Template generation method, device and storage medium for artificial intelligence AI solutions

Abstract

The present invention discloses a method, an apparatus, and a storage medium for generating a template of an artificial intelligence (AI) solution related to artificial intelligence and associated with technologies such as computer vision and natural language processing. The method includes: obtaining an addition operation instruction triggered in a workflow configuration interface, where the addition operation instruction is used to indicate adding a set of work components matching a target scenario to a workflow template, and the work components include AI components composed of functional modules encapsulating AI algorithms; configuring component attribute information to be configured in the work components to obtain target component attribute information; configuring running relationships to be configured between the work components to obtain target running relationships; and generating an AI solution template applied to the target scenario according to the target component attribute information and the target running relationships. The present invention solves the technical problem of low efficiency in generating templates for AI solutions.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence (AI), and more particularly, to a method and apparatus for generating a template of an AI solution and a storage medium. Background Art

[0002] In recent years, the demand for solutions to in-depth industry-specific problems in various sub-vertical scenarios has been increasing. However, in the case of accessing a new customer project in the prior art, a large amount of R & D manpower needs to be invested. By manually writing relevant scripts, a template of a solution that meets the requirements is obtained to achieve access to the customer scenario. However, manual writing not only increases costs but also causes waste of time due to complex processes and redundant steps. Therefore, there is a problem of low efficiency in generating templates for AI solutions in the prior art.

[0003] In response to the above problems, no effective solution has been proposed yet. Summary of the Invention

[0004] Embodiments of the present invention provide a method and apparatus for generating a template of an AI solution and a storage medium, so as to at least solve the technical problem of low efficiency in generating templates for AI solutions.

[0005] According to one aspect of the embodiments of the present invention, a method for generating a template of an AI solution is provided, including: obtaining an add operation instruction triggered in a workflow configuration interface, where the add operation instruction is used to indicate adding a set of work components matching a target scenario in a workflow template, and the work components include AI components composed of functional modules encapsulating AI algorithms; configuring component attribute information to be configured in the work components to obtain target component attribute information; configuring an operation relationship to be configured between the work components to obtain a target operation relationship; and generating an AI solution template applied to the target scenario according to the target component attribute information and the target operation relationship.

[0006] According to another aspect of the embodiments of the present invention, there is also provided a template generation device for an artificial intelligence (AI) solution, including: an acquisition unit, configured to acquire an addition operation instruction triggered in a workflow configuration interface, where the addition operation instruction is used to indicate adding a set of work components matching a target scenario in a workflow template, and the work components include AI components composed of functional modules encapsulating AI algorithms; a first configuration unit, configured to configure component attribute information to be configured in the work components to obtain target component attribute information; a second configuration unit, configured to configure an operation relationship to be configured between the work components to obtain a target operation relationship; and a first generation unit, configured to generate an AI solution template applied to the target scenario according to the target component attribute information and the target operation relationship.

[0007] According to yet another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium storing a computer program, where the computer program is configured to execute the above-mentioned template generation method for an artificial intelligence (AI) solution when running.

[0008] According to yet another aspect of the embodiments of the present invention, there is also provided an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the above-mentioned template generation method for an artificial intelligence (AI) solution through the computer program.

[0009] In the embodiments of the present invention, an addition operation instruction triggered in a workflow configuration interface is acquired, where the addition operation instruction is used to indicate adding a set of work components matching a target scenario in a workflow template, and the work components include AI components composed of functional modules encapsulating AI algorithms; the component attribute information to be configured in the work components is configured to obtain target component attribute information; the operation relationship to be configured between the work components is configured to obtain a target operation relationship; and an AI solution template applied to the target scenario is generated according to the target component attribute information and the target operation relationship. Through a set of components that can be flexibly added and combined, an AI solution template applicable to the target scenario is efficiently generated, thereby achieving the purpose of reducing redundant steps in generating the AI solution template, and thus realizing the technical effect of improving the generation efficiency of the AI solution template, and further solving the technical problem of low generation efficiency of the template for the artificial intelligence (AI) solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0011] Figure 1 It is a schematic diagram of an application environment of a template generation method for an optional artificial intelligence (AI) solution according to an embodiment of the present invention;

[0012] Figure 2 It is a schematic diagram of a flowchart of a template generation method for an optional artificial intelligence (AI) solution according to an embodiment of the present invention;

[0013] Figure 3 It is a schematic diagram of a template generation method for an optional artificial intelligence (AI) solution according to an embodiment of the present invention;

[0014] Figure 4 It is a schematic diagram of another optional template generation method for an artificial intelligence (AI) solution according to an embodiment of the present invention;

[0015] Figure 5 It is a schematic diagram of another optional template generation method for an artificial intelligence (AI) solution according to an embodiment of the present invention;

[0016] Figure 6 It is a schematic diagram of another optional template generation method for an artificial intelligence (AI) solution according to an embodiment of the present invention;

[0017] Figure 7 It is a schematic diagram of another optional template generation method for an artificial intelligence (AI) solution according to an embodiment of the present invention;

[0018] Figure 8 It is a schematic diagram of another optional template generation method for an artificial intelligence (AI) solution according to an embodiment of the present invention;

[0019] Figure 9 It is a schematic diagram of another optional template generation method for an artificial intelligence (AI) solution according to an embodiment of the present invention;

[0020] Figure 10 It is a schematic diagram of another optional template generation method for an artificial intelligence (AI) solution according to an embodiment of the present invention;

[0021] Figure 11 It is a schematic diagram of another optional template generation method for an artificial intelligence (AI) solution according to an embodiment of the present invention;

[0022] Figure 12 It is a schematic diagram of an optional template generation device for an artificial intelligence (AI) solution according to an embodiment of the present invention;

[0023] Figure 13 It is a schematic diagram of the structure of an optional electronic device according to an embodiment of the present invention. Detailed implementation manners

[0024] To enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0025] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0026] Artificial Intelligence (AI) is a theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a way similar to human intelligence. Artificial intelligence also studies the design principles and implementation methods of various intelligent machines to enable the machines to have the functions of perception, reasoning, and decision-making.

