Application creation method and device, equipment and storage medium
By selecting the model to call the function blocks and configuring the interface during the application creation process, creating applications based on the processing nodes, solving the complexity and stability problems of developers during the development process, and achieving simplification and stability improvement of application development.
Patent Information
- Application Number
- CN202411118036.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-14
- Publication Date
- 2024-12-06
AI Technical Summary
In the existing technology, developers need to deeply understand the operating principles of the model, prompt word engineering and functional block construction technologies, which makes application development difficult and requires repeated configuration and debugging to form stable and usable applications, hoping to reduce the difficulty of development. To achieve simpler and faster application development.
By selecting the model calling function block from the multiple function blocks for the multiple processing nodes applied to the target to be created, presenting the configuration interface in response to the configuration request, receiving configuration information, and consolidating based on this information and processing nodes The workflow creates target applications, realizing the configurability and visualization of model calling methods.
It reduces the difficulty of application development with model calling capabilities, improves the stability of model output, simplifies the development process, and enables developers to create high-quality applications more quickly.
Smart Images

Figure CN119088478A_ABST
Abstract
Description
Technical Field
[0001] Example embodiments of the present disclosure generally relate to the field of computers, and more particularly, to application creation methods, devices, apparatuses, and computer-readable storage media. Background Art
[0002] With the development of information technology, various terminal devices can provide various services to people in work and life. Applications that provide services can be deployed in terminal devices. Terminal devices present corresponding content and interact with users through the user interface of the application to meet various needs of users. In some cases, developers may want to develop applications simply and quickly. Therefore, how to reduce the difficulty of application development has become a concern. Summary of the invention
[0003] In a first aspect of the present disclosure, an application creation method is provided. The method includes: for a model calling node among multiple processing nodes of a target application to be created, selecting a model calling function block for the model calling node to call a target model from multiple function blocks; in response to a configuration request for the model calling function block, presenting a configuration interface of the model calling function block; receiving configuration information for the model calling function block through the configuration interface; and creating a target application based on the configuration information and a workflow composed of multiple processing nodes.
[0004] In a second aspect of the present disclosure, an application creation device is provided. The device includes: a selection module configured to select a model calling function block for a model calling node to call a target model from multiple function blocks for a model calling node among multiple processing nodes of a target application to be created; a presentation module configured to present a configuration interface of the model calling function block in response to a configuration request for the model calling function block; a receiving module configured to receive configuration information for the model calling function block through the configuration interface; and a creation module configured to create a target application based on the configuration information and a workflow composed of multiple processing nodes.
[0005] In a third aspect of the present disclosure, an electronic device is provided. The device includes at least one processing unit; and at least one memory, the at least one memory is coupled to the at least one processing unit and stores instructions for execution by the at least one processing unit. When the instructions are executed by the at least one processing unit, the electronic device executes the method of the first aspect.
[0006] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, wherein a computer program is stored on the medium, and when the computer program is executed by a processor, the method of the first aspect is implemented.
[0007] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the method according to the first aspect of the present disclosure is implemented.
[0008] It should be understood that the content described in this section is not intended to limit the key features or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:
[0010] Figure 1 A schematic diagram showing an example environment in which embodiments of the present disclosure can be implemented;
[0011] Figure 2 A flowchart showing a process of application creation according to some embodiments of the present disclosure;
[0012] FIG. 3A to FIG. 3D Schematic diagrams respectively showing examples of application creation interfaces according to some embodiments of the present disclosure;
[0013] Figure 4 A flowchart showing a process of generating a data query instruction according to some embodiments of the present disclosure is shown;
[0014] Figure 5 A flowchart showing a process of data processing according to some embodiments of the present disclosure;
[0015] Figure 6 A schematic structural block diagram of an application creation device according to some embodiments of the present disclosure is shown; and
[0016] Figure 7 A block diagram of an electronic device in which one or more embodiments of the present disclosure may be implemented is shown. DETAILED DESCRIPTION
[0017] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.
[0018] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The term "some embodiments" should be understood as "at least some embodiments". Other explicit and implicit definitions may also be included below.
[0019] Herein, unless explicitly stated, executing a step “in response to A” does not mean executing the step immediately after “A” but may include one or more intermediate steps.
[0020] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition, use, storage or deletion of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.
[0021] It is understandable that before using the technical solutions disclosed in the various embodiments of the present disclosure, the types, scopes of use, usage scenarios, etc. of the information involved in the present disclosure should be informed to relevant users and their authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations. The relevant users may include any type of right holders, such as individuals, enterprises, and groups.
[0022] For example, in response to receiving an active request from a user, a prompt message is sent to the relevant user to clearly prompt the relevant user that the operation requested to be performed will require obtaining and using the information of the relevant user, so that the relevant user can independently choose whether to provide information to software or hardware such as an electronic device, application, server or storage medium that executes the operation of the technical solution of the present disclosure based on the prompt message.
[0023] As an optional but non-limiting implementation, in response to receiving an active request from a relevant user, a prompt message is sent to the relevant user, for example, in the form of a pop-up window, in which the prompt message may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to select "agree" or "disagree" to provide information to the electronic device.
[0024] It is understandable that the above notification and the process of obtaining user authorization are only illustrative and do not limit the implementation of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation of the present disclosure. The activation of the digital assistant related functions, the acquired data, the processing and storage of the data, etc. in the embodiment of the present disclosure shall obtain the prior authorization of the user and other rights holders associated with the user, and shall comply with the relevant laws and regulations and the agreement rules between the rights holders.
[0025] As used herein, the term "model" can learn the association between the corresponding input and output from the training data, so that after the training is completed, the corresponding output can be generated for a given input. The generation of the model can be based on machine learning technology. Deep learning is a machine learning algorithm that processes inputs and provides corresponding outputs by using multi-layer processing units. A neural network model is an example of a model based on deep learning. In this article, "model" may also be referred to as "machine learning model", "learning model", "machine learning network" or "learning network", and these terms are used interchangeably in this article.
[0026] Figure 1 A schematic diagram of an example environment 100 in which embodiments of the present disclosure can be implemented is shown. The environment 100 involves an application management platform 110, which can support the creation of applications and / or the running of applications. In some embodiments, the portion of the application management platform 110 used to support the creation of applications can also be referred to as an application creation portion 111. In some embodiments, the portion of the application management platform 110 used to support the running of applications can also be referred to as an application running portion 112.
[0027] like Figure 1 As shown, the application creation part 111 can provide the user 105 with an environment for creating and publishing applications. User 105 can be referred to as an application creation user or creator. In some embodiments, the application creation part 111 can be a low-code platform that provides a collection of tools for application creation. The application creation part 111 can support visual development of various types of applications, so that developers can skip the manual coding process and speed up the application development cycle and cost. The application creation part 111 can support any appropriate platform for users to develop one or more types of applications, and may include, for example, a platform based on Application Platform as a Service (aPaaS). Such a platform can support users to efficiently develop applications and implement operations such as application creation and application function adjustment.
