A method, device and storage medium for arranging a business plan

Through the target business flow chart, the business plan is arranged, the files of functional nodes are obtained and packaged, and the performance of the hardware configuration plan is estimated, which solves the problem that existing AI service platforms are difficult to deal with complex business scenarios and cannot guarantee optimal resource performance, and realizes the optimization and difficulty of business plan.

CN115421706BActive Publication Date: 2025-05-23HANGZHOU HIKVISION DIGITAL TECHNOLOGY CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202211056100.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-08-31
Publication Date
2025-05-23
Estimated Expiration
2042-08-31

AI Technical Summary

Technical Problem

Existing AI service platforms are difficult to deal with complex business scenarios and cannot guarantee that users choose business solutions with the best resource performance.

Method used

The business plan is orchestrated through the target business flow chart, the executable files and configuration files of each functional node are obtained, packaged into engine packages, and the algorithm performance and resource consumption of each hardware configuration plan are estimated, and the user is prompted to choose the optimal solution.

Benefits of technology

It reduces the difficulty of program development, optimizes the performance of business solutions, and helps users choose business solutions with the best resource performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115421706B_ABST
    Figure CN115421706B_ABST
Patent Text Reader

Abstract

The present application discloses a method, device and storage medium for arranging a business plan, which relates to the field of artificial intelligence technology and can be used to reduce the difficulty of program development and optimize the performance of business plans. The method includes: obtaining a target business process diagram, the target business process diagram includes multiple functional nodes and the connection relationship between the multiple functional nodes; obtaining executable files and configuration files corresponding to each functional node in the target business process diagram, and packaging the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package; estimating the algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme in at least one hardware configuration scheme; prompting the algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method, device and storage medium for arranging a business plan. Background Art

[0002] With the continuous popularization of AI (artificial intelligence) technology and its application in various industries, the application scenarios of AI technology are becoming more and more abundant, and the complexity of the demand for AI technology in various industries is also increasing, which also makes the development cycle of AI project development longer and longer.

[0003] In order to facilitate the rapid implementation of AI-related projects by enterprises, the AI ​​service platform system in related technologies mainly provides the function of training the corresponding AI algorithm model by uploading data. However, this type of platform service can only handle business scenarios with single functions and cannot guarantee that users can determine the business plan with the best resource performance. Summary of the invention

[0004] The embodiments of the present application provide a method, device and storage medium for arranging a business solution, which can be used to reduce the difficulty of program development and optimize the performance of the business solution.

[0005] In a first aspect, an embodiment of the present application provides a method for orchestrating a business solution, the method comprising: obtaining a target business process diagram, the target business process diagram comprising multiple functional nodes and connection relationships between the multiple functional nodes, the target business process diagram being used to describe a process for implementing a target business; obtaining executable files and configuration files corresponding to each functional node in the target business process diagram, and packaging the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package; estimating the algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme in at least one hardware configuration scheme; and prompting the algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme.

[0006] The technical solution provided by the embodiment of the present application brings at least the following beneficial effects: by arranging the business plan through the target business process diagram, the dependence on program development experience can be reduced, thereby reducing the difficulty of program development; at the same time, the algorithm performance and consumed resources of the engine package corresponding to the business plan in each hardware configuration plan are evaluated, and the estimated results are prompted to the user so that the user can determine the business plan with the best resource performance, thereby optimizing the performance of the business plan.

[0007] In some embodiments, the above-mentioned acquisition of the target business process diagram includes: acquiring business keywords input by the user, and displaying multiple business process diagrams related to the business keywords based on the business keywords; and acquiring the target business process diagram selected by the user from the multiple business process diagrams.

[0008] Based on this, relevant business process diagrams can be displayed according to the user's business keywords to provide users with reference, so that users can directly select existing flowchart templates, eliminating the process of building business process diagrams, saving time, and improving the efficiency of business plan arrangement.

[0009] In some embodiments, the above-mentioned acquisition of the target business process diagram includes: obtaining business keywords input by the user, and displaying multiple business process diagrams related to the business keywords based on the business keywords; obtaining a reference business process diagram selected by the user from multiple business process diagrams; in response to the user's modification instruction to the reference business process diagram, modifying the reference business process diagram, and using the modified reference business process diagram as the target business process diagram.

[0010] Based on this, relevant business process diagrams can be displayed according to the user's business keywords, allowing users to modify existing business process diagrams instead of building business process diagrams from scratch, thereby improving the efficiency of business plan orchestration.

[0011] In some embodiments, the above-mentioned acquisition of the target business process diagram includes: displaying a graphical draggable component pool, the component pool including multiple functional nodes; in response to a user's instruction to drag a functional node in the component pool to an orchestration area and connect the functional nodes in the orchestration area, the business process diagram generated after connecting the functional nodes in the orchestration area is used as the target process diagram.

[0012] Based on this, users can freely arrange the flowchart corresponding to the business plan according to their needs, which improves the flexibility of business plan arrangement. At the same time, the arrangement process does not rely on program development experience, which reduces the difficulty of program development.

[0013] In some embodiments, the above-mentioned acquisition of executable files and configuration files corresponding to each functional node of the target business process diagram includes: acquiring configuration parameters of each functional node in the target business process diagram; determining whether the connection relationship between each functional node of the target business process diagram and the configuration parameters of each functional node of the target business process diagram meet preset rules; if the preset rules are met, selecting executable files and configuration files corresponding to each functional node of the target business process diagram from the database.

[0014] Based on this, the target business process diagram can be verified, so that only when the connection relationship between the various functional nodes of the target business process diagram and the configuration parameters of each functional node meet the preset rules, the files required to implement the target business process diagram (that is, the executable files and configuration files corresponding to the various functional nodes of the target business process diagram) are further selected from the database, thereby saving computing power.

[0015] In some embodiments, the configuration parameters of the above-mentioned functional node include a model file corresponding to the functional node, and the model file corresponding to the functional node is determined according to the following method: receiving training data corresponding to the functional node uploaded by a user, and obtaining the model file corresponding to the functional node by training according to the training data; or, receiving the model file uploaded by the user for the functional node, and using the model file uploaded by the user for the functional node as the model file corresponding to the functional node.

[0016] It should be understood that the model file of the functional node trained according to the training data uploaded by the user will be more in line with the user's actual business needs. In addition, the model file uploaded by the user for the functional node can be a model file set by the user himself that is more in line with the expectations. Based on this, the function customization of the functional node can be achieved, thereby improving the adaptability of the functional node to the business.

[0017] In some embodiments, the above method also includes: receiving a test data set uploaded by a user; the above packaging of the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package includes: performing simulation testing on the test data set according to the executable files and configuration files corresponding to each functional node in the target business process diagram; if the results after the simulation test meet the user requirements, packaging the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package.