[0027] Artificial intelligence technology is an interdisciplinary subject involving a wide range of fields, including both hardware-level technologies and software-level technologies. The basic technologies of artificial intelligence generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, and mechatronics. The software technologies of artificial intelligence mainly include several major directions such as computer vision technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0028] Computer Vision Technology (CV) Computer vision is a science that studies how to enable machines to "see". More specifically, it refers to machine vision that uses cameras and computers to replace human eyes for tasks such as object recognition, tracking, and measurement, and further performs image processing to make the computer-processed images more suitable for human eye observation or transmission to instrument detection. As a scientific discipline, computer vision studies related theories and technologies and attempts to build artificial intelligence systems that can obtain information from images or multi-dimensional data. Computer vision technology usually includes technologies such as image processing, image recognition, image semantic understanding, image retrieval, OCR, video processing, video semantic understanding, video content / behavior recognition, 3D object reconstruction, 3D technology, virtual reality, augmented reality, simultaneous localization and mapping, etc., and also includes common biometric recognition technologies such as face recognition and fingerprint recognition.

[0029] Natural Language Processing (NLP) is an important direction in the fields of computer science and artificial intelligence. It studies various theories and methods that can achieve effective communication between humans and computers in natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Therefore, the research in this field will involve natural language, that is, the language people use in daily life, so it has a close connection with the research of linguistics. Natural language processing technology usually includes technologies such as text processing, semantic understanding, machine translation, robot question answering, knowledge graph, etc.

[0030] With the research and progress of artificial intelligence technology, artificial intelligence technology has been studied and applied in multiple fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, driverless, autonomous driving, drones, robots, smart healthcare, smart customer service, etc. It is believed that with the development of technology, artificial intelligence technology will be applied in more fields and play an increasingly important role.

[0031] The solution provided by the embodiments of this application involves technologies such as natural language processing and computer vision of artificial intelligence, and will be specifically described through the following embodiments:

[0032] According to one aspect of the embodiments of the present invention, a method for generating a template of an artificial intelligence (AI) solution is provided. Optionally, as an alternative implementation manner, the above method for generating a template of an AI solution can be, but is not limited to, applied to, for example Figure 1In the environment shown. Among them, it may but is not limited to including user device 102, network 110 and server 112. Among them, the user device 102 may but is not limited to include a display 108, a processor 106 and a memory 104. The display 108 may but is not limited to be used to display a workflow configuration interface 1024, and a workflow template 1026 displayed on the workflow configuration interface 1024, and work components 1022 that can be added to the workflow template.

[0033] The specific process can be as follows:

[0034] Step S102, the user device 102 obtains an add operation instruction triggered in the workflow configuration interface 1024, where the add operation instruction is used to indicate adding the work component 1022 to the workflow template 1026

[0035] Step S104, the user device 102 obtains a configuration operation instruction triggered in the workflow configuration interface 1024, and obtains the component attribute information of all configured work components and the connection relationship between the work components. The configuration operation instruction is used to configure the component attribute information of all work components (including the added work component 1022) in the workflow template 1026 and the connection relationship between the components;

[0036] Steps S106 - S108, the user device 102 sends the add operation instruction, the component attribute information and the connection relationship between the components to the server 112 through the network 110;

[0037] Step S110, the server 112 searches for relevant data corresponding to the operation instruction, the component attribute information and the connection relationship between the components through the database 114, and processes the operation instruction, the component attribute information and the connection relationship between the components through the processing engine 116, so as to generate an AI solution template;

[0038] Steps S112 - S114, the server 112 sends the AI solution template to the user device 102 through the network 110. The processor 106 in the user device 102 displays the generated AI solution template on the display 108, and stores the generated AI solution template in the memory 104.

[0039] It should be noted that the above method for generating a template of the artificial intelligence AI solution is only described as an example, and there is no restriction on the triggering order of the add operation instruction and the configuration operation instruction.

[0040] Optionally, as an alternative implementation, as Figure 2 shown, the method for generating a template of the artificial intelligence AI solution includes:

[0041] S202, obtain the addition operation instruction triggered in the workflow configuration interface, where the addition operation instruction is used to indicate adding a set of work components matching the target scenario to the workflow template, and the work components include AI components composed of functional modules encapsulating AI algorithms;

[0042] S204, configure the component attribute information to be configured in the work components to obtain the target component attribute information;

[0043] S206, configure the running relationship to be configured between the work components to obtain the target running relationship;

[0044] S208, generate an AI solution template applied to the target scenario according to the target component attribute information and the target running relationship.

[0045] Optionally, in this embodiment, the method for generating the template of the artificial intelligence AI solution may be, but is not limited to, applied to the scenario of solving business problems. The triggering method of the addition operation instruction may be, but is not limited to, a triggering method of moving or transferring the identifier representing the component to be added from the first position area (such as the component library on one side of the workflow configuration interface) to the second position area (such as the workflow template on one side of the workflow configuration interface), such as dragging, path transfer, etc. The work components may be, but are not limited to, functional modules that encapsulate AI capabilities and / or links in the AI process in the form of components and can be selectively added in the sidebar of the visual interface (workflow configuration interface). The component attribute information to be configured may be, but is not limited to, a set of configuration and function setting information required for the current component, such as name, type, attribute, etc. The running relationship to be configured may be, but is not limited to, used to represent the connection relationship and running logic between components, such as the first component executes first, and the second component executes when the first component finishes execution or the execution process meets a preset degree, etc. The AI solution template may be, but is not limited to, a template that can be flexibly configured or matched for solving the target AI task in the feature scenario. For example, an AI task is created based on the generated AI solution template, and a solution is run with the data collected by the device or the data stored in the data center, where the solution is used to solve the AI task.

[0046] It should be noted that an add operation instruction triggered in the workflow configuration interface is obtained, wherein the add operation instruction is used to instruct the addition of a group of work components matching the target scenario in the workflow template, wherein the work components include AI components composed of functional modules encapsulating AI algorithms; the component attribute information to be configured in the work components is configured to obtain the target component attribute information; the operating relationship to be configured between the work components is configured to obtain the target operating relationship; and an AI solution template applied in the target scenario is generated according to the target component attribute information and the target operating relationship.

