***及计算设备 |
| CN112130849A (zh) * | 2020-09-27 | 2020-12-25 | 建信金融科技有限责任公司 | 代码自动生成方法及装置 |
| CN112416338A (zh) * | 2020-11-26 | 2021-02-26 | 上海睿成软件有限公司 | 一种基于标签的代码仓库*** |
| CN112883116A (zh) * | 2021-02-24 | 2021-06-01 | 深圳市爱云信息科技有限公司 | 基于区块链的供应链金融AI DaaS算法仓库平台 |
| CN114153482A (zh) * | 2022-02-09 | 2022-03-08 | 深圳市爱云信息科技有限公司 | 基于数字孪生DaaS平台的深度学习编程方法及*** |
| CN114693220A (zh) * | 2022-05-30 | 2022-07-01 | 深圳市爱云信息科技有限公司 | 基于数字孪生DaaS平台的算法仓库管理方法及*** |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114693220A (zh) | 2022-07-01 |
| CN114693220B (zh) | 2022-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109086031B (zh) | 一种基于规则引擎的业务决策方法和装置 | |
| US10453165B1 (en) | Computer vision machine learning model execution service | |
| WO2021036447A1 (zh) | 基于Flink引擎的业务数据处理方法及装置 | |
| WO2019227573A1 (zh) | 协同办公数据流处理方法、装置、计算机设备及存储介质 | |
| TW201820165A (zh) | 用於雲端巨量資料運算架構之伺服器及其雲端運算資源最佳化方法 | |
| US11677861B2 (en) | System and method for interoperable communication between entities with different structures | |
| US20210312295A1 (en) | Information processing method, information processing device, and information processing program | |
| WO2023231526A1 (zh) | 基于数字孪生DaaS平台的算法仓库管理方法及*** | |
| CN111209077A (zh) | 深度学习框架设计方法 | |
| CN112965710B (zh) | 计算图的处理方法、装置和*** | |
| WO2021008576A1 (zh) | 上下文信息管理方法、装置、***及计算机可读存储介质 | |
| CN110737425B (zh) | 一种计费平台***的应用程序的建立方法及装置 | |
| WO2023093375A1 (zh) | 一种计算资源获取方法、装置、电子设备和存储介质 | |
| US20240095529A1 (en) | Neural Network Optimization Method and Apparatus | |
| CN113806429A (zh) | 基于大数据流处理框架的画布式日志分析方法 | |
| CN110442753A (zh) | 一种基于opc ua的图数据库自动建立方法及装置 | |
| CN116302448B (zh) | 任务调度方法和*** | |
| CN113495723B (zh) | 一种调用功能组件的方法、装置及存储介质 | |
| WO2021227293A1 (zh) | 一种通用的人工智能模型训练方法及*** | |
| CN114253550A (zh) | 优化策略生成方法和算子构建方法 | |
| CN114170004A (zh) | 基于多事件的评分决策方法、装置、设备及存储介质 | |
| CN112367205A (zh) | 一种对http调度请求的处理方法及调度*** | |
| CN115437757A (zh) | 调度方法、***、服务器和计算机可读存储介质 | |
| CN112632113A (zh) | 一种算子管理方法和算子管理*** | |
| CN111159203A (zh) | 一种数据关联分析的方法、平台、电子设备及存储介质 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23814723 Country of ref document: EP Kind code of ref document: A1 |