[0018] Based on this, before packaging the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package, it is possible to first determine whether the simulation test results of the executable files and configuration files corresponding to the target process diagram meet the user requirements, so that the executable files and configuration files corresponding to each functional node are packaged only when they meet the requirements, thereby saving the amount of calculation. In addition, the feasibility and reliability of the business plan are guaranteed.

[0019] In some embodiments, the above method also includes: in response to a user's selection instruction for a hardware configuration scheme, determining a target hardware configuration scheme from at least one hardware configuration scheme; sending the engine package to the target hardware platform according to the target hardware configuration scheme; receiving the data to be processed uploaded by the user, and sending the data to be processed to the target hardware platform, so that the target hardware platform runs the engine package according to the target hardware configuration scheme to perform business processing on the data to be processed, and obtains a business processing result corresponding to the target business.

[0020] Based on this, users can select the target hardware configuration plan that best meets their needs based on the estimated results of the algorithm performance and resource consumption of the engine package corresponding to the business plan in each hardware configuration plan, and further run the engine package corresponding to the target business process diagram to process business data according to the plan.

[0021] In the second aspect, an embodiment of the present application provides a business plan orchestration device, including: a process acquisition module, used to obtain a target business process diagram, the target business process diagram includes multiple functional nodes and connection relationships between the multiple functional nodes, and the target business process diagram is used to describe the process of realizing the target business; a file configuration module, used to obtain executable files and configuration files corresponding to each functional node in the target business process diagram, and package the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package; a resource estimation module, used to estimate the algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme in at least one hardware configuration scheme; and prompt the algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme.

[0022] In some embodiments, the process acquisition module is specifically used to acquire business keywords input by the user, and display multiple business process diagrams related to the business keywords according to the business keywords; and acquire a target business process diagram selected by the user from the multiple business process diagrams.

[0023] In some embodiments, the above-mentioned process acquisition module is specifically used to obtain business keywords input by the user, and display multiple business process diagrams related to the business keywords based on the business keywords; obtain a reference business process diagram selected by the user from multiple business process diagrams; in response to the user's modification instruction to the reference business process diagram, modify the reference business process diagram, and use the modified reference business process diagram as the target business process diagram.

[0024] In some embodiments, the above-mentioned process acquisition module is specifically used to display a graphical draggable component pool, which includes multiple functional nodes; in response to the user's instruction to drag the functional node in the component pool to the orchestration area and connect the various functional nodes in the orchestration area, the business process diagram generated after connecting the various functional nodes in the orchestration area is used as the target business process diagram.

[0025] In some embodiments, the above-mentioned file configuration module is specifically used to obtain the configuration parameters of each functional node in the target business process diagram; determine whether the connection relationship between the various functional nodes of the target business process diagram and the configuration parameters of each functional node of the target business process diagram meet the preset rules; if the preset rules are met, select the executable files and configuration files corresponding to the various functional nodes of the target business process diagram from the database; the configuration parameters of the above-mentioned functional nodes include the model files corresponding to the functional nodes, and the model files corresponding to the functional nodes are determined according to the following methods: receiving the training data corresponding to the functional nodes uploaded by the user, and obtaining the model files corresponding to the functional nodes according to the training data; or, receiving the model files uploaded by the user for the functional nodes, and using the model files uploaded by the user for the functional nodes as the model files corresponding to the functional nodes.

[0026] In some embodiments, the above-mentioned file configuration module is also used to receive a test data set uploaded by a user; the above-mentioned file configuration module is specifically used to perform simulation testing on the test data set according to the executable files and configuration files corresponding to each functional node in the target business process diagram; if the result after the simulation test meets the user requirements, the executable files and configuration files corresponding to each functional node in the target business process diagram are packaged into an engine package.

[0027] In some embodiments, the above-mentioned device also includes a business processing module, which is used to respond to the user's selection instruction for the hardware configuration plan, determine the target hardware configuration plan from at least one hardware configuration plan; send the engine package to the target hardware platform according to the target hardware configuration plan; receive the data to be processed uploaded by the user, and send the data to be processed to the target hardware platform, so that the target hardware platform runs the engine package according to the target hardware configuration plan to perform business processing on the data to be processed, and obtains the business processing result corresponding to the target business.

[0028] In a third aspect, an embodiment of the present application provides a business solution orchestration device, comprising: a memory and a processor; the memory and the processor are coupled; the memory is used to store computer program code, and the computer program code includes computer instructions; wherein, when the processor executes the computer instructions, the business solution orchestration device executes a business solution orchestration method such as the first aspect and any possible design method thereof.

[0029] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium includes computer instructions. When the computer instructions are executed on a computer, the computer executes the method provided in the first aspect and possible implementations.

[0030] In a fifth aspect, an embodiment of the present application provides a computer program product comprising computer instructions, which, when executed on a computer, enables the computer to execute the method provided in the first aspect and possible implementations described above.

[0031] It should be noted that the above computer instructions may be stored in whole or in part on a computer-readable storage medium, wherein the computer-readable storage medium may be packaged together with the processor of the controller, or may be packaged separately from the processor of the controller, which is not limited in this application.

[0032] The specific description of the second to fifth aspects and their various implementations in this application can refer to the detailed description of the first aspect and its various implementations. The beneficial effects of the second to fifth aspects and their various implementations can refer to the beneficial effect analysis of the first aspect and its various implementations, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A schematic diagram of the structure of an orchestration system for a service solution according to some embodiments Figure 1 ;

[0034] Figure 2 A schematic diagram of the structure of an orchestration system for a service solution according to some embodiments Figure 2 ;

[0035] Figure 3 A flowchart of a method for arranging a service solution according to some embodiments;

[0036] Figure 4 is a flow chart of a target business according to some embodiments;

[0037] Figure 5 A schematic diagram of a structure of an arrangement device for a service solution according to some embodiments Figure 1 ;

[0038] Figure 6 A schematic diagram of a structure of an arrangement device for a service solution according to some embodiments Figure 2 . DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in this application to clearly and completely describe the technical solutions in this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0040] It should be noted that, in the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way. The terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, "multiple" means two or more.

[0041] As described in the background technology, the application scenarios of AI technology are becoming more and more abundant, and the complexity of the demand for AI technology in various industries is also increasing, which also makes the development cycle of AI project development longer and longer. In order to facilitate enterprises to promote the rapid implementation of AI-related projects, the AI ​​service platform system in the related technology mainly provides the function of training the corresponding AI algorithm model by uploading data. However, this type of platform service can only handle business scenarios with a single function, and cannot guarantee that users can determine the business plan with the best resource performance.

[0042] In this regard, an embodiment of the present application provides a method for orchestrating a business solution. The core idea of ​​the method is that the business solution is orchestrated through a target business process diagram that includes multiple functional nodes and connection relationships between the multiple functional nodes, so as to cope with more complex business scenarios. In addition, the process of AI productization is simplified, and the degree of dependence on program development experience and the difficulty of program development are reduced. At the same time, the algorithm performance and consumed resources of the engine package corresponding to the business solution in each hardware configuration solution are evaluated, and the estimated results are prompted to the user so that the user can determine the business solution with the best resource performance, thereby optimizing the performance of the business solution.