[0047] Further examples are optional, such as Figure 3 As shown, it includes a workflow configuration interface 302, a workflow template 304 in the workflow configuration interface 302, and a visualization interface 306 for placing addable work components, wherein the visualization interface 306 includes work components (ordinary cameras 308). The specific steps are as follows: Step S302, obtain the add operation instruction triggered in the workflow configuration interface 302, and the front-end server adds the operation instruction accordingly. The display screen is to move the ordinary camera 308 from the visualization interface 306 where the work component can be added to the workflow template 304. Optionally, the ordinary camera 308 in the visualization interface 306 is retained, and an ordinary camera 308 is generated on the workflow template 304.

[0048] Further examples are optional, such as Figure 4 As shown, it is possible but not limited to triggering configuration operation instructions on the component attribute information configuration interface 402 to modify the component attribute information of the ordinary camera 305 in the workflow template 304, wherein the component attribute information of the ordinary camera 305 that can be modified includes, for example, device component type, device component parameter name, device component attribute value, and adding or deleting device components.

[0049] Further examples are optional, such as Figure 5 As shown, the visualization interface 306 includes work components that can be added to the workflow template 304, such as a general camera 305, a human attribute request 502, a human attribute 504, a human attribute result 506, a data storage 508, etc.;

[0050] Furthermore, at the current moment, the workflow template 304 includes a group of work components, namely, ordinary camera 305, human attribute request 502, human attribute 504, human attribute result 506, and data storage 508, and there is a connection relationship between the above group of work components (as shown by the arrows), wherein the connection relationship indicates the operation relationship between the work components, for example, the ordinary camera 308 is executed first, and the data storage 508 is executed last.

[0051] Through the embodiments provided in this application, an addition operation instruction triggered in the workflow configuration interface is obtained. The addition operation instruction is used to indicate adding a set of work components that match the target scenario in the workflow template. The work components include AI components composed of functional modules encapsulating AI algorithms. Configure the component attribute information to be configured in the work components to obtain target component attribute information. Configure the running relationship to be configured between the work components to obtain the target running relationship. According to the target component attribute information and the target running relationship, generate an AI solution template applied to the target scenario. Through a set of components that can be flexibly added and combined, an AI solution template that can be applied to the target scenario is efficiently generated, thereby achieving the purpose of reducing the redundant steps in generating the AI solution template, and thus realizing the technical effect of improving the generation efficiency of the AI solution template.

[0052] As an optional solution, configuring the component attribute information to be configured in the work components to obtain the target component attribute information includes:

[0053] S1. Determine the component type of the work component;

[0054] S2. According to the component type, determine the component attribute information to be configured;

[0055] S3. Obtain a first configuration operation instruction, where the first configuration operation instruction is used to configure the component attribute information to be configured.

[0056] It should be noted that determine the component type of the work component; according to the component type, determine the component attribute information to be configured; obtain a first configuration operation instruction, where the first configuration operation instruction is used to configure the component attribute information to be configured. Optionally, the component type is used to represent the way of encapsulating the AI capabilities and the links in the AI process in the form of components. For example, the component encapsulating the data storage link in the process can be but is not limited to a data storage component, etc.

[0057] For further illustration by example, optionally, for example Figure 6 As shown, determine that the component type of the ordinary camera 308 is the device component type 602, and then obtain the component attribute information to be configured corresponding to the device component type 602;

[0058] Further, obtain a first configuration operation instruction on the component attribute information configuration interface 402, and configure the component attribute information of the ordinary camera 308 of the device component type 602.

[0059] Through the embodiments provided in this application, the component type of the working component is determined; according to the component type, the component attribute information to be configured is determined; a first configuration operation instruction is obtained, where the first configuration operation instruction is used to configure the component attribute information to be configured, thereby achieving the purpose of specifically matching the corresponding component attribute information to be configured for different component types, and thus realizing the effect of improving the configuration efficiency of the component attribute information to be configured.

[0060] As an optional solution, determining the component attribute information to be configured according to the component type includes:

[0061] S1. When the component type of the working component includes the AI component type, determining the component attribute information to be configured includes at least one of the following: function configuration information, material library information;

[0062] S2. When the component type is the data storage component type, determining the component attribute information to be configured includes data storage information;

[0063] S3. When the component type includes the device component type, determining the component attribute information to be configured includes device configuration information;

[0064] S4. When the component type is the script component type, determining the component attribute information to be configured includes script content information;

[0065] S5. When the component type is the data access component type, determining the component attribute information to be configured includes access configuration information.

[0066] It should be noted that when the component type of the working component includes the AI component type, determining the component attribute information to be configured includes at least one of the following: function configuration information, material library information; when the component type is the data storage component type, determining the component attribute information to be configured includes data storage information; in this case, determining the component attribute information to be configured includes device configuration information; when the component type is the script component type, determining the component attribute information to be configured includes script content information; when the component type is the data access component type, determining the component attribute information to be configured includes access configuration information.

[0067] For further illustration by way of example, optionally, for example Figure 7As shown, obtain the first configuration operation instruction. When the first configuration operation instruction is used for the configuration data storage 508 (the shaded part indicates being selected) in the workflow template 304, display the component attribute information interface 702 on the workflow configuration interface 302. The component attribute information interface 702 includes modifiable component configuration attributes corresponding to the working component data storage 508, such as storage type, storage medium, database name, table name, etc., and also includes non-modifiable component configuration attributes, such as component name. Among them, a new database table for storing data can be created by, but not limited to, an operation instruction triggered at the "New Table" position on the component attribute information interface 702.