[0028] The application creation part 111 may be deployed locally on the terminal device of the user 105, and / or may be supported by a server device. For example, the terminal device of the user 105 may run a client of the application creation part 111, and the client may support the interaction between the user and the application creation part 111 provided by the server. In the case where the application creation part 111 runs locally on the user's terminal device, the user 105 may directly interact with the local application creation part 111 using the terminal device. In the case where the application creation part 111 runs on the server device, the server device may provide services to the client running in the terminal device based on the communication connection with the terminal device. The application creation part 111 may present a corresponding interface 130 to the user 105 based on the operation of the user 105, so as to output information related to application creation to the user 105 and / or receive information from the user 105.
[0029] In some embodiments, the application creation part 111 can be associated with a corresponding database, which stores data or information required for the application creation process supported by the application creation part 111. For example, the database can store the code and description information corresponding to each functional module used to constitute the application. The application creation part 111 can also perform operations such as calling, adding, deleting, and updating the functional modules in the database. The database can also store operations that can be performed on different functional blocks. Exemplarily, in the scenario where an application is to be created, the application creation part 111 can call the corresponding functional blocks from the database to build the application.
[0030] In an embodiment of the present disclosure, the user 105 can create a target application 120 (also referred to as a target application to be created) as needed on the application creation part 111, and publish the target application 120. The target application 120 can be published to any appropriate application running part 112, as long as the application running part 112 can support the running of the target application 120.
[0031] After publishing, the target application 120' is available. The target application 120' can be used to be operated by one or more end users 145. The end user 145 can operate the target application 120' through an associated terminal device 146, and then interact with the application management platform 110. The end user 145 can be referred to as the end user of the target application 120'. In some embodiments, the target application 120' may include or be implemented as a digital assistant.
[0032] The digital assistant can be configured to have the ability of intelligent conversation. The digital assistant can be integrated into the target application 120` as part of the target application 120` to assist in executing task processing within the target application 120`. In other examples, the digital assistant can be configured as an independently running application, such as a web application or other types of applications. In such an example, the digital assistant and the target application 120` can be regarded as the same application. The digital assistant is provided to assist users with various task processing needs in different applications and scenarios. During the interaction with the digital assistant, the user inputs an interactive message, and the digital assistant provides a reply message in response to the user input. Generally, the digital assistant can support users to input questions in natural language, and perform tasks and provide replies based on the understanding of natural language input and logical reasoning ability.
[0033] In some embodiments, the digital assistant can interact with the terminal user 145 as a contact. For example, the digital assistant can be implemented in an instant messaging (IM) application. The digital assistant can interact with the terminal user 145 in a single chat session with the terminal user 145. In some embodiments, the digital assistant can interact with multiple users in a group chat session including multiple users.
[0034] For each terminal user 145, the client of the application running part 112 can present an interactive window 142 of the target application 120' or the digital assistant in the client interface, such as a conversation window with the digital assistant. The terminal user 145 can enter a conversation message in the conversation window, and the target application 120' can determine the reply message of the digital assistant based on the created configuration information and present it to the user in the interactive window 142. In some embodiments, depending on the configuration of the target application 120', the interactive message with the target application 120' may include a message in a multimodal form, such as a text message (e.g., natural language text), a voice message, an image message, a video message, and the like.
[0035] Similar to the application creation part 111, the application running part 112 can be deployed locally on the terminal device of each terminal user 145, and / or can be supported by a server device. For example, the terminal device of the terminal user 145 can run a client of the application running part 112, and the client can support the interaction between the user and the application running part 112 provided by the server. In the case where the application running part 112 runs locally on the user's terminal device, the terminal user 145 can directly use the terminal device to interact with the local application running part 112. In the case where the application running part 112 runs on the server device, the server device can realize the service provision of the client running in the terminal device based on the communication connection with the terminal device. The application running part 112 can present the corresponding application page to the terminal user 145 based on the operation of the terminal user 145, so as to output and / or receive information related to the use of the application to the terminal user 145 and / or from the terminal user 145.
[0036] In some embodiments, the implementation of at least part of the functions of the target application 120' and / or the implementation of at least part of the functions of the digital assistant in the target application 120' can be implemented based on the model. During the creation or operation of the target application 120', one or more models 155, such as the capabilities of the model 155, can be called. In the target application 120', the digital assistant can use the model 155 to understand the user input and provide a response to the user based on the output of the model 155.
[0037] During the creation process, the application management platform 110 needs to use the model 155 to determine whether the operation result of the target application 120 meets expectations when testing the target application 120. During the operation process, in response to different operation requests of the user of the target application 120, the application operation part 112 may need to use the model 155 to determine the response result to the user.
[0038] Although shown as being independent of the application management platform 110, one or more models 155 may run on the application management platform 110, or other remote servers. In some embodiments, the model 155 may be a machine learning model, a deep learning model, a learning model, a neural network, etc. In some embodiments, the model may be based on a language model (LM). The language model can have question-answering capabilities by learning from a large amount of corpus. The model 155 may also be based on other appropriate models.
[0039] The application management platform 110 can run on appropriate electronic devices. The electronic devices here can be any type of device with computing capabilities, including terminal devices or server devices. The terminal device can be any type of mobile terminal, fixed terminal or portable terminal, including mobile phones, desktop computers, laptop computers, notebook computers, netbook computers, tablet computers, media computers, multimedia tablets, personal communication systems (PCS) devices, personal navigation devices, personal digital assistants (PDAs), audio / video players, digital cameras / camcorders, positioning devices, television receivers, radio broadcast receivers, e-book devices, gaming devices, or any combination of the foregoing, including accessories and peripherals of these devices or any combination thereof. The server device can include, for example, a computing system / server, such as a mainframe, an edge computing node, a computing device in a cloud environment, and the like. In some embodiments, the management platform 110 can be implemented based on cloud services.
[0040] It should be understood that the structure and functionality of environment 100 are described for exemplary purposes only and do not imply any limitation on the scope of the present disclosure. For example, although a single user interacting with application creation portion 111 and a single user interacting with application execution portion 112 are shown, in reality multiple users may access application management platform 110 to each create a digital assistant, and each digital assistant may be used to interact with multiple users.
[0041] As mentioned above, applications that provide services can be deployed in terminal devices. Traditionally, developers developing applications with high stability and high availability require high software development capabilities, and usually require technicians with professional software development skills to complete application development.
[0042] With the development of machine learning technology, its application in terminal device scenarios is becoming more and more extensive. Some applications provide services to users by calling one or more machine learning models during operation. This application method reduces the difficulty of application development to a certain extent, allowing more developers to participate in the application development process. For example, some companies can develop enterprise-level applications based on the usage needs in enterprise scenarios.
[0043] Although the use of machine learning models can reduce the difficulty of application development, developers still need to have a deep understanding of the principles of model operation, prompt word engineering, and function block construction technologies, and then they need to repeatedly configure, debug, and evaluate the application to form an application with high stability and high usability. Therefore, developers still hope to further reduce the difficulty of application development so that they can develop applications more simply and quickly.