[0043] To facilitate the subsequent description, the following first introduces an overall arrangement system of a business solution involved in this application. Figure 1 As shown, Figure 1 It is a structural diagram of an orchestration system for a business solution provided by this application.

[0044] Reference Figure 1 , the orchestration system 100 of the business solution includes an application layer 101 , a platform layer 102 , an engine layer 103 , and an algorithm layer 104 .

[0045] Among them, the application layer 101 is used to implement the application interaction service of the user end. Optionally, the application layer 101 can implement the visual arrangement function; exemplarily, the application layer can display a graphical draggable component pool, which includes multiple functional nodes, so that the user can generate a business process diagram by dragging and connecting multiple functional nodes in the component pool. Optionally, the application layer 101 can implement the template suite selection function; exemplarily, the application layer can display a template suite of business process diagrams of various common requirements, so that the user can directly import and use them or import and use them after adjustment. For example, the application layer can display a suite template of multiple business process diagrams related to the business keywords entered by the user for the user to use. Optionally, the application layer 101 can implement the account management function; exemplarily, the application layer 101 can display the user's personal account, so that the user can set up a personal account on the orchestration system 100 of the business solution, and further, the user can save commonly used business solutions on the personal account. Optionally, the application layer 101 can also display the cloud verification results corresponding to the business process diagram.

[0046] The platform layer 102 connects the application layer 101 and the engine layer 103, can set the model file of each functional node, and verify the business process diagram. Optionally, the platform layer 102 can be used to implement the training platform service. In some examples, the platform layer 102 can implement the customized model production function; exemplarily, a customized model can be trained based on the data uploaded by the user and the model required by the business. Optionally, the platform layer 102 can be used to implement the verification platform service function. In some examples, the verification platform service function includes the orchestration diagram verification function; exemplarily, the platform layer 102 can determine whether the connection relationship between each functional node in the business process diagram and the configuration parameters of each functional node in the business process diagram meet the preset rules. In some examples, the verification platform service includes a simulation data function; exemplarily, the platform layer 102 can perform simulation tests on the test data set uploaded by the user to determine whether the results of the simulation test meet the requirements. In some examples, the verification platform service includes a node capability set, in which the rule constraints of algorithm orchestration are preset. The platform layer 102 can determine whether the connection relationship between each functional node in the business process diagram meets the rule constraints of algorithm orchestration based on the node capability set; wherein the rule constraints of algorithm orchestration include but are not limited to: the attribute of the input data of the functional node should be consistent with the attribute of the output data of the functional node connected to the input end of the functional node, for example, if the attribute of the input data of the functional node is image data, then the attribute of the output data of the functional node connected to the input end of the functional node should also be image data; the connection of each functional node does not have logical errors, for example, it should be ensured that when there is a circular connection relationship between each functional node, there is a loop termination condition, etc.

[0047] The engine layer 103 connects the platform layer 102 and the algorithm layer 104. Optionally, the engine layer 103 includes a node parallel scheduling function. Exemplarily, for different business processes, the engine layer 103 can perform parallel scheduling processing on functional nodes that can be parallel, and use a multi-threaded approach to schedule the solution corresponding to the overall business process diagram. Optionally, the engine layer 103 includes a hardware platform adaptation function and a suite processing flow. Exemplarily, for different hardware platforms, the engine layer 103 can select the underlying dynamic library for adaptability as needed to improve the efficiency and performance of the algorithm operation. Optionally, the engine layer 103 includes a component packaging function. Exemplarily, after completing the node connection at the engine level, the engine layer 103 can package the engine components and send them to the remote device platform to deploy the application.

[0048] The algorithm layer 104 is connected to the engine layer 103. Optionally, the algorithm layer 104 includes an AI reasoning operator function to support the function of reasoning on a model trained using a deep learning framework. Exemplarily, the AI ​​reasoning operator function can perform algorithm processing such as preprocessing, feature extraction, forward reasoning, and post-processing on the input information through a specific deep learning network. Optionally, the algorithm layer 104 includes an algorithm node template function. Exemplarily, the algorithm layer 104 can template the public algorithm nodes provided to users for free arrangement, and provide different node configuration templates; wherein the above algorithm nodes include one or more of the following: detection, classification, instance segmentation, semantic segmentation, text recognition, graphic comparison, background modeling, face recognition, target association, target screening, branch selection, and 3D size measurement algorithm nodes; it should be understood that the content of the above algorithm nodes is only an example, and other algorithm content can be added according to actual business needs, and this application does not limit this. Optionally, the algorithm layer 104 includes a custom development framework function. For example, the algorithm layer 104 can provide a customized algorithm development framework to support the expansion of third-party algorithm capabilities for different business scenarios. In addition, the algorithm layer 104 can also provide corresponding underlying dynamic library files on different platforms and different data sources.

[0049] In some embodiments, reference Figure 2 , the business solution orchestration system 100 can be logically divided into Figure 2 The solution recommendation module 105, the custom node module 106, and the resource evaluation module 107.

[0050] The function of the solution recommendation module 105 is implemented by the above application layer and the above platform layer, and the solution recommendation module 105 can recommend an available orchestration algorithm template to the user. Optionally, the orchestration algorithm template can be a business process diagram template; the orchestration algorithm template can include several sets of commonly used orchestration algorithm templates, or it can be an orchestration algorithm template provided according to the business keywords input by the user.

[0051] The functions of the custom node module 106 can be implemented by the platform layer and the engine layer, and the custom node module 106 can be used to meet the user's needs for customizing the orchestration algorithm. Optionally, the custom node module 106 can receive the self-developed function node uploaded by the user, and the function node is the smallest logical unit when executing the orchestration method of the business plan; optionally, the custom node module 106 can receive the model file uploaded by the user, and the model file is configured in the function node to implement the algorithm function of the function node.

[0052] The functions of the resource evaluation module 107 can be implemented by the above-mentioned engine layer and the above-mentioned algorithm layer. The resource evaluation module 107 can make corresponding evaluation references based on the performance range of the underlying algorithm library and the resources required by the upper-layer engine, so that users can choose an evaluation plan that meets their needs, and then the engine scheduling layer selects the corresponding algorithm library and configuration file according to the actual hardware platform environment and packages and deploys it to the device platform to implement the application.

[0053] It should be understood that Figure 2 The division of the orchestration system 100 of the above service solution is exemplary, and is only a division of a logical function. In actual implementation, there may be other division methods. The present application embodiment does not impose any specific limitation on this.