[0068] For further illustration by example, optionally, for example, an application embodiment in a certain traffic scenario Figure 8 As shown, obtain the first configuration operation instruction. When the first configuration operation instruction is used for the AI configuration of the ordinary camera 308 (the shaded part indicates being selected) in the workflow template 304, display the component attribute information interface 802 on the workflow configuration interface 302. The component attribute information interface 802 includes modifiable AI configuration attributes corresponding to the working component ordinary camera 308, such as "Video Area Recognition Configuration", "Select View Library", "Search by Image"; among them, optionally, the modification of the AI configuration attributes can meet different functional requirements in different scenarios. For example, if it is necessary to count the number of people, then select "Video Area Recognition Configuration" in the settings and modify "Video Area Recognition Configuration" to "Configuration by Location".

[0069] It should be noted that the working component is not limited to a component of only a single component type. For example, the device component includes an AI component, or a combination of other types of components.

[0070] For further illustration by example, optionally, for example Figure 8As shown, the ordinary camera 308 of the device component type has an AI configuration and can implement the function of real-time video analysis. Among them, real-time video analysis can, but is not limited to, the video of camera devices (including ordinary cameras and AI cameras) accessed through the device center can be viewed in real time, and the results output by the AI tasks of the video recognition type in the AI studio can also be displayed on the video screen, so as to achieve the effects of project demonstration, POC, and external user experience. Specifically, the implemented real-time video analysis function can, but is not limited to, include at least one of the following: cameras (device components) can be selected according to the grouped structure, the video can be viewed in real time after the camera is selected, if the camera is associated with a certain AI task, the task name is displayed, and the results of the AI task are superimposed on the video. If there are multiple AI results, the user can select which one to superimpose, and only one can be superimposed at a time. A warning can be triggered when the threshold is reached. In addition, not all AI tasks require the function of real-time video analysis. In other words, the examples here are only for AI tasks whose data source is a camera, and do not limit the application scenarios of this application in any way.

[0071] Through the embodiments provided by this application, when the component type of the working component includes the AI component type, the determined component attribute information to be configured includes at least one of the following: function configuration information, material library information; when the component type is the data storage component type, the determined component attribute information to be configured includes data storage information; in this case, the determined component attribute information to be configured includes device configuration information; when the component type is the script component type, the determined component attribute information to be configured includes script content information; when the component type is the data access component type, the determined component attribute information to be configured includes access configuration information, thereby achieving the purpose that the component attribute information to be configured can meet different functional requirements in different scenarios, and thus realizing the effect of improving the configuration flexibility of the component attribute information.

[0072] As an optional solution, configuring the running relationship to be configured between working components to obtain the target running relationship includes:

[0073] S1, determining the node type of the working component;

[0074] S2, determining the running relationship to be configured between working components according to the node type;

[0075] S3, obtaining a second configuration operation instruction, where the second configuration operation instruction is used to configure the running relationship to be configured between working components.

[0076] Optionally, the node type can be, but is not limited to, the distribution type of work components in the form of nodes on the workflow template, such as the connection type with other nodes (single-line connection, branch connection, convergence connection, etc.), the running status type (such as single run, multiple runs, loop run, etc.).

[0077] It should be noted that the node type of the work component is determined; according to the node type, the running relationship to be configured between work components is determined; a second configuration operation instruction is obtained, where the second configuration operation instruction is used to configure the running relationship to be configured between work components. Optionally, the running relationship can be, but is not limited to, the running relationship between work components, or can be, but is not limited to, the running logic relationship of a single component itself, etc.

[0078] For further illustration by way of example, optionally Figure 9 As shown, the component node information interface 902 of the ordinary camera 308 is obtained, where the component node information interface 902 is used to represent the running logic and connection relationship (connection relationship with other nodes) of the components in the form of nodes on the workflow template 304, etc.;

[0079] Furthermore, for the ordinary camera 308 in the form of a node on the workflow template 304, the connection method with other components in the form of nodes on the workflow template 304 is a single-line connection 904. Then, optionally, the connection relationship displayed on the component node information interface 902 is "single-line connection (Normal)"; the second configuration operation instruction triggered on the component node information interface 902 is obtained, and the running relationship of the ordinary camera 308 is configured, such as configuring the "running logic" to be "single run (single)" or "loop run (loop)", modifying the connection relationship between nodes (components on the workflow template 304), etc.

[0080] Through the embodiments provided in the present application, the node type of the work component is determined; according to the node type, the running relationship to be configured between work components is determined; a second configuration operation instruction is obtained, where the second configuration operation instruction is used to configure the running relationship to be configured between work components, thereby achieving the purpose of flexibly configuring the running relationship between components, and thus realizing the effect of improving the configuration flexibility of the running relationship to be configured between work components.

[0081] As an optional solution, determining the running relationship to be configured between work components according to the node type includes:

[0082] S1, in the case where the node type is a single-line connection type, determining that the next work component having a single-line connection relationship with the first work component includes the second work component;

[0083] S2. When the node type is the branch connection type, determine that the next working components having a branch connection relationship with the first working component include a second working component and a third working component;

[0084] S3. When the node type is the convergence connection type, determine that the previous working components having a convergence connection relationship with the first working component include a fourth working component and a fifth working component.

[0085] It should be noted that when the node type is the single-line connection type, determine that the next working component having a single-line connection relationship with the first working component includes the second working component; when the node type is the branch connection type, determine that the next working components having a branch connection relationship with the first working component include the second working component and the third working component; when the node type is the convergence connection type, determine that the previous working components having a convergence connection relationship with the first working component include the fourth working component and the fifth working component. Optionally, the branch connection can, but is not limited to, assign weights to the branches of the connection. The node type can, but is not limited to, be used to represent the position information of the node in the overall process, such as the start node, the end node, etc.

[0086] For further illustration by way of example, optionally, for example Figure 10 As shown, the workflow template 304 includes a set of components connected in a certain relationship. Specifically, the ordinary camera 308, the human attribute request 502, and the data storage 508 are connected in a convergence connection 1002 manner, that is, the data storage 508 is the next node of the ordinary camera 308 and the human attribute request 502; the human attribute request 502 is connected to the human attribute 504 and the data storage 508 in a parallel connection manner 1004, that is, the human attribute request 502 is the previous node of the human attribute 504 and the data storage 508; the data storage 508 is connected to the human attribute 504 in a single-line connection manner 1006, that is, the data storage 508 is the next node of the human attribute 504.