[0044] In view of this, in an embodiment of the present disclosure, an improved solution for application creation is provided. In the solution, for a model calling node among multiple processing nodes of a target application to be created, a model calling function block for the model calling node to call a target model is selected from multiple function blocks; in response to a configuration request for the model calling function block, a configuration interface of the model calling function block is presented; configuration information for the model calling function block is received through the configuration interface; and based on the configuration information and a workflow composed of multiple processing nodes, a target application is created.
[0045] In this way, by using engineering methods, the calling method of the target model can be made configurable, and the configuration of the calling method of the target model can be visualized, which can reduce the difficulty of developing applications with model calling capabilities and help improve the stability of model output.
[0046] Some example embodiments of the present disclosure will be described in detail below with reference to examples of the accompanying drawings.
[0047] The task management process described in the embodiments of the present disclosure may be implemented on an application management platform, a terminal device on which the application management platform is installed, and / or a server corresponding to the application management platform. In the following examples, for the sake of discussion, the description is from the perspective of the application management platform, for example Figure 1 The application management platform 110 shown. The user interface presented by the application management platform 110 can be presented via the terminal device of the user 145, and the application management platform 110 can receive user input via the terminal device of the user 145. In some embodiments of the present disclosure, the user 145 is the terminal user of the target application 120. It should be understood that the user interface presented by the application management platform 110 can also be presented via the terminal device of the user 105, and the application management platform 110 can also receive user input via the terminal device of the user 105. In some embodiments of the present disclosure, the user 105 is the creator, manager or maintainer of the target application 120.
[0048] Figure 2 1 is a flowchart of a process 200 for creating an application according to some embodiments of the present disclosure. The process 200 may be implemented in the application management platform 110, for example, by the application creation part 111 of the application management platform 110. Figure 1 To describe Figure 2 Process 200 is shown.
[0049] In block 210, the application management platform 110 selects a model calling function block from a plurality of function blocks for the model calling node to call the target model, for the model calling node in the plurality of processing nodes of the target application 120 to be created. The target application 120 here may be any type of application, such as a data analysis application, a text generation application, a personnel management application, a financial application, a performance application, an entertainment application, an electronic office application, and the like.
[0050] The processing node here should be understood as an execution point, decision point or task point in the data processing flow, workflow or business process of the application, which can represent one or more operations, decisions or tasks corresponding to the processing node. The target application 120 to be created may include multiple processing nodes, such as input nodes, nodes for specific data processing, function call nodes, interface call nodes, model call nodes, record nodes, logical branch nodes, output nodes, etc. It can be understood that the above-mentioned applications and processing nodes of the applications are exemplary. In the actual creation process of the application, the processing nodes contained in different applications may not be the same.
[0051] The function block here should be understood as a set of programs or systems that can be added to each processing node to implement the operation, decision or task represented by the processing node. Examples of function blocks may include, but are not limited to, function blocks for performing specific data processing, function blocks for generating data query instructions, function blocks for generating data processing instructions, function blocks for analyzing data, function blocks for calling functions, function blocks for calling interfaces, function blocks for calling models, function blocks for reading and writing, function blocks for decision making or logical judgment, and the like.
[0052] In some embodiments, the application management platform 110 may present an application creation interface in response to receiving a creation request for the target application 120. In response to the configuration operation of the workflow of the target application, node identifiers corresponding to multiple processing nodes, and the connection relationship between the multiple node identifiers are presented in the application creation interface to indicate the workflow composed of multiple processing nodes through the node identifiers and the connection relationship. The application creation interface may include one or more interfaces, such as an application management interface, a process design interface, a page design interface, a data management interface, a permission setting interface, and the like. The node identifier can be used to identify the processing node and / or the functional block corresponding to the processing node.
[0053] In some embodiments, the application management platform 110 may present a node identifier corresponding to the model call function block in the application creation interface in response to the operation of selecting a model call function block for the model call node. For example, if the model call node does not select the corresponding model call function block, the node identifier may present a first display state. If the model call node has determined the corresponding model call function block, the node identifier presents a second display state.
[0054] To illustrate the creation of a processing node and the selection of a functional block of the target application 120, the following will be combined with FIG. 3A to FIG. 3D An example is given. FIG. 3A to FIG. 3D Schematic diagrams of examples 300A to 300D of application creation interfaces according to some embodiments of the present disclosure are shown respectively. It should be understood that the interfaces shown in the drawings are merely examples, and various interface designs may actually exist. The various graphical elements in the interface may have different arrangements and different visual representations, one or more of which may be omitted or replaced, and one or more other elements may also exist. The embodiments of the present disclosure are not limited in this respect.
[0055] The user 105 can configure basic information such as application name, application description, and application icon through the application management interface in the application management platform 110 to start the process of creating the target application 120. In response to starting the process of creating the target application 120, the application management platform 110 presents the interface shown in example 300A, which shows an example of an application creation interface (e.g., a workflow design interface).
[0056] The example 300A includes node identifications 310A, 320A, 330A, 340A, 350A, 360A, etc. of multiple processing nodes. The display status of the node identifications 310A, 320A, 330A, 340A indicates that the corresponding processing nodes have selected the corresponding functional blocks, and the display status of the node identification 350A indicates that the corresponding processing node has not selected the corresponding functional block.
[0057] For example, the initial state of the application creation interface may not display the node identification, or may display one or more node identifications in a display state as shown by the node identification 350A. The node identification 350A includes a function block adding entry 351. The application management platform 110 may present a function block selection interface in response to a trigger operation of the function block adding entry 351. The function block selection interface includes a plurality of candidate function blocks, and the candidate function blocks have a predetermined processing logic and preset information for a predetermined operation, and the preset information includes preset parameters, preset variables, preset conditions, preset prompt words, and the like.
[0058] In response to the selection operation for any one of the plurality of candidate function blocks, the application management platform 110 may determine the selected function block as the model call function block of the model call node corresponding to the node identifier 350. Afterwards, the node identifier 350 switches to a display state similar to the node identifiers 310A, 320A, 330A, and 340A, presenting an icon of the selected function block, description information of the function block, and the like.
[0059] Example 300A includes connecting lines 312, 322, etc. between node identifiers. Connecting lines 312, 322, etc. can be used to identify the connection relationship between node identifiers. The connection relationship between node identifiers can indicate the association relationship between corresponding processing nodes, such as a timing relationship, a logical relationship, a data connection relationship, etc. A flowchart can be formed by node identifiers and connecting lines, and the flowchart can refer to a workflow composed of multiple processing nodes. Alternatively or additionally, connecting lines 312, 322 can be preset, or they can be added by the application management platform 110 in the interface shown in Example 300A in response to the operation of the user 105.
[0060] For example, the interface shown in example 300A presents node identifiers 310A, 330A, and a connection line 312 between node identifiers 310A and 330A in the initial state, and a node adding entry 313 is provided on the connection line 312. In response to a trigger operation on the node adding entry 313, the application management platform 110 can add a node identifier 320A between the node identifier 310A and the node identifier 330A, connect the node identifier 310A and the node identifier 320A through the connection line 312, and connect the node identifier 320A and the node identifier 330A through the connection line 322.
[0061] Of course, the application creation interface may also include, for example, a node adding control, such as displaying the node adding control in the top menu bar or the side task bar of the application creation interface. The application management platform 110 may display the node identification corresponding to the added processing node in the application creation interface in response to the triggering operation of the node adding control.