[0054] It should be noted that the embodiments of the present application do not limit the specific application scenarios of the method for arranging the business solutions provided by the present application, and can be applied to any business processing scenario. In addition, the method for arranging the business solutions provided by the embodiments of the present application can be executed by an arrangement device for the business solutions. For example, the arrangement device for the business solutions can be a computer device. For another example, the arrangement device for the business solutions can be the above-mentioned Figure 1 or Figure 2The orchestration system of the business solution shown. For another example, the orchestration device of the business solution can be a server. For another example, the orchestration device of the business solution can be any electronic chip with processing function, such as a central processing unit (CPU), a graphics processing unit (GPU), a general processor, a network processor (NP), a neural network processor (NPU), a digital signal processor (DSP), a microprocessor, a microcontroller, a programmable logic device (PLD) or any combination thereof; for another example, the orchestration device of the business solution can be a mobile phone, a tablet computer, a desktop, a laptop, a handheld computer, a notebook computer, an ultra-mobile personal computer (UMPC), a netbook, and a cellular phone, a personal digital assistant (PDA), an augmented reality (AR)\virtual reality (VR) device, etc. This application does not impose any special restrictions on the specific form of the orchestration device of the business solution.

[0055] For ease of understanding, the following is a detailed introduction to the arrangement method of the business solution provided by this application in conjunction with the accompanying drawings.

[0056] Figure 3 A method for arranging a business plan proposed in the embodiment of the present application is applied to the field of artificial intelligence technology. The following takes the execution subject as a computer device as an example. Figure 3 The method shown is described in detail. Figure 3 As shown, the service solution arrangement method includes the following steps S101 to S104:

[0057] S101. Obtain a target business process diagram.

[0058] The target business process diagram is used to describe the process of achieving the target business. In addition, the target business process diagram includes multiple function nodes and the connection relationship between the multiple function nodes. The function node is the smallest logical unit when executing the orchestration method of the business plan.

[0059] In some embodiments, the function node can implement various algorithm functions. Exemplarily, the functions that the function node can implement include one or more of the following: detection, classification, instance segmentation, semantic segmentation, text recognition, graphic comparison, background modeling, face recognition, target association, target screening, branch selection, and 3D size measurement. It should be understood that the functions of the above-mentioned function nodes are only examples, and this application does not impose specific restrictions on this.

[0060] In some embodiments, in order to standardize the use of functional nodes and coordinate the connection relationship between various functional nodes, the configuration parameters of the functional nodes may be further set. For ease of understanding, some configuration parameter setting methods are exemplarily given below.

[0061] Optionally, the configuration parameters of a function node may include attributes of the input data and output data of the function node. Exemplarily, if the function implemented by a function node is face recognition, the attributes of the input data of the function can be set to image data. Furthermore, if the attributes of the output data of the function node are set to image type data, the function node can output specific face image data after face recognition; if the attributes of the output data of the function node are set to logic type data, the function node can output discrimination information after face recognition, for example, it can output logic data used to characterize whether there is a face in the image data input by the function node. It should be understood that the above-mentioned image type data and logic type data are only examples, and the attributes of the input data and the attributes of the output data can also be other types of data, and the present application does not impose specific restrictions on this.

[0062] Optionally, the configuration parameters of the function node may include a model file corresponding to the function node, so that the data input to the function node is processed accordingly through the model file corresponding to the function node. In some examples, the model file corresponding to the function node may be a general model file by default to save the user's model training time. In other examples, the model file corresponding to the function node can be determined in the following manner: receiving the training data corresponding to the function node uploaded by the user, and training the model file corresponding to the function node according to the training data; based on this, the effect of the model file of the function node can be made more in line with the actual business needs of the user. In some other examples, the model file corresponding to the function node can be determined in the following manner: receiving the model file uploaded by the user for the function node, and using the model file uploaded by the user for the function node as the model file corresponding to the function node; based on this, the user can design a model file that meets the expectations, and use the model file as the model file corresponding to the function node to improve the adaptability of the function node to the business.

[0063] It should be noted that the above configuration parameters (such as the attributes of input data) are only examples, and are only for the convenience of relevant technicians in the field to understand the role of configuration parameters. In actual application, the configuration parameters of the function node may also include other parameters, and may be added or deleted according to user needs, and this application does not impose specific restrictions on this.

[0064] In some embodiments, the connection relationship between multiple functional nodes includes logical connection relationships such as input-output relationship (i.e., sequential relationship), conditional selection relationship, and loop relationship between multiple functional nodes. Among them, the connection relationship between multiple functional nodes can be configured in the configuration parameters of multiple functional nodes, or can be presented in the business processing flow chart in the form of a connection, or can be presented in the business processing flow chart in the form of a logical node, wherein the logical node is used to assist multiple functional nodes in realizing the above-mentioned logical connection relationship, for example, the logical node can record the judgment condition in the conditional selection relationship, or record the loop termination condition in the loop relationship. It should be understood that the present application does not specifically limit the connection method between multiple functional nodes.

[0065] In some embodiments, the above-mentioned acquisition of the target business process diagram is specifically implemented as the following steps: acquiring the business keyword input by the user, and displaying multiple business process diagrams related to the business keyword according to the business keyword; acquiring the target business process diagram selected by the user from the multiple business process diagrams. Based on this, the relevant business process diagrams can be displayed according to the user's business keyword to provide a reference for the user, so that the user can directly select an existing flowchart template, eliminating the process of building a business process diagram, saving time, and improving the efficiency of arranging business plans.

[0066] In other embodiments, the above-mentioned acquisition of the target business process diagram is specifically implemented as the following steps: acquiring the business keyword input by the user, and displaying multiple business process diagrams related to the business keyword according to the business keyword; acquiring the reference business process diagram selected by the user from multiple business process diagrams; in response to the user's modification instruction for the reference business process diagram, modifying the reference business process diagram, and using the modified reference business process diagram as the target business process diagram. Based on this, the relevant business process diagram can be displayed according to the user's business keyword, so that the user can modify the existing business process diagram instead of building a business process diagram from scratch, thereby improving the efficiency of business plan arrangement.

[0067] In some other embodiments, the above-mentioned acquisition of the target business process diagram is specifically implemented as the following steps: displaying a graphical draggable component pool, the component pool including multiple functional nodes; responding to the user's instruction to drag the functional node in the component pool to the arrangement area and connect the functional nodes in the arrangement area, the business process diagram generated after connecting the functional nodes in the arrangement area is used as the target process diagram. Based on this, the user can freely arrange the flow chart corresponding to the business plan according to the needs, improve the flexibility of the business plan arrangement, and at the same time, do not rely on the experience of program development in the arrangement process, reducing the difficulty of program development.

[0068] S102: Obtain executable files and configuration files corresponding to each functional node in the target business process diagram, and package the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package.

[0069] In some embodiments, the above-mentioned acquisition of executable files and configuration files corresponding to each functional node in the target business process diagram can be specifically implemented as follows: acquiring configuration parameters of each functional node in the target business process diagram; determining whether the connection relationship between each functional node in the target business process diagram and the configuration parameters of each functional node in the target business process diagram meet the preset rules; if the preset rules are met, then selecting executable files and configuration files corresponding to each functional node in the target business process diagram from the database. Furthermore, if the preset rules are not met, then issuing a first warning message to prompt the user to check the target business process diagram.