[0087] Through the embodiments provided by the present application, when the node type is the single-line connection type, determine that the next working component having a single-line connection relationship with the first working component includes the second working component; when the node type is the branch connection type, determine that the next working components having a branch connection relationship with the first working component include the second working component and the third working component; when the node type is the convergence connection type, determine that the previous working components having a convergence connection relationship with the first working component include the fourth working component and the fifth working component, thereby achieving the purpose of quickly determining the to-be-configured running relationship between the working components, and thus realizing the effect of improving the determination efficiency of the to-be-configured running relationship between the working components.

[0088] As an alternative solution, after performing a configuration operation on the working components to obtain an AI solution template for application in the target scenario, it includes:

[0089] S1. Obtain a target test instruction, and input the target test data into the sixth working component, where the component type of the sixth working component is a data access component type, and the target test data is used to test the working performance of the AI solution template in the target scenario;

[0090] S2. Obtain the target test result and display the result of the target test on the workflow configuration interface.

[0091] It should be noted that obtain a target test instruction, input the target test data into the sixth working component, where the component type of the sixth working component is a data access component type, and the target test data is used to test the working performance of the AI solution template in the target scenario; obtain the target test result and display the result of the target test on the workflow configuration interface.

[0092] For further illustration by example, optionally, connect the generated AI solution template to the dataset that has been collected in the data center for debugging, and listen to and display the logs through a long connection. At the same time, anchor and mark in red the module (working component) with problems to test whether the operation process of the generated AI solution template meets the requirement conditions.

[0093] Through the embodiments provided in this application, obtain a target test instruction, input the target test data into the sixth working component, where the component type of the sixth working component is a data access component type, and the target test data is used to test the working performance of the AI solution template in the target scenario; obtain the target test result and display the result of the target test on the workflow configuration interface, thereby achieving the purpose of testing whether the AI solution template meets the requirement conditions, and thus realizing the effect of improving the operation quality of the AI solution template.

[0094] As an alternative solution, after displaying the result of the target test on the workflow configuration interface, it includes:

[0095] S1. In the case where the target test result indicates that the working performance of the AI solution template meets the preset conditions, obtain the task information of the target AI task to be solved currently, where the scenario where the target AI task is located is the target scenario;

[0096] S2. Generate a target AI solution that matches the target AI task according to the AI solution template and the task information of the target AI task.

[0097] It should be noted that when the target test result indicates that the working performance of the AI solution template meets the preset conditions, the task information of the target AI task to be solved currently is obtained, where the scenario where the target AI task is located is the target scenario; according to the AI solution template and the task information of the target AI task, a target AI solution matching the target AI task is generated.

[0098] For further illustration by way of example, optionally, after testing or debugging, the AI solution template is officially released, and an AI task is established. When creating the task, the AI solution template is bound to specific hardware sensor devices and data sets to implement a specific instance based on the AI solution template to solve the target AI task in the target scenario.

[0099] Through the embodiments provided in this application, when the target test result indicates that the working performance of the AI solution template meets the preset conditions, the task information of the target AI task to be solved currently is obtained, where the scenario where the target AI task is located is the target scenario; according to the AI solution template and the task information of the target AI task, a target AI solution matching the target AI task is generated, thereby achieving the purpose of using the AI solution template to solve the target AI task in the target scenario, and thus realizing the effect of improving the solving efficiency of the target AI task in the target scenario.

[0100] As an optional solution, the method for generating the template of the artificial intelligence AI solution is applied in a specific embodiment, including:

[0101] Package middleware (working components) such as AI components, device components, data access components, data storage components, and script components in the AI solution link;

[0102] Drag components on the canvas of the AI template, and fill in the check items of the component configuration and the check items of the global workflow configuration of the template, branch path scheduling weights, etc. in sequence to implement a complete self - contained workflow;

[0103] Connect the edited template to the data set that has been collected in the data center for debugging. Through the long - link method, monitor the logs and display them, and at the same time anchor and mark the problem - causing module in red so that users can run through the template process they edited;

[0104] After debugging, officially release the template, and establish an AI task. When creating the task, bind the template to specific hardware sensor devices and data sets to implement a specific instance based on the template, and a scenario - based AI solution can be officially delivered to B - end users.

[0105] Optionally, WorkFlow Pipeline: By connecting processes in series through wires, custom templates can be implemented, and debugging of simulated tasks can be carried out. At the same time, when creating an AI task, an AI template can be bound, and form items can be flexibly determined according to the template configuration to achieve the delivery of an AI solution that meets the business scenario;

[0106] AI Component: Encapsulate AI capabilities and links in the AI process in the form of components. It is a middleware with flexible input and output parameters, which divides the original complex AI solution into functional modules such as AI components (i.e., algorithm components), device components, script components, data access components, and data storage components, and provides solutions for the workflow by dragging and connecting in series;

[0107] Console Monitor: Listen to the real-time data pushed by the service on the long connection channel through a long connection (socket), monitor the execution of the custom workflow, and debug the configuration items and input and output parameters between modules in the draggable and reconfigurable solution to achieve the purpose of running through the solution. In this solution, users can not only see the real-time logs but also see the specific modules marked in red;

[0108] AI Layout: By abstracting some abstract capabilities of the AI solution, a paradigm is formed through drag-and-drop arrangement. Data can be collected from the algorithm samples stored in the data center for debugging. After the template is determined to pass the debugging, it can be officially released. After release, it can be integrated into a specific AI task to form a directly usable solution for users;

[0109] AI Task: Based on the successfully released AI template, select real data sources (from real hardware devices such as cameras and all-in-one machines or data sets stored in the data center) to implement a solution that meets the user scenario and can be delivered to B-side users;

[0110] WorkFlow Conig: Determine the type of each module and the calibration of the next module, and carry the key information of each category of module. Each module is set to one of the three types: normal, switch, and combin. Weights for task execution can be assigned to the branch paths, thus supporting the connection of modules in series to edit the template;

[0111] Workflow front-end configuration (Web Config): It involves information related to data visualization and information related to form feedback display to achieve repeated editing of custom templates, connect AI templates and AI tasks, organize the concatenation relationship of module sets and parameter injection, and form a flexible and deliverable visual AI studio solution.