[0062] Similarly, the application creation interface may also include a connection line adding control, and the user adds a connection line to the application creation interface by operating the connection line adding control, for example, by clicking or dragging the connection line adding control to add a connection line between two node identifiers.
[0063] It should be noted that the above-mentioned creation process of the processing node and the selection process of the function block are only exemplary. In actual application, the application creation page may also only display the node identification, or display the brief interface of the function block corresponding to the processing node, etc.
[0064] In block 220, the application management platform 110 presents a configuration interface of the model call function block in response to a configuration request for the model call function block. The user may input a configuration request to the application management platform 110 in a variety of ways. In some embodiments, the application management platform may present a configuration interface of the model call function block in response to receiving a trigger operation for a node identifier corresponding to a model call node.
[0065] For example, Figure 3A and Figure 3B As shown, when the user needs to configure the functional module corresponding to the node identifier 320A, the user can trigger the node identifier 320A through the application management platform 110 or an attached device of the application management platform 110. The application management platform 110 can present the configuration interface 320B shown in example 300B in response to the triggering operation on the node identifier 320A.
[0066] It is understandable that the above-mentioned method of triggering the application management platform 110 to present the configuration interface of the model calling function block is only exemplary. In specific implementation, the application management platform 110 may be triggered to present the configuration interface of the model calling function block in other ways. For example, the user may select the node identifier 320A through the application management platform 110 or the attached device of the application management platform 110, and may trigger the application management platform 110 to present the configuration interface by selecting the right-click menu or the configuration control in the task bar.
[0067] In box 230, the application management platform 110 receives configuration information for the model calling function block through the configuration interface. The configuration information may include various parameters, variables, modes, strategies or conditions required for the model calling function block to perform corresponding operations, tasks or decision-making processes. Correspondingly, the configuration interface may include an entry, input box or control for receiving this information, etc. It is understandable that due to the different operations required to be performed by different function blocks, the configuration information required by different function blocks may be different, and the elements, layouts and visual representations contained in the configuration interfaces of different function blocks may also be different. The configuration interface will be exemplified in conjunction with some examples below, but it should not be understood that the configuration interface is limited to the interface design shown below.
[0068] In some embodiments, the model calling function block is configured to call the target model based on the configuration information to generate a predetermined type of content. That is, after the target application 120 is created, the application running part 112 can call the target model based on the configuration information of the model calling function block by running the model calling function block to generate a predetermined type of content. The predetermined type of content here may include but is not limited to text, images, audio, video, program code, structured data with a specific structure, etc. Taking program code as an example, the predetermined type of program code may include, for example, data query instructions (such as database query instructions written in structured query language), data processing instructions (such as data processing instructions written in Python programming language), etc. Of course, the above content is only exemplary, and those skilled in the art can choose to call the content generated by the model according to actual needs.
[0069] In some embodiments, the application management platform 110 can receive the specified information for the target model as at least a part of the configuration information through the model configuration entry in the configuration interface. The specified information is used to specify the target model to the model call function block to generate a predetermined type of content. Alternatively or additionally, the specified information may include various information that can identify and call the target model, such as the name, number, interface, address, etc. of the target model.
[0070] There are many implementation methods for the model configuration entry. In some embodiments, the model configuration entry can be implemented as an input box including a call address of the target model, and the user can input the call address of the target model into the input box through the application management platform 110 or a device attached to the application management platform 110. In this way, after the target application 120 is released, the application running part 112 can call the target model based on the call address.
[0071] In other embodiments, the application management platform 110 may present at least one candidate model identifier in response to receiving a trigger operation on a model configuration entry. The application management platform 110 may receive a selection operation on a target identifier among the identifiers of at least one candidate model, and then, in response to the selection operation, determine that the candidate model corresponding to the target identifier is designated as the target model.
[0072] For example, Figure 3B and Figure 3CAs shown, the interface shown in Example 300B presents a model configuration entry 326, and the model configuration entry 326 may include a control 327. In response to the trigger operation of space 327, the application management platform 110 may switch the model configuration entry 326 to the display state in Example 300C, and display the identifiers 371, 372, and 373 of multiple candidate models in the model configuration entry 326, respectively, for the user to select a target model. In response to the user's selection operation on any one of the identifiers 371, 372, and 373, the application management platform 110 may designate the model corresponding to the selected identifier as the target model. For example, if the user selects identifier 373, the application management platform may designate the model corresponding to identifier 373 as the target model, and display identifier 373 in the model configuration entry 326, as shown in FIG. Figure 3B shown.
[0073] In some embodiments, the predetermined type of content may include a data query instruction. The application management platform 110 may receive at least one query condition for generating a data query instruction as at least a part of the configuration information through the query configuration entry of the configuration interface. The query condition here is used to filter out data that meets specific conditions. The query condition may include but is not limited to a data query range (e.g., a database, a table to be queried, a row in a table, a column in a table, etc.), a data value screening condition (e.g., greater than, equal to, less than, greater than or equal to, etc.), a logical screening condition (e.g., and, or, not, etc.), a fuzzy condition, etc.
[0074] For example, Figure 3C As shown, the interface shown in example 300C may include a query configuration entry 324, which can be used to receive a table to be queried. In response to a trigger operation on the query configuration entry 324, the application management platform 110 may present an interface including one or more candidate tables. The user may select a table to be queried from the candidate tables presented in the interface, and the application management platform 110 may, in response to the selection operation on the candidate table, use the selected candidate table as the table to be queried, and present an identifier 374 of the table to be queried in the query configuration entry 324. Of course, the user may also input an identifier 374 of the table to be queried (e.g., the name of the table) into the query configuration entry 324.
[0075] In some embodiments, the predetermined type of content may include structured data. The application management platform 110 can receive structural information indicating the data structure of the structured data through the output configuration entry of the configuration interface as at least a part of the configuration information. Structured data should be understood as data in a standardized format. Alternatively or additionally, structured data may include but is not limited to structures, objects, strings, arrays, tables, and the like. Taking a structure as an example, a structure may include multiple fields, each of which may include different types of data. It is understandable that for different types of structured data, the data structure and structural information may be different, resulting in different interface elements and layouts of the configuration interface.
[0076] For example, Figure 3A and Figure 3D As shown, the function block corresponding to the node identifier 330A can be used to call the target model to generate an article. The application management platform 110 can present the configuration interface 330B shown in example 300D in response to the trigger operation of the node identifier 330A in the interface shown in example 300A. The configuration interface 330B includes an output configuration entry 335, and the output configuration entry 335 may include multiple groups of input boxes 336, 337, and 338, wherein a group of input boxes 336 includes input boxes for inputting parameter names and input boxes for inputting the type of structured data. The user can enter "response" in the input box for inputting the parameter name, and can also enter, for example, "object" in the input box for inputting the type of structured data. A group of input boxes 337 and a group of input boxes 338 can respectively include input boxes for inputting field names, field types, and field descriptions. The user can enter "content", "string", and "article content" in the group of input boxes 337. The user can also enter "title", "string", and "article title" in the group of input boxes 338. In this way, the function block corresponding to the node identifier 330A can call the target model to generate an article, and can output the generated article as an object named "Response", which includes a "title" field and a "content" field. The data contained in the "title" field is the article title, and the data contained in the "content" field is the article content.