[0070] Based on this, the target business process diagram can be verified, so that only when the connection relationship between the various functional nodes of the target business process diagram and the configuration parameters of each functional node meet the preset rules, the files required to implement the target business process diagram are further selected from the database, thereby saving calculation amount.

[0071] For ease of understanding, the above preset rules are exemplarily described below.

[0072] In some embodiments, if the configuration parameters of a function node include the attributes of the input data and the output data of the function node, the preset rules may include: the attributes of the input data of the function node should be consistent with the attributes of the output data of the function node connected to the input end of the function node; and the attributes of the output data of the function node should be consistent with the attributes of the input data of the function node connected to the output end of the function node. Based on this, the accurate flow of data between each function node can be guaranteed, and the reliability of the target business process diagram can be improved.

[0073] In some embodiments, if the relationship between a function node and other function nodes forms a loop, the preset rule may include: the function node in the loop needs to have a judgment condition for terminating the loop; illustratively, the configuration parameters of the function node may include a maximum number of executions, and if the node execution number of the function node exceeds the maximum number of executions, the loop is terminated. Based on this, it is possible to avoid the formation of an infinite loop between the function nodes.

[0074] It should be understood that the above-mentioned preset rules are only exemplary descriptions. In actual applications, the setting of the preset rules makes the configuration parameters of each functional node match and the connection relationship between each functional node is logical. This application does not impose specific restrictions on this.

[0075] S103: Estimate the algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme in at least one hardware configuration scheme.

[0076] In some embodiments, the hardware resources include one or more of the following: computing resources, storage resources, and bandwidth resources. The algorithm performance includes service processing time and / or service processing effect. It should be understood that the hardware resources and algorithm performance are only examples, and this application does not impose specific restrictions on this.

[0077] In some embodiments, the above hardware configuration scheme may be a hardware configuration scheme generated after performing default splitting, merging, parallel processing or memory sharing on the corresponding executable files and configuration files in the target business process diagram that has been orchestrated.

[0078] In other embodiments, the above hardware configuration scheme may be a configuration scheme generated according to hardware allocation parameters set by a user.

[0079] S104: Prompt the algorithm performance of the engine package corresponding to each hardware configuration solution and the consumed hardware resources.

[0080] In some embodiments, when prompting the algorithm performance of the engine package corresponding to each hardware configuration scheme and the hardware resources consumed, the weights of various evaluation contents (such as storage resources or bandwidth resources) set by the user can also be received, and the algorithm performance of the engine package corresponding to each hardware configuration scheme and the hardware resources consumed can be scored according to the weights of each evaluation content.

[0081] Exemplarily, if the above-mentioned hardware resources include: computing power resources, storage resources, and bandwidth resources, and the above-mentioned algorithm performance includes: business processing time, and business processing effect; the evaluation weight of computing power resources is set to 1, and the weights of other evaluation contents are set to 0, then the scheme with the least computing power resources in each hardware configuration scheme can be evaluated. Alternatively, the evaluation weight of business processing time is set to 0.5, the evaluation weight of business processing effect is set to 0.5, and the weights of other evaluation contents are set to 0, then the scheme with better business processing time and business processing effect in each hardware configuration scheme can be evaluated. Furthermore, users can also select from each hardware configuration scheme the scheme with the least storage resources, the scheme with the least bandwidth resources, the scheme with the least processing time, the scheme with the best processing effect, the scheme with the best comprehensive hardware resources, or the scheme with the best comprehensive algorithm performance. It should be understood that the setting of the above evaluation weights is only an example, and the setting of the evaluation weights can be determined according to the specific needs of the user, and this application does not limit this.

[0082] In some embodiments, when prompting the algorithm performance of the engine package corresponding to each hardware configuration scheme and the hardware resources consumed, the management page of the hardware allocation parameters can also be displayed, so that when the user is not satisfied with the estimated results, the hardware allocation parameters can be reset to generate a new hardware configuration scheme. Exemplarily, the management page of the hardware allocation parameters includes hardware allocation parameters 1 and hardware allocation parameters 2; if adjusting hardware allocation parameters 1 can reduce business processing time but consume more computing power resources and bandwidth resources, and adjusting hardware allocation parameters 2 can improve business processing effects but occupy more storage resources, then the hardware allocation parameters set by the user can be received, so that the user can set the hardware allocation parameters by himself and generate different hardware configuration schemes.

[0083] Furthermore, after a new hardware configuration scheme is generated, the above steps S103 to S104 are repeatedly executed to re-estimate and prompt the algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme until the estimated result meets the user's requirements.

[0084] Based on this, users can select or set solutions that suit their device platforms based on resource evaluation results (the algorithm performance of the engine packages corresponding to the above-mentioned hardware configuration solutions and the hardware resources consumed). For example, for video AI tasks, the algorithm performance and verification results (including the hardware resources consumed) of the video AI task are estimated while meeting the bit rate and / or resolution supported by the device platform, so that users can select the hardware configuration solution that is suitable for the video AI task, and then send the output results (that is, the engine package corresponding to the hardware configuration solution) from the platform to the actual device platform for implementation. Furthermore, the problem of long cycles for iterative optimization of solution parameters of different AI algorithms on different hardware platforms in specific businesses can be solved, the version iteration cost caused by the balance between system resources and algorithm performance indicators is reduced, and the technical threshold for carrying out this work is also lowered.

[0085] It should be understood that Figure 3 The provided technical solution brings at least the following beneficial effects: by arranging the business solution through the target business process diagram, the reliance on program development experience can be reduced, thereby reducing the difficulty of program development; at the same time, the algorithm performance and consumed resources of the engine package corresponding to the business solution in each hardware configuration solution are evaluated, and the estimated results are prompted to the user so that the user can determine the business solution with the best resource performance, thereby optimizing the performance of the business solution.

[0086] In some embodiments, when the executable files and configuration files corresponding to each functional node are packaged into an engine package, simulation tests are also performed on the executable files and configuration files corresponding to each functional node. Figure 3 The method shown also includes: receiving a test data set uploaded by a user; wherein, in step S102, the executable files and configuration files corresponding to each functional node in the target business process diagram are packaged into an engine package, which is specifically implemented as the following steps: according to the executable files and configuration files corresponding to each functional node in the target business process diagram, the test data set is simulated and tested; if the result after the simulation test meets the user's requirements, the executable files and configuration files corresponding to each functional node in the target business process diagram are packaged into an engine package. Optionally, the above-mentioned simulation test can be performed on a specified hardware platform or on a cloud server in order to save test resources, and this application does not impose specific restrictions on this. Furthermore, if the result of the actual test does not meet the test requirements, a second warning message is issued to prompt the user.