[0112] Further, taking the generation of an AI template for completing an AI task in a target scenario as an example, as Figure 11 shown, the specific steps are as follows:

[0113] Step S1102, establish a template workflow and connect AI middleware;

[0114] Step S1104, obtain the configuration information of the module and the configuration information of the template. Among them, the configuration information of the module and the configuration information of the template can be, but are not limited to, realized by checking the configuration items of each module and the template configuration items during the dragging process;

[0115] Step S1106, test the generated AI template. Among them, the test can be, but is not limited to, debugging logs and dynamic pipelines (socket implementation) for the execution of results;

[0116] Step S1108, establish a deliverable AI task based on the template (configuration information) that has been officially released after debugging.

[0117] In this embodiment, by designing solutions according to business scenarios, dragging components to generate AI templates, debugging the AI templates to observe the logs and then running the templates successfully, and then publishing the AI templates that can run successfully, and then creating AI tasks based on the published AI templates to make the AI tasks run completely. Through the visual process and service orchestration capabilities, the AI project implementation costs of various industry application developers are significantly reduced. For example, in the scenario of designing AI solutions, with the visual process orchestration capabilities, algorithms such as Natural Language Processing (NLP) and Optical Character Recognition (OCR) can be optionally and organically integrated into the OCR-NLP process. Each algorithm and operation is abstracted into a node in the process, and can be freely modified, added or deleted and finally orchestrated and executed. On this basis, the architecture is robust and scalable, and the scheduling is flexible. The data center currently has four data access capabilities: local upload, external push, online pull, and COS synchronization. The external push mode applied to the underwriting environment is just an application practice of one of the capabilities. Through the capabilities of the data center, the business does not need to care about the docking work of the underlying storage medium and the management mechanism of business data, which are all uniformly processed by the data center, greatly saving the time for the business to store data and improving the ability of the business to access data, providing a basic guarantee for the underwriting application to be applicable to more scenarios. At the same time, the data center continuously optimizes the logic, shortens the processing time, and improves the processing ability, providing more stable and reliable services for the underwriting business. In the real-time video parsing module, this application has the ability to select a device in the left device tree and perform real-time video parsing demonstration according to the AI task associated with the device.

[0118] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0119] According to another aspect of the embodiments of the present invention, there is also provided an AI solution template generation device for implementing the above-mentioned AI solution template generation method.

[0120] As Figure 12 shown, the device includes:

[0121] An acquisition unit 1202, configured to acquire an addition operation instruction triggered in a workflow configuration interface, where the addition operation instruction is used to indicate adding a set of work components matching a target scenario to a workflow template, and the work components include AI components composed of functional modules encapsulating AI algorithms;

[0122] A first configuration unit 1204, configured to configure component attribute information to be configured in the work components to obtain target component attribute information;

[0123] A second configuration unit 1206, configured to configure an operation relationship to be configured between the work components to obtain a target operation relationship;

[0124] A first generation unit 1208, configured to generate an AI solution template applied to the target scenario according to the target component attribute information and the target operation relationship.

[0125] Optionally, in this embodiment, the template generation device for the artificial intelligence AI solution may be, but is not limited to, applied to scenarios for solving business problems. The triggering method of the addition operation instruction may be, but is not limited to, a triggering method of moving or transferring an identifier representing a component to be added from a first position area (such as a component library on one side of the workflow configuration interface) to a second position area (such as a workflow template on one side of the workflow configuration interface), such as dragging, path transfer, etc. The work components may be, but are not limited to, functional modules that encapsulate AI capabilities and / or links in the AI process in the form of components and can be selectively added to a sidebar in a visual interface (workflow configuration interface). The component attribute information to be configured may be, but is not limited to, a set of configuration and function setting information required for the current component, such as name, type, attribute, etc. The operation relationship to be configured may be, but is not limited to, used to represent the connection relationship and operation logic between components, such as the first component executes first, and the second component executes when the first component finishes execution or the execution process meets a preset degree, etc. The AI solution template may be, but is not limited to, a template that can be flexibly configured or matched for solving target AI tasks in a feature scenario. For example, an AI task is created based on the generated AI solution template, and a solution is run with data collected by a device or data stored in a data center, where the solution is used to solve the AI task.

[0126] It should be noted that an addition operation instruction triggered in the workflow configuration interface is obtained. The addition operation instruction is used to indicate adding a set of work components matching the target scenario in the workflow template. The work components include AI components composed of functional modules encapsulating AI algorithms. The component attribute information to be configured in the work components is configured to obtain target component attribute information. The running relationship to be configured between the work components is configured to obtain a target running relationship. An AI solution template applicable to the target scenario is generated based on the target component attribute information and the target running relationship.

[0127] Specific embodiments may refer to the examples shown in the above method for generating an AI solution template. These examples are not elaborated herein.

[0128] Through the embodiments provided in this application, an addition operation instruction triggered in the workflow configuration interface is obtained. The addition operation instruction is used to indicate adding a set of work components matching the target scenario in the workflow template. The work components include AI components composed of functional modules encapsulating AI algorithms. The component attribute information to be configured in the work components is configured to obtain target component attribute information. The running relationship to be configured between the work components is configured to obtain a target running relationship. An AI solution template applicable to the target scenario is generated based on the target component attribute information and the target running relationship. By means of a set of components that can be flexibly added and combined, an AI solution template applicable to the target scenario is efficiently generated, thereby achieving the purpose of reducing the redundant steps in generating the AI solution template, and thus realizing the technical effect of improving the generation efficiency of the AI solution template.

[0129] As an alternative solution, the first configuration unit 1204 includes:

[0130] A first determination module, configured to determine the component type of the work component;

[0131] A second determination module, configured to determine the component attribute information to be configured according to the component type;

[0132] A first acquisition module, configured to acquire a first configuration operation instruction, where the first configuration operation instruction is used to configure the component attribute information to be configured.