[0077] It is understandable that the object is not limited to including the "title" field and the "content" field. The user can also trigger the control 375 in the interface shown in example 300D, and the application management platform 110 can present one or more groups of input boxes in response to the triggering operation of the control 375. The user can enter field information such as the field name, field type, and field description of one or more other fields of the object through the one or more groups of input boxes.
[0078] In some embodiments, the application management platform 110 can receive preset prompt information through the prompt input entry in the configuration interface, and the preset prompt information at least includes the generation requirements of the predetermined type of content. The preset prompt information here can include the generation requirements of the predetermined type of content written in natural language.
[0079] For example, Figure 3C As shown, the function block corresponding to the node identifier 320A can be used to call the target model to generate a data query instruction. The configuration interface 320B includes a prompt input entry 325 for inputting a prompt word (prompt). The user can use natural language to input query requirements into the prompt input entry 325 to guide the model to generate data query instructions that meet the requirements.
[0080] For example, Figure 3D As shown, the function block corresponding to the node identifier 330A can be used to call the target model to generate an article, and the configuration interface 330B includes a prompt input entry 334 for inputting a prompt word. The user can use natural language to input clear requirements for the article into the prompt input entry 334 to guide the model to generate an article that meets the requirements.
[0081] In block 240 , the application management platform 110 creates a target application 120 based on the configuration information and a workflow consisting of a plurality of processing nodes.
[0082] In some embodiments, after receiving the configuration information for each functional block, the application management platform 110 may also test each functional block. If the test results indicate that each functional block has passed the test, the application management platform 110 may create a target application 120 based on the configuration information of each functional block and a workflow composed of multiple processing nodes.
[0083] For example, Figure 3C As shown, the application management platform 110 may present an operation result interface 320C in response to the test operation of the function block corresponding to the node identifier 320A. The operation result interface 320C may include an area 328 for presenting the operation result, and the operation result of the function block may be presented in the area 328, such as the query instruction at the data generated by the function block calling the target model. The application management platform 110 may determine whether the function block corresponding to the node identifier 320A passes the test based on the operation result.
[0084] For example, Figure 3DAs shown, the application management platform 110 may present an operation result interface 330C in response to the test operation of the function block corresponding to the node identifier 330A. The operation result interface 330 may also include an area 339 for presenting the operation result, and the operation result of the function block may be presented in the area 339, for example, the function block calls the article generated by the target model, and outputs an object including the "title" field and the "content" field based on the generated article. The application management platform 110 may determine whether the function block corresponding to the node identifier 330A passes the test based on the operation result.
[0085] Exemplarily, the application management platform 110 may determine that the test results of the functional blocks corresponding to the node identifiers 310A, 320A, 330A, 340A, 350A, and 360A all indicate that the test has passed, and then create the target application 120 based on the configuration information of each functional block and a flowchart (such as a timing relationship, a logical relationship, a data transmission relationship, etc.) that can indicate a workflow composed of multiple processing nodes. The application file of the target application 120 created by the application management platform 110 records the configuration information of each functional block and the workflow composed of multiple processing nodes.
[0086] It is understandable that in the actual application creation process, it is not limited to process design, but also includes configuration processes in multiple dimensions such as page design, data management, permission setting, account management, etc., so the actual application creation process is much more complicated than the process described in this article. However, since the configuration processes in other dimensions during the application creation process are not related to the improved solutions of the embodiments of the present disclosure, this article does not introduce in detail the configuration processes in other dimensions such as page design, data management, permission setting, account management, etc.
[0087] The following is an exemplary description of the running process of the target application 120' after being released. The running process of the target application can be implemented in the application management platform 110, for example, it can be implemented by the application running part 112 of the application management platform 110. Figure 1 To describe the running process of the target application.
[0088] In an embodiment of the present disclosure, the application management platform 110 may generate target prompt information based on the user input and configuration information in response to receiving the user input through the target application 120'. The target prompt information is provided to the target model to obtain a model output. Thereafter, the application management platform 110 may determine response information for the user input based on the model output.
[0089] Alternatively or additionally, the user input may be input from any suitable user, for example, it may be user input from user 145. The user input may be any suitable type of user input, such as text, voice, image, video, and the like.
[0090] Alternatively or additionally, upon receiving user input, the application management platform 110 may determine a model calling function block that responds to the user input from a plurality of function blocks of the target application 120'. If the model calling function block includes a model calling function block for calling a target model to generate a predetermined type of content, target prompt information may be generated based on the configuration information of the model calling function block and the user input. The target prompt information may be used to prompt the target model of the user input requirement, and the target prompt information may include, for example, a prompt word or a model input. After receiving the target prompt information, the target model may generate a model output based on the target prompt information and provide the model data to the application management platform 110.
[0091] Alternatively or additionally, the application management platform 110 may obtain historical interaction information that matches the user input. For example, the user may interact with the target application 120' based on multimodal information such as text, voice, image, video, etc., and the data generated during this interaction process may be stored as historical interaction information. In the process of answering the user's current input, the application management platform 110 may obtain the stored historical interaction information, such as the question input by the user or the response fed back by the target application 120', etc. The application management platform 110 may obtain the structural information and preset prompt information in the configuration information, the structural information indicating the data structure of the structured data, and the preset prompt information including the generation requirements of the structured data. The target prompt information for indicating the target model to generate the structured data is generated based on the user input, the historical interaction information, the structural information and the preset prompt information by using the model call function block. Afterwards, the application management platform 110 may provide the target prompt information to the target model, indicating that the target model generates the model output based on the target prompt information. The application management platform 110 may output a reply information for the user input based on the model. In this case, the reply information may include structured data having the data structure indicated by the structural information. Afterwards, the reply information may also be stored as historical interaction information.
[0092] Regarding the application management platform 110 determining the reply information for the user input based on the model output, in one example, the application management platform 110 may present at least part of the content in the model output to the user as the reply information. For example, if the user input instructs the target application to generate an article, the target model may generate an article for the user input based on the target prompt information, and the application management platform 110 may present the specific content of the article to the user.
[0093] In another example, the application management platform 110 may call the target model through a model call function block to obtain the model output. Afterwards, the application management platform 110 may also call another function block based on the model output to perform another operation to generate reply information for the user input. For example, the user input may instruct the target application to perform a data query operation, and the application management platform 110 may call the target model through a model call function block to generate a data query instruction, and then perform a data query operation based on the data query instruction through another function block, and use the queried data as the reply information for the user input.
[0094] The following combination Figure 4 and Figure 5 The operation process of the functional blocks in the target application is exemplified. Figure 4 A flowchart of a process 400 for generating a data query instruction according to some embodiments of the present disclosure is shown. Figure 1 The environment in 100, and will Figure 3A The function block corresponding to the node identifier 320A is executed as a model call function block. Figure 4 The process 400 shown is taken as an example. Figure 4 Process 400 is shown for illustration.