[0087] Based on this, before packaging the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package, it is possible to first determine whether the simulation test results of the executable files and configuration files corresponding to the target process diagram meet the user requirements, so that the executable files and configuration files corresponding to each functional node are packaged only when they meet the requirements, thereby saving the amount of calculation. In addition, the feasibility and reliability of the business plan are guaranteed.

[0088] In some embodiments, after step S101 to step S104, the following steps are further included:

[0089] S105. In response to the user's selection instruction for the hardware configuration plan, a target hardware configuration plan is determined from at least one hardware configuration plan; the engine package is sent to the target hardware platform according to the target hardware configuration plan; the data to be processed uploaded by the user is received, and the data to be processed is sent to the target hardware platform, so that the target hardware platform runs the engine package according to the target hardware configuration plan to perform business processing on the data to be processed, and obtains a business processing result corresponding to the target business.

[0090] In some examples, the business processing result may include one or more of the following: a final result, an intermediate module output result (ie, an output result of each functional node), or a resource evaluation result (algorithm performance and / or consumed hardware resources).

[0091] For example, take the business of "license plate information recognition" as the target business. Figure 4 As shown, if the flowchart corresponding to the target business is constructed as: detection function node → segmentation function node → text recognition function node, the corresponding business processing results may include one or more of the following: license plate information, output results of the detection function node (such as license plate position information), output results of the segmentation function node (such as license plate image), text recognition function node (such as license plate text content), or resource evaluation results (such as license plate recognition rate, consumed storage resources, or consumed bandwidth resources, etc.).

[0092] Based on this, users can select the target hardware configuration plan that best meets their needs based on the estimated results of the algorithm performance and consumed resources of the engine package corresponding to the business plan in each hardware configuration plan, and further run the engine package corresponding to the target business flow chart according to the plan to process business data and obtain the business processing results corresponding to the target business.

[0093] To further illustrate the arrangement method of the service plan provided in the embodiment of the present application, the following takes "license plate information recognition" as an example of the target service to specifically illustrate the arrangement method of the service plan. Among them, the arrangement method of the service plan corresponding to "license plate information recognition" can be specifically implemented as the following process:

[0094] Obtaining the business keywords "license plate" and "identification" input by the user, displaying multiple business process diagrams related to "license plate" and "identification", and obtaining a target business process diagram selected by the user from the multiple business process diagrams; or,

[0095] Obtaining the business keywords "license plate" and "identification" input by the user, displaying multiple business process diagrams related to "license plate" and "identification", and obtaining a reference business process diagram selected by the user from the multiple business process diagrams; in response to the user's modification instruction for the reference business process diagram, modifying the reference business process diagram, and using the modified reference business process diagram as the target business process diagram; or,

[0096] A graphical draggable component pool is displayed, the component pool includes multiple function nodes; in response to the user dragging the function node in the component pool to the orchestration area and connecting the function nodes in the orchestration area, the business process diagram generated after connecting the function nodes in the orchestration area is used as the target process diagram. The following is an exemplary method for orchestrating a business process diagram through multiple function nodes in the component pool:

[0097] For example, the above-mentioned component pool includes reusable "detection function nodes", "segmentation function nodes" and "text recognition function nodes". Among them, the detection function node is used to perform semantic detection and positioning on the image. The user can define the detection type according to his needs. For example, the detection type can be human body, cat, sidewalk, water accumulation, vehicle, or license plate, etc. There is no restriction on this. The detection function node outputs the position of the image area corresponding to the detection type. The "segmentation function node" is used to segment the image area corresponding to the position information from the input image based on the input position information. The "text recognition function node" is used to recognize the meaning of the text in the image. If the logical implementation of the "license plate information recognition" business is arranged as: detect the vehicle in the image → segment the vehicle image → recognize the license plate information in the vehicle image, it is still like this Figure 4As shown, the flowchart corresponding to the target business can be constructed as: detection function node → segmentation function node → text recognition function node. Further, the attribute of the input data in the configuration parameters of the detection function node in the flowchart corresponding to the target business is set to image data, the attribute of the output data is set to position information and image data, and the detected content type is selected as "license plate"; the attribute of the input data in the configuration parameters of the segmentation function node is set to position information and image data, and the attribute of the output data is set to image data; the attribute of the input data in the configuration parameters of the text recognition function node in the flowchart corresponding to the target business is set to image data, and the attribute of the output data is set to text content. Based on this, when the initial image is input into the flowchart corresponding to the target business, the position of the license plate is detected and located in the detection function node, and the license plate position and the original image are output to the segmentation function node; the segmentation function node segments the license plate image from the original image according to the received license plate position and the original image, and outputs the segmented license plate image to the text recognition function node; the text recognition function node recognizes the text content of the received license plate image, thereby realizing license plate information recognition, and completing the construction of the target business flowchart.

[0098] Get the configuration parameters of each functional node in the target business process diagram.

[0099] If the connection relationship between the functional nodes of the target business process diagram and the configuration parameters of the functional nodes of the target business process diagram meet the preset rules, the executable files and configuration files corresponding to the functional nodes of the target business process diagram are selected from the database.

[0100] Receive the test data set uploaded by the user; perform simulation test on the test data set according to the executable files and configuration files corresponding to each functional node in the target business process diagram; if the result after the simulation test meets the user's requirements, package the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package.

[0101] The algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme in at least one hardware configuration scheme are estimated.

[0102] The algorithm performance of the engine package corresponding to each hardware configuration scheme and the hardware resources consumed are prompted so that users can select the hardware configuration scheme that meets their needs according to the prompts.

[0103] In response to a user's selection instruction for a hardware configuration plan, a target hardware configuration plan is determined from at least one hardware configuration plan; an engine package is sent to a target hardware platform according to the target hardware configuration plan; data to be processed uploaded by the user is received, and the data to be processed is sent to the target hardware platform, so that the target hardware platform runs the engine package according to the target hardware configuration plan to perform business processing on the data to be processed, and obtains a business processing result corresponding to the target business.

[0104] It can be seen that the above mainly introduces the solution provided by the embodiment of the present application from the perspective of the method. In order to achieve the above functions, the embodiment of the present application provides a hardware structure and / or software module corresponding to each function. It should be easily appreciated by those skilled in the art that, in combination with the modules and algorithm steps of each example described in the embodiment disclosed herein, the embodiment of the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0105] like Figure 5 As shown, the embodiment of the present application provides a business plan arrangement device for executing Figure 3 The arrangement method of the business plan shown in the figure. The arrangement device 300 of the business plan includes: a process acquisition module 301, a file configuration module 302, and a resource estimation module 303.

[0106] The process acquisition module 301 is used to acquire a target business process diagram, which includes multiple function nodes and connection relationships between the multiple function nodes. The target business process diagram is used to describe the process of implementing the target business.

[0107] The file configuration module 302 is used to obtain executable files and configuration files corresponding to each functional node in the target business process diagram, and package the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package.