[0133] Specific embodiments may refer to the examples shown in the above method for generating an AI solution template. These examples are not elaborated herein.

[0134] As an alternative solution, the second determination module includes:

[0135] The first determination subunit is configured to determine that the component attribute information to be configured includes at least one of the following: function configuration information and material library information when the component type of the working component includes the AI component type.

[0136] The second determination subunit is configured to determine that the component attribute information to be configured includes data storage information when the component type is the data storage component type.

[0137] The third determination subunit is configured to determine that the component attribute information to be configured includes device configuration information when the component type includes the device component type.

[0138] The fourth determination subunit is configured to determine that the component attribute information to be configured includes script content information when the component type is the script component type.

[0139] The fifth determination subunit is configured to determine that the component attribute information to be configured includes access configuration information when the component type is the data access component type.

[0140] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for generating a template of the artificial intelligence AI solution, and details are not described herein again.

[0141] As an alternative solution, the second configuration unit 1206 includes:

[0142] The third determination module is configured to determine the node type of the working component.

[0143] The fourth determination module is configured to determine the running relationship to be configured between the working components according to the node type.

[0144] The second acquisition module is configured to acquire a second configuration operation instruction, where the second configuration operation instruction is used to configure the running relationship to be configured between the working components.

[0145] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for generating a template of the artificial intelligence AI solution, and details are not described herein again.

[0146] As an alternative solution, the fourth determination module includes:

[0147] The sixth determination subunit is configured to determine that the next working component having a single-line connection relationship with the first working component includes the second working component when the node type is the single-line connection type.

[0148] The seventh determination subunit is configured to determine that the next working components having a branch connection relationship with the first working component include the second working component and the third working component when the node type is the branch connection type.

[0149] The eighth determination subunit is configured to determine that the previous working component having a converging connection relationship with the first working component includes a fourth working component and a fifth working component when the node type is a converging connection type.

[0150] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for generating a template of an AI solution. Such examples are not elaborated herein.

[0151] As an alternative solution, it includes:

[0152] The second acquisition unit is configured to, after performing a configuration operation on a working component to obtain an AI solution template applied in a target scenario, acquire a target test instruction, and input target test data into a sixth working component, where the component type of the sixth working component is a data access component type, and the target test data is used to test the working performance of the AI solution template in the target scenario;

[0153] The third acquisition unit is configured to, after performing a configuration operation on a working component to obtain an AI solution template applied in a target scenario, acquire a target test result, and display the result of the target test on a workflow configuration interface.

[0154] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for generating a template of an AI solution. Such examples are not elaborated herein.

[0155] As an alternative solution, it includes:

[0156] The fourth acquisition unit is configured to, after displaying the result of the target test on a workflow configuration interface, acquire task information of a target AI task to be solved currently when the target test result indicates that the working performance of the AI solution template meets a preset condition, where the scenario where the target AI task is located is the target scenario;

[0157] The second generation unit is configured to, after displaying the result of the target test on a workflow configuration interface, generate a target AI solution that matches the target AI task according to the AI solution template and the task information of the target AI task.

[0158] For specific embodiments, reference may be made to the examples shown in the above-mentioned method for generating a template of an AI solution. Such examples are not elaborated herein.

[0159] According to another aspect of the embodiments of the present invention, there is also provided an electronic device for implementing the above-mentioned method for generating a template of an AI solution, such as Figure 13As shown, the electronic device includes a memory 1302 and a processor 1304. A computer program is stored in the memory 1302, and the processor 1304 is configured to execute the steps in any of the above method embodiments through the computer program.

[0160] Optionally, in this embodiment, the above electronic device may be at least one network device among multiple network devices in a computer network.

[0161] Optionally, in this embodiment, the above processor may be configured to execute the following steps through a computer program:

[0162] S1. Obtain an add operation instruction triggered in a workflow configuration interface, where the add operation instruction is used to indicate adding a set of work components matching a target scenario in a workflow template, and the work components include AI components composed of functional modules encapsulating AI algorithms;

[0163] S2. Configure the component attribute information to be configured in the work components to obtain target component attribute information;

[0164] S3. Configure the running relationship to be configured between the work components to obtain a target running relationship;

[0165] S4. Generate an AI solution template applied to the target scenario according to the target component attribute information and the target running relationship.

[0166] Optionally, those of ordinary skill in the art can understand that Figure 13 the structure shown is only schematic, and the electronic device may also be a terminal device such as a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a handheld computer, and a Mobile Internet Device (MID), a PAD, etc. Figure 13 It does not limit the structure of the above electronic device. For example, the electronic device may further include more or fewer components (such as a network interface, etc.) than those shown Figure 13 in the figure, or have a different configuration from that shown Figure 13 in the figure.

[0167] Among them, the memory 1302 can be used to store software programs and modules, such as the program instructions / modules corresponding to the template generation method and device of the artificial intelligence (AI) solution in the embodiments of the present invention. The processor 1304 executes various functional applications and data processing by running the software programs and modules stored in the memory 1302, that is, implements the above-mentioned template generation method of the AI solution. The memory 1302 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some instances, the memory 1302 may further include a memory remotely disposed relative to the processor 1304, and these remote memories can be connected to the terminal through a network. Examples of the above network include but are not limited to the Internet, enterprise intranet, local area network, mobile communication network, and combinations thereof. Among them, the memory 1302 can specifically but not limitedly be used to store operation instructions, working components, target component attribute information, target running relationships, and AI solution templates, etc. As an example, as Figure 13 shown, the above-mentioned memory 1302 may include, but is not limited to, the acquisition unit 1202, the first configuration unit 1204, the second configuration unit 1206, and the first generation unit 1208 in the above-mentioned template generation device of the AI solution. In addition, it may also include, but is not limited to, other module units in the above-mentioned template generation device of the AI solution, which will not be elaborated in this example.