[0095] In block 401, the application management platform 110 may receive a data query request. The data query request may be triggered by a user, for example, the user may input the data query request through the application management platform 110 or a device attached to the application management platform 110. The data query request may also be triggered by another functional block of the application management platform 110. For example, the user input may instruct the target application to generate an article, and the user input requirement is determined by the other functional block, and the data query request is generated in response to the user input requirement.
[0096] In block 402, the application management platform 110 may obtain metadata of a table to be queried. The table to be queried may be recorded in the configuration information of the model calling function block, or may be determined based on user input. For example, at least one table to be queried that matches the requirement may be determined based on the requirement input by the user. After determining the at least one table to be queried, metadata of the at least one table to be queried may be obtained. Metadata is used to describe the table structure and attributes of the table to be queried, such as table name, row information, column information, index, description information, etc.
[0097] In block 403 , the application management platform 110 may generate a prompt based on the metadata, the configuration information of the model calling function block, and the user input.
[0098] Alternatively or additionally, before generating the prompt word, the application management platform 110 may pre-process the metadata. The pre-processing may include replacing identification information such as the table name of the table to be queried in the metadata with a number (ID) to reduce the amount of metadata data. The pre-processing may also include standardizing the metadata to process the metadata into data in a specific format. For example, all or part of the data content in the metadata may be processed into data that complies with the Database Markup Language (DBML).
[0099] Alternatively or additionally, the application management platform 110 is not limited to generating prompt words based on the above information. For example, the application management platform 110 can also obtain environmental information (such as date, time, current events, cultural background, etc.), and preset prompt words in the preset information of the model calling function block. The application management platform 110 can use this information together with the preset prompt words in the configuration information, user input, and metadata to generate prompt words (Prompt).
[0100] In block 404, the application management platform 110 may determine whether the capacity of the generated prompt word exceeds a capacity threshold. The capacity here can indicate the amount of data of the prompt word, such as the number of words, bytes, tokens, etc. In order to prevent the prompt word from being too long and affecting the response speed of the target model, a capacity threshold may be set in advance. After obtaining the prompt word, the capacity of the prompt word may be compared with the capacity threshold.
[0101] If it is determined in box 404 that the capacity of the prompt word exceeds the capacity threshold, process 400 returns to box 403. In box 403, the information used to generate the prompt word can be trimmed. For example, in the case where the data query request includes multiple data query processes, the multiple data query processes can be split, and prompt words can be generated for each data query process respectively to reduce the length of each prompt word. For another example, the fields in the meta information can be trimmed to reduce the data volume of the meta information. For another example, the target model can be called to determine the target table from the table to be queried to further reduce the number of tables to be queried. Afterwards, the information used to generate the prompt word can be updated, the prompt word can be generated based on the updated information, and it is determined whether the capacity of the regenerated prompt word exceeds the capacity threshold.
[0102] If it is determined in block 404 that the capacity of the prompt word does not exceed the capacity threshold, process 400 proceeds to block 405. In block 405, application management platform 110 provides the prompt word to the target model, instructing the target model to generate one or more data query instructions.
[0103] Alternatively or additionally, the application management platform 110 may provide a prompt word to the target model to instruct the target model to determine the number of data query instructions to be generated (eg, N-1, where N is a positive integer). Afterwards, N-1 data query instructions may be generated by the target model.
[0104] In block 406, the application management platform 110 receives the data query instruction generated by the target model and can determine whether the generation of the data query instruction is complete, for example, whether N-1 data query instructions have been generated. If it is determined in block 406 that the target model has completed the generation of the data query instruction (for example, N-1 data query instructions have been generated), the process 400 ends.
[0105] If it is determined in block 406 that the target model has not completed the generation of the data query instruction, the process 400 proceeds to block 407. In block 407, the application management platform 110 provides the generated data query instruction as part of the prompt word to the target model, instructing the target model to generate one or more subsequent data query instructions (e.g., data query instructions that conform to the relational database language).
[0106] Figure 5 A flowchart of a data processing process 500 according to some embodiments of the present disclosure is shown. Figure 1 100 pairs of environments Figure 5 Process 500 is shown for illustration.
[0107] In block 501, the application management platform 110 may receive a data processing request. The data processing request may be triggered by a user, for example, the user may trigger the data processing request through the application management platform 110 or a device attached to the application management platform 110. The data processing request may also be triggered by another functional block of the application management platform 110. For example, the user input indicates that the target application generates an article, and the article needs to include a figure and / or a table. The other functional block may generate a data processing request, requesting the model call functional block to call the target model to generate a data processing instruction for the figure and / or the table.
[0108] In block 502, the application management platform 110 may initialize a trusted execution environment in response to a data processing request. The trusted execution environment may include, but is not limited to, a sandbox.
[0109] In block 503, the application management platform 110 may input target data into the trusted execution environment. The target data may include various data required to generate data processing instructions, such as configuration information and preset information of the model calling function block, user input, and processing results of the previous function block.
[0110] In block 504, the application management platform 110 may generate a prompt word based on the target data. Alternatively or additionally, before generating the prompt word, the application management platform 110 may perform feature extraction on the target data to obtain feature information of the target data, and may generate the prompt word based on the feature information.
[0111] At block 505 , the application management platform 110 may provide prompt words to a target model to generate a model output.
[0112] At block 506, the application management platform 110 may determine whether the model output includes data processing instructions. Alternatively or additionally, the application management platform 110 may determine whether the model data includes data processing instructions that meet a predetermined condition, such as data processing instructions that meet the Python language. Of course, the predetermined condition may also include other conditions, such as other protocols, other programming languages, and the like.
[0113] If it is determined in block 506 that the model output contains data layer care instructions, process 500 proceeds to block 507. In block 507, the application management platform 110 may perform data processing operations based on the data processing instructions. Alternatively or additionally, the data processing instructions may be used to perform any type of data processing, including but not limited to data conversion, data cleaning, sorting, various numerical operations, chart generation, etc.
[0114] Exemplarily, the data processing instruction can be used to perform an icon generation operation. Based on the data processing instruction, the application management platform 110 can perform operations such as chart type selection, chart dimension information generation, chart data calculation, and then output the generated chart, such as a chart in JSON (JavaScript Object Notation) format.
[0115] In block 508, the application management platform 110 may determine whether the execution result of the data processing operation is correct. It is understandable that the judgment conditions for determining whether the execution result is correct may be different when the data processing operation is different. For example, in the case of performing a data sorting operation, it may be determined whether the data sorting has been performed, whether the sorting meets the requirements input by the user, etc. In the case of performing an icon generation operation, it may be determined whether the execution result is correct from multiple dimensions of the chart, for example, it may be determined whether the information of each dimension such as the name, row, column, graphic, index, etc. of the chart meets the requirements input by the user or the requirements of the data processing request, etc.
[0116] If it is determined in block 508 that the execution result of the data processing operation is correct, process 500 proceeds to block 509. In block 509, the application management platform 110 determines whether the data processing operation has been executed. The judgment condition for determining whether the data processing operation has been executed depends mainly on the services provided by the target application, the requirements input by the user, the data processing request, and the specific type of the data processing operation. For example, in the case where the data processing operation is a chart generation operation, it can be determined whether a chart that meets the requirements input by the user has been generated. If it is determined in block 509 that the data processing operation has been executed, process 500 ends.