[0108] The resource estimation module 303 is used to estimate the algorithm performance and the hardware resources consumed by the engine package corresponding to each hardware configuration scheme in at least one hardware configuration scheme; and prompt the algorithm performance and the hardware resources consumed by the engine package corresponding to each hardware configuration scheme.

[0109] In some embodiments, the process acquisition module 301 is specifically used to acquire business keywords input by the user, and display multiple business process diagrams related to the business keywords according to the business keywords; and acquire a target business process diagram selected by the user from the multiple business process diagrams.

[0110] In some embodiments, the above-mentioned process acquisition module 301 is specifically used to obtain business keywords input by the user, and display multiple business process diagrams related to the business keywords based on the business keywords; obtain a reference business process diagram selected by the user from multiple business process diagrams; in response to the user's modification instruction to the reference business process diagram, modify the reference business process diagram, and use the modified reference business process diagram as the target business process diagram.

[0111] In some embodiments, the above-mentioned process acquisition module 301 is specifically used to display a graphical draggable component pool, which includes multiple functional nodes; in response to the user's instruction to drag the functional node in the component pool to the orchestration area and connect the various functional nodes in the orchestration area, the business process diagram generated after connecting the various functional nodes in the orchestration area is used as the target business process diagram.

[0112] In some embodiments, the above-mentioned file configuration module 302 is specifically used to obtain the configuration parameters of each functional node in the target business process diagram; determine whether the connection relationship between the various functional nodes of the target business process diagram and the configuration parameters of each functional node of the target business process diagram meet the preset rules; if the preset rules are met, select the executable files and configuration files corresponding to each functional node of the target business process diagram from the database; the configuration parameters of the above-mentioned functional nodes include the model files corresponding to the functional nodes, and the model files corresponding to the functional nodes are determined according to the following methods: receiving the training data corresponding to the functional nodes uploaded by the user, and obtaining the model files corresponding to the functional nodes according to the training data; or, receiving the model files uploaded by the user for the functional nodes, and using the model files uploaded by the user for the functional nodes as the model files corresponding to the functional nodes.

[0113] In some embodiments, the above-mentioned file configuration module 302 is also used to receive a test data set uploaded by a user; the above-mentioned file configuration module 302 is specifically used to perform simulation testing on the test data set according to the executable files and configuration files corresponding to each functional node in the target business process diagram; if the result after the simulation test meets the user requirements, the executable files and configuration files corresponding to each functional node in the target business process diagram are packaged into an engine package.

[0114] In some embodiments, the above-mentioned device also includes a business processing module 304, which is used to respond to the user's selection instruction for the hardware configuration plan, determine the target hardware configuration plan from at least one hardware configuration plan; send the engine package to the target hardware platform according to the target hardware configuration plan; receive the data to be processed uploaded by the user, and send the data to be processed to the target hardware platform, so that the target hardware platform runs the engine package according to the target hardware configuration plan to perform business processing on the data to be processed, and obtains the business processing result corresponding to the target business.

[0115] It should be noted that Figure 5 The division of modules in the example is schematic and is only a logical function division. There may be other division methods in actual implementation. For example, two or more functions may be integrated into one processing module. The above integrated modules may be implemented in the form of hardware or software function modules.

[0116] In the case of implementing the functions of the above-mentioned integrated modules in the form of hardware, the embodiment of the present application provides another possible structural diagram of the arrangement device of the service scheme involved in the above-mentioned embodiment. Figure 6 As shown, the service solution arrangement device 400 includes: a processor 402 and a bus 404. Optionally, the service solution arrangement device may further include a memory 401; optionally, the service solution arrangement device may further include a communication interface 403.

[0117] The processor 402 may be a processor that implements or executes various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of the present application. The processor 402 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of the present application. The processor 402 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0118] The communication interface 403 is used to connect with other devices via a communication network, such as Ethernet, wireless access network, wireless local area network (WLAN), etc.

[0119] The memory 401 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program codes in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0120] As a possible implementation, the memory 401 may exist independently of the processor 402, and the memory 401 may be connected to the processor 402 via a bus 404 to store instructions or program codes. When the processor 402 calls and executes the instructions or program codes stored in the memory 401, the method for arranging the service solution provided in the embodiment of the present application can be implemented.

[0121] In another possible implementation, the memory 401 may also be integrated with the processor 402 .

[0122] The bus 404 may be an extended industry standard architecture (EISA) bus, etc. The bus 404 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0123] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the orchestration device of the business plan is divided into different functional modules to complete all or part of the functions described above.

[0124] The embodiment of the present application also provides a computer-readable storage medium. All or part of the processes in the above method embodiments can be completed by computer instructions to instruct the relevant hardware, and the program can be stored in the above computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments. The computer-readable storage medium can be the memory or memory of any of the above embodiments. The above computer-readable storage medium can also be an external storage device of the above-mentioned orchestration device based on the business solution, such as a plug-in hard disk, a smart memory card (smart media card, SMC), a secure digital (secure digital, SD) card, a flash card (flash card), etc. equipped on the orchestration device of the above business solution. Further, the above computer-readable storage medium can also include both the internal storage unit of the orchestration device of the above business solution and an external storage device. The above computer-readable storage medium is used to store the above computer program and other programs and data required by the orchestration device of the above business solution. The above computer-readable storage medium can also be used to temporarily store data that has been output or is to be output.

[0125] An embodiment of the present application also provides a computer program product, which includes a computer program. When the computer program product runs on a computer, the computer executes any one of the service scheme arrangement methods provided in the above embodiments.

[0126] Although the present application is described herein in conjunction with various embodiments, in the process of implementing the claimed application, those skilled in the art may understand and implement other changes to the disclosed embodiments by viewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "one" or "an" does not exclude multiple situations. A single processor or other unit can implement several functions listed in the claims. Certain measures are recorded in different dependent claims, but this does not mean that these measures cannot be combined to produce good results.

[0127] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely exemplary illustrations of the present application as defined by the appended claims, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

[0128] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for arranging a business plan, It is characterized in that The method comprises: Acquire a target business process diagram, wherein the target business process diagram includes a plurality of function nodes and connection relationships between the plurality of function nodes, and the target business process diagram is used to describe a process of implementing a target business; Obtaining executable files and configuration files corresponding to each functional node in the target business process diagram, and packaging the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package; estimating the algorithm performance and the hardware resources consumed by the engine package corresponding to each hardware configuration scheme in at least one hardware configuration scheme; the hardware configuration scheme is generated by splitting, merging, parallel processing or memory sharing processing of executable files and configuration files in the engine package, or the hardware configuration scheme is generated according to hardware allocation parameters set by a user; The algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme are prompted, and a management page of hardware allocation parameters is displayed. The management page of hardware allocation parameters is used to adjust the hardware allocation parameters to generate a new hardware configuration scheme through the adjusted hardware allocation parameters.