[0168] Optionally, the above-mentioned transmission device 1306 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wired network and a wireless network. In one instance, the transmission device 1306 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers through a network cable, so as to communicate with the Internet or a local area network. In one instance, the transmission device 1306 is a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0169] In addition, the above-mentioned electronic device further includes: a display 1308, which is used to display the above-mentioned operation instructions, working components, target component attribute information, target running relationships, and AI solution templates, etc.; and a connection bus 1310, which is used to connect each module component in the above-mentioned electronic device.

[0170] According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, in which a computer program is stored, and wherein the computer program is configured to execute the steps in any one of the above method embodiments when running.

[0171] Optionally, in this embodiment, the above computer-readable storage medium may be configured to store a computer program for performing the following steps:

[0172] S1. Obtain an addition operation instruction triggered in the workflow configuration interface, where the addition operation instruction is used to indicate adding a set of work components matching the target scenario to the workflow template, and the work components include AI components composed of functional modules encapsulating AI algorithms;

[0173] S2. Configure the component attribute information to be configured in the work components to obtain target component attribute information;

[0174] S3. Configure the running relationships to be configured between the work components to obtain target running relationships;

[0175] S4. Generate an AI solution template applied to the target scenario according to the target component attribute information and the target running relationships.

[0176] Optionally, in this embodiment, those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by a program instructing the relevant hardware of the terminal device. The program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc.

[0177] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0178] If the integrated unit in the above embodiments is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in the above computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in the storage medium and includes several instructions for causing one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention.

[0179] In the above embodiments of the present invention, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0180] In several embodiments provided by this application, it should be understood that the disclosed client can be implemented in other ways. Among them, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling, or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of units or modules can be in an electrical or other form.

[0181] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0182] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0183] The above is only the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for generating a template of an artificial intelligence (AI) solution, characterized in that, Including: Obtain an addition operation instruction triggered in the workflow configuration interface, where the addition operation instruction is used to indicate adding a set of work components matching the target scenario to the workflow template, and the work component is a functional module that encapsulates the AI capabilities and / or links in the AI process in component form and can be selectively added to a sidebar in the visual interface; Configure the component attribute information to be configured in the work component to obtain target component attribute information, where the component attribute information to be configured is a set of configuration required for the current component and the AI function setting information to be achieved; Configure the running relationship to be configured between the work components to obtain a target running relationship, where the target running relationship is the running relationship between multiple work components and the running logic relationship of a single component itself; Generate an AI solution template applied to the target scenario according to the target component attribute information and the target running relationship.

2. The method according to claim 1, wherein The configuring the component attribute information to be configured in the work component to obtain target component attribute information includes: Determine the component type of the work component; According to the component type, determine the component attribute information to be configured; Obtain a first configuration operation instruction, where the first configuration operation instruction is used to configure the component attribute information to be configured.

3. The method according to claim 2, wherein The determining the component attribute information to be configured according to the component type includes: In the case where the component type of the work component includes the AI component type, determine that the component attribute information to be configured includes at least one of the following: function configuration information, material library information; In the case where the component type is the data storage component type, determine that the component attribute information to be configured includes data storage information; In the case where the component type includes the device component type, determine that the component attribute information to be configured includes device configuration information; In the case where the component type is the script component type, determine that the component attribute information to be configured includes script content information; In the case where the component type is the data access component type, determine that the component attribute information to be configured includes access configuration information.

4. The method according to claim 1, wherein The configuring the running relationship to be configured between the work components to obtain a target running relationship includes: Determine the node type of the work component; According to the node type, determine the running relationship to be configured between the work components; Obtain a second configuration operation instruction, where the second configuration operation instruction is used to configure the running relationship to be configured between the work components.

5. The method according to claim 4, wherein The determining the running relationship to be configured between the work components according to the node type includes: In the case where the node type is the single-line connection type, determine that the next work component having a single-line connection relationship with the first work component includes the second work component; In the case where the node type is the branch connection type, determine that the next work components having a branch connection relationship with the first work component include the second work component and the third work component; When the node type is the aggregation connection type, determining the previous working components having an aggregation connection relationship with the first working component includes a fourth working component and a fifth working component.

6. The method according to claim 1, wherein After performing a configuration operation on the working component to obtain an AI solution template applied in the target scenario, it includes: Obtaining a target test instruction, and inputting target test data into a sixth working component, where the component type of the sixth working component is a data access component type, and the target test data is used to test the working performance of the AI solution template in the target scenario; Obtaining a target test result, and displaying the result of the target test on the workflow configuration interface.

7. The method according to claim 6, wherein After displaying the result of the target test on the workflow configuration interface, it includes: When the target test result indicates that the working performance of the AI solution template meets a preset condition, obtaining task information of a target AI task to be solved currently, where the scenario where the target AI task is located is the target scenario; Generating a target AI solution matching the target AI task according to the AI solution template and the task information of the target AI task.

8. A template generation device for an artificial intelligence (AI) solution, characterized in that, It includes: An obtaining unit, configured to obtain an add operation instruction triggered in the workflow configuration interface, where the add operation instruction is used to indicate adding a set of working components matching the target scenario to the workflow template, and the working component is a functional module that encapsulates AI capabilities and / or links in the AI process in the form of a component and can be selectively added in a sidebar of the visual interface; A first configuration unit, configured to configure component attribute information to be configured in the working component to obtain target component attribute information, where the component attribute information to be configured is a set of information used to represent the configuration required by the current component and the AI function settings to be achieved; A second configuration unit, configured to configure the running relationship to be configured between the working components to obtain a target running relationship, where the target running relationship is the running relationship between multiple working components and the running logic relationship of a single component itself; A first generating unit, configured to generate an AI solution template applied in the target scenario according to the target component attribute information and the target running relationship.

9. A computer-readable storage medium, the computer-readable storage medium including a stored program, wherein, When the program runs, it executes the method described in any one of claims 1 to 7 above.

10. An electronic device, comprising a memory and a processor, characterized in that, A computer program is stored in the memory, and the processor is configured to execute the method described in any one of claims 1 to 7 through the computer program.

Images