[0117] If it is determined in box 509 that the data processing operation has not been completed, process 500 proceeds to box 510. In box 510, the application management platform 110 generates a prompt word based on the execution result, provides the prompt word to the target model, continues to generate data processing instructions through the target model, and executes the data processing operation until it is determined that the data processing operation is completed.
[0118] If it is determined in block 508 that the execution result is an error, process 500 proceeds to block 511. In block 511, application management platform 110 determines whether the current number of retries exceeds the retry number threshold. If it is determined in block 511 that the current number of retries exceeds the retry number threshold, process 500 ends.
[0119] If it is determined in block 511 that the current number of retries does not exceed the retry number threshold, process 550 proceeds to block 512. In block 512, the application management platform regenerates prompt words according to the error information, provides the prompt words to the target model, regenerates data processing instructions through the target model, and executes data processing operations. The error information is used to identify or describe the specific error of the execution result, and the error information may include but is not limited to the error type, error prompt, error description, etc. Alternatively or additionally, the error information may include the error information of the current execution result, and may also include the error information of the previous multiple execution results.
[0120] If it is determined at block 506 that the model output of the target model does not include data processing instructions, process 500 proceeds to block 513. At block 513, application management platform 110 destroys the trusted execution environment and the target data input into the trusted execution environment. For example, a sandbox and the target data input into the sandbox may be destroyed.
[0121] In box 514, the application management platform 110 may determine whether it is still necessary to perform the data processing operation. The application management platform 110 may determine whether it is necessary to continue to perform the data processing operation based on the operation type of the data processing operation, the data processing capability of the target application, and the requirements of the user input. For example, the data processing request is used to request the execution of a chart generation operation. If it is determined that the target model does not output a data processing instruction for performing the chart generation operation, it is determined whether the target application includes a function block that can generate a chart independently of the target model. If it is determined that the target application is not configured with a function block that can generate a chart independently of the target model, the process 500 may be terminated. If it is determined that the target application has a function block for chart generation that is independent of the target model, it is determined whether the function block for chart generation can meet the requirements of the user input. If it is determined that the function block for chart generation cannot generate a chart that meets the requirements of the user input, the process 500 may be terminated.
[0122] If it is determined in block 514 that the data processing operation needs to be continued, the process 500 proceeds to block 515. In block 515, the application management platform 110 continues to perform the data processing operation corresponding to the data processing request. In the case of obtaining a data processing result that meets the data processing request, the data processing process 500 can be terminated. For example, the application management platform 110 calls a function block for chart generation that is not dependent on the target model to perform the chart generation operation. In the case of obtaining a chart that meets the data processing request or the requirements input by the user, the data processing process can be terminated.
[0123] In summary, according to the embodiments of the present disclosure, the calling method of the target model is made configurable by using an engineering approach, and the configuration of the calling method of the target model is visualized, which can reduce the development difficulty of users (such as developers) and is conducive to improving the stability of model output.
[0124] The embodiments of the present disclosure also provide corresponding devices for implementing the above methods or processes. Figure 6 A schematic structural block diagram of an apparatus 600 for application creation according to some embodiments of the present disclosure is shown. The apparatus 600 may be implemented in or included in the application management platform 110, for example. Each module / component in the apparatus 600 may be implemented by hardware, software, firmware, or any combination thereof.
[0125] As shown in the figure, the device 600 may include a selection module 610, a presentation module 620, a receiving module 630 and a creation module 640. The selection module 610 is configured to select a model calling function block for the model calling node to call the target model from multiple function blocks for the model calling node in multiple processing nodes of the target application to be created. The presentation module 620 is configured to present a configuration interface of the model calling function block in response to a configuration request for the model calling function block. The receiving module 630 is configured to receive configuration information for the model calling function block through the configuration interface. The creation module 640 is configured to create the target application based on the configuration information and the workflow composed of multiple processing nodes.
[0126] In some embodiments, the model calling function block is configured to call the target model based on the configuration information to generate a predetermined type of content.
[0127] In some embodiments, the receiving module 630 is further configured to: receive specified information for the target model as at least a part of the configuration information through a model configuration entry in the configuration interface.
[0128] In some embodiments, the receiving module 630 is further configured to: in response to receiving a trigger operation on a model configuration entry, present an identifier of at least one candidate model; receive a selection operation on a target identifier among the identifiers of at least one candidate model; and in response to the selection operation, determine that the candidate model corresponding to the target identifier is designated as the target model.
[0129] In some embodiments, the predetermined type of content includes structured data, and the receiving module 630 is further configured to: receive structural information indicating a data structure of the structured data as at least a part of the configuration information through an output configuration entry of the configuration interface.
[0130] In some embodiments, the predetermined type of content includes a data query instruction, and the receiving module 630 is further configured to: receive at least one query condition for generating the data query instruction as at least a part of the configuration information through a query configuration entry of the configuration interface.
[0131] In some embodiments, the receiving module 630 is further configured to: receive preset prompt information through a prompt input entry in the configuration interface, where the preset prompt information at least includes a generation requirement for a predetermined type of content.
[0132] In some embodiments, the device 600 also includes a response module, which is configured to: present an application creation interface in response to receiving a creation request for a target application; and present node identifiers corresponding to multiple processing nodes, respectively, and connection relationships between multiple node identifiers in the application creation interface in response to a configuration operation on the workflow of the target application, so as to indicate the workflow through the node identifiers and the connection relationships.
[0133] In some embodiments, the response module is further configured to: in response to an operation of selecting a model calling function block for a model calling node, present a node identifier corresponding to the model calling function block on the application creation interface.
[0134] In some embodiments, the presentation module 620 is further configured to present a configuration interface of the model calling function block in response to receiving a trigger operation on a node identifier corresponding to the model calling node.
[0135] In some embodiments, the device 600 also includes an operation module, which is configured to generate target prompt information based on the user input and configuration information in response to receiving user input through the target application; provide the target prompt information to the target model to obtain a model output; and determine response information for the user input based on the model output.
[0136] In some embodiments, the running module is further configured to: obtain historical interaction information matching the user input; obtain structural information and preset prompt information in the configuration information, the structural information indicating the data structure of the structured data, and the preset prompt information including the generation requirements of the structured data; and use the model calling function block to generate target prompt information for instructing the target model to generate structured data based on the user input, historical interaction information, structural information and preset prompt information.
[0137] The units and / or modules included in the device 600 can be implemented in various ways, including software, hardware, firmware, or any combination thereof. In some embodiments, one or more units and / or modules can be implemented using software and / or firmware, such as machine executable instructions stored on a storage medium. In addition to or as an alternative to machine executable instructions, some or all of the units and / or modules in the device 600 can be implemented at least in part by one or more hardware logic components. As an example and not limitation, exemplary types of hardware logic components that can be used include field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.
[0138] Figure 7 1 shows a block diagram of an electronic device 700 in which one or more embodiments of the present disclosure may be implemented. It should be understood that Figure 7 The electronic device 700 shown is merely exemplary and should not constitute any limitation on the functionality and scope of the embodiments described herein. Figure 7 The electronic device 700 shown may include or be implemented as Figure 1 application management platform 110, or Figure 6 Device 600.
[0139] like Figure 7 As shown, the electronic device 700 is in the form of a general electronic device. The components of the electronic device 700 may include, but are not limited to, one or more processors or processing units 710, a memory 720, a storage device 730, one or more communication units 740, one or more input devices 750, and one or more output devices 760. The processing unit 710 may be an actual or virtual processor and is capable of performing various processes according to a program stored in the memory 720. In a multi-processor system, multiple processing units execute computer executable instructions in parallel to improve the parallel processing capability of the electronic device 700.
[0140] The electronic device 700 typically includes a plurality of computer storage media. Such media may be any accessible media that is accessible to the electronic device 700, including but not limited to volatile and non-volatile media, removable and non-removable media. The memory 720 may be a volatile memory (e.g., registers, caches, random access memory (RAM)), a non-volatile memory (e.g., a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. The storage device 730 may be a removable or non-removable medium, and may include a machine-readable medium, such as a flash drive, a disk, or any other medium, which may be capable of being used to store information and / or data and may be accessed within the electronic device 700.
[0141] The electronic device 700 may further include additional removable / non-removable, volatile / non-volatile storage media. Figure 7 As shown in , a disk drive for reading or writing from a removable, non-volatile disk (e.g., a "floppy disk") and an optical drive for reading or writing from a removable, non-volatile optical disk may be provided. In these cases, each drive may be connected to the bus (not shown) by one or more data media interfaces. The memory 720 may include a computer program product 725 having one or more program modules that are configured to perform various methods or actions of various embodiments of the present disclosure.
[0142] The communication unit 740 implements communication with other electronic devices through a communication medium. Additionally, the functions of the components of the electronic device 700 can be implemented with a single computing cluster or multiple computing machines that can communicate through a communication connection. Therefore, the electronic device 700 can operate in a networked environment using a logical connection with one or more other servers, a network personal computer (PC), or another network node.
[0143] The input device 750 may be one or more input devices, such as a mouse, a keyboard, a tracking ball, etc. The output device 760 may be one or more output devices, such as a display, a speaker, a printer, etc. The electronic device 700 may also communicate with one or more external devices (not shown) through the communication unit 740 as needed, such as a storage device, a display device, etc., communicate with one or more devices that allow a user to interact with the electronic device 700, or communicate with any device that allows the electronic device 700 to communicate with one or more other electronic devices (e.g., a network card, a modem, etc.). Such communication may be performed via an input / output (I / O) interface (not shown).
[0144] According to an exemplary implementation of the present disclosure, a computer-readable storage medium is provided, on which computer-executable instructions are stored, wherein the computer-executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the method described above.
[0145] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices, equipment, and computer program products implemented according to the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.
[0146] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.
[0147] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, so that the instructions executed on the computer, other programmable data processing apparatus, or other device implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.
[0148] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple implementations of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of a module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some implementations as replacements, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be realized by a special hardware-based system that performs the function or action of the specification, or can be realized by a combination of special hardware and computer instructions.
[0149] The above descriptions of various implementations of the present disclosure are exemplary, non-exhaustive, and not limited to the disclosed implementations. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described implementations. The selection of terms used herein is intended to best explain the principles of the implementations, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the various implementations disclosed herein.
Claims
1. A method for creating an application, include: For a model calling node among a plurality of processing nodes of a target application to be created, selecting a model calling function block for the model calling node to call a target model from a plurality of function blocks; In response to a configuration request for the model calling function block, presenting a configuration interface of the model calling function block; Receiving configuration information for the model calling function block through the configuration interface; as well as The target application is created based on the configuration information and a workflow composed of the plurality of processing nodes. 2 . The method according to claim 1 , wherein the model calling function block is configured to call a target model based on the configuration information to generate a predetermined type of content.
3. The method according to claim 2, wherein the configuration information for calling the function block of the model is received through the configuration interface include: The specified information for the target model is received through the model configuration entry in the configuration interface as at least a part of the configuration information.
4. The method according to claim 3, wherein the specified information of the target model is received through the model configuration entry include: In response to receiving a triggering operation on the model configuration entry, presenting an identification of at least one candidate model; receiving a selection operation on a target identifier among the identifiers of the at least one candidate model; as well as In response to the selection operation, a candidate model corresponding to the target identifier is determined to be designated as the target model.
5. The method according to claim 2, wherein the predetermined type of content includes structured data, and receiving configuration information for calling a function block of the model through the configuration interface include: Structural information indicating a data structure of the structured data is received through an output configuration entry of the configuration interface as at least a part of the configuration information.
6. The method according to claim 2, wherein the predetermined type of content includes a data query instruction, and receiving configuration information for calling a function block of the model through the configuration interface include: At least one query condition for generating the data query instruction is received through the query configuration entry of the configuration interface as at least a part of the configuration information.
7. The method according to claim 2, wherein the configuration information for calling the function block of the model is received through the configuration interface include: The preset prompt information is received through the prompt input entry in the configuration interface, and the preset prompt information at least includes the generation requirement of the predetermined type of content.
8. The method according to claim 1, further comprising: include: In response to receiving a request to create the target application, presenting an application creation interface; as well as In response to a configuration operation on the workflow of the target application, node identifiers corresponding to the multiple processing nodes and connection relationships between the multiple node identifiers are presented in the application creation interface to indicate the workflow through the node identifiers and the connection relationships.
9. The method according to claim 8, further comprising: include: In response to the operation of selecting the model calling function block for the model calling node, a node identifier corresponding to the model calling function block is presented on the application creation interface.
10. The method according to claim 9, wherein a configuration interface of the model calling function block is presented include: In response to receiving a trigger operation on a node identifier corresponding to the model calling node, a configuration interface of the model calling function block is presented.
11. The method according to claim 1, further comprising: include: In response to receiving user input through the target application, generating target prompt information based on the user input and the configuration information using the model calling function block; Providing the target prompt information to the target model by using the model calling function block to obtain a model output; as well as Based on the model output, response information for the user input is determined.
12. The method according to claim 11, wherein the target prompt information is generated include: Acquire historical interaction information matching the user input; Acquire structural information and preset prompt information in the configuration information, wherein the structural information indicates a data structure of the structured data, and the preset prompt information includes generation requirements of the structured data; as well as The target prompt information for instructing the target model to generate structured data is generated by using the model calling function block based on the user input, the historical interaction information, the structure information and the preset prompt information.
13. An application creation device, include: A selection module is configured to select, for a model calling node among a plurality of processing nodes of a target application to be created, a model calling function block for the model calling node to call a target model from a plurality of function blocks; A presentation module, configured to present a configuration interface of the model calling function block in response to a configuration request for the model calling function block; A receiving module, configured to receive configuration information for calling a function block of the model through the configuration interface; as well as The creation module is configured to create the target application based on the configuration information and a workflow composed of the multiple processing nodes.
14. An electronic device, include: at least one processing unit; as