2. The method according to claim 1, It is characterized in that The step of obtaining the target business process diagram includes: Acquire a business keyword input by a user, and display multiple business process diagrams related to the business keyword according to the business keyword; A target business process diagram selected by a user from the multiple business process diagrams is obtained.

3. The method according to claim 1, It is characterized in that The step of obtaining the target business process diagram includes: Acquire a business keyword input by a user, and display multiple business process diagrams related to the business keyword according to the business keyword; Obtaining a reference business process diagram selected by a user from the multiple business process diagrams; In response to a user's modification instruction for the reference business process diagram, the reference business process diagram is modified, and the modified reference business process diagram is used as the target business process diagram.

4. The method according to claim 1, It is characterized in that The step of obtaining the target business process diagram includes: Displaying a graphical draggable component pool, wherein the component pool includes a plurality of function nodes; In response to the user's instruction to drag the function nodes in the component pool to the orchestration area and connect the function nodes in the orchestration area, the business process diagram generated after connecting the function nodes in the orchestration area is used as the target business process diagram.

5. The method according to claim 1, It is characterized in that The obtaining of executable files and configuration files corresponding to each functional node of the target business process diagram includes: Obtaining configuration parameters of each functional node in the target business process diagram; Determine whether the connection relationship between each functional node of the target business process diagram and the configuration parameters of each functional node of the target business process diagram meet the preset rules; If the preset rules are met, the executable files and configuration files corresponding to the functional nodes of the target business process diagram are selected from the database.

6. The method according to claim 5, It is characterized in that The configuration parameters of the functional node include a model file corresponding to the functional node, and the model file corresponding to the functional node is determined according to the following method: receiving training data corresponding to the function node uploaded by a user, and obtaining a model file corresponding to the function node through training according to the training data; or, A model file uploaded by a user for the function node is received, and the model file uploaded by the user for the function node is used as the model file corresponding to the function node.

7. The method according to claim 1, It is characterized in that The method further comprises: Receive test data sets uploaded by users; The step of packaging the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package includes: Performing simulation testing on the test data set according to the executable files and configuration files corresponding to each functional node in the target business process diagram; If the result after the simulation test meets the user's requirements, the executable files and configuration files corresponding to each functional node in the target business process diagram are packaged into an engine package.

8. The method according to any one of claims 1 to 7, It is characterized in that The method further comprises: In response to a user's selection instruction for a hardware configuration scheme, determining a target hardware configuration scheme from the at least one hardware configuration scheme; Sending the engine package to the target hardware platform according to the target hardware configuration solution; Receive the data to be processed uploaded by the user, and send the data to be processed to the target hardware platform, so that the target hardware platform runs the engine package to perform business processing on the data to be processed according to the target hardware configuration plan, and obtains the business processing result corresponding to the target business.

9. A device for arranging a business plan, It is characterized in that include: A process acquisition module is used to acquire a target business process diagram, wherein the target business process diagram includes a plurality of functional nodes and connection relationships between the plurality of functional nodes, and the target business process diagram is used to describe a process of realizing a target business; A file configuration module, used to obtain executable files and configuration files corresponding to each functional node in the target business process diagram, and package the executable files and configuration files corresponding to each functional node in the target business process diagram into an engine package; A resource estimation module, used to estimate the algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme in at least one hardware configuration scheme; prompt the algorithm performance and consumed hardware resources of the engine package corresponding to each hardware configuration scheme, and display the management page of hardware allocation parameters; the hardware configuration scheme is generated by splitting, merging, parallel processing or memory sharing processing of executable files and configuration files in the engine package, or the hardware configuration scheme is generated according to the hardware allocation parameters set by the user; The management page of the hardware allocation parameters is used to adjust the hardware allocation parameters so as to generate a new hardware configuration scheme through the adjusted hardware allocation parameters.

10. The device according to claim 9, It is characterized in that It also includes a business processing module; The process acquisition module is specifically used to acquire the business keyword input by the user, and display multiple business process diagrams related to the business keyword according to the business keyword; and acquire the target business process diagram selected by the user from the multiple business process diagrams; The process acquisition module is specifically used to acquire the business keywords input by the user, and display multiple business process diagrams related to the business keywords according to the business keywords; and acquire a reference business process diagram selected by the user from the multiple business process diagrams; In response to a modification instruction of the user on the reference business process diagram, modify the reference business process diagram, and use the modified reference business process diagram as the target business process diagram; The process acquisition module is specifically used to display a graphical draggable component pool, wherein the component pool includes multiple function nodes; in response to a user's instruction to drag a function node in the component pool to an orchestration area and connect each function node in the orchestration area, a business process diagram generated by connecting each function node in the orchestration area is used as a target business process diagram; The file configuration module is specifically used to obtain the configuration parameters of each functional node in the target business process diagram; determine whether the connection relationship between the functional nodes of the target business process diagram and the configuration parameters of the functional nodes of the target business process diagram meet the preset rules; if the preset rules are met, select the executable files and configuration files corresponding to the functional nodes of the target business process diagram from the database; The configuration parameters of the functional node include a model file corresponding to the functional node, and the model file corresponding to the functional node is determined according to the following method: receiving training data corresponding to the functional node uploaded by a user, and obtaining the model file corresponding to the functional node through training according to the training data; or receiving a model file uploaded by a user for the functional node, and using the model file uploaded by the user for the functional node as the model file corresponding to the functional node; The file configuration module is further used to receive a test data set uploaded by a user; the file configuration module is specifically used to perform simulation testing on the test data set according to the executable files and configuration files corresponding to each functional node in the target business process diagram; If the result after the simulation test meets the user's requirements, the executable files and configuration files corresponding to each functional node in the target business process diagram are packaged into an engine package; The business processing module is used to determine a target hardware configuration scheme from the at least one hardware configuration scheme in response to a user's selection instruction for the hardware configuration scheme; Sending the engine package to the target hardware platform according to the target hardware configuration solution; Receive the data to be processed uploaded by the user, and send the data to be processed to the target hardware platform, so that the target hardware platform runs the engine package to perform business processing on the data to be processed according to the target hardware configuration plan, and obtains the business processing result corresponding to the target business.

11. A device for arranging a business plan, It is characterized in that include: A memory and a processor; the memory and the processor are coupled; the memory is used to store computer program code, the computer program code includes computer instructions; When the processor executes the computer instructions, the service plan arrangement device executes the service plan arrangement method according to any one of claims 1 to 8.

12. A computer-readable storage medium, It is characterized in that The computer-readable storage medium includes computer instructions; Wherein, when the computer instructions are executed on a device for arranging a business plan, the device for arranging a business plan executes the method for arranging a business plan as claimed in any one of claims 1 to 8.

Citation Information

Patent Citations

  • Service processing method, device and equipment

    CN112099848A

  • Business process arrangement method and device, computer equipment and storage medium

    CN113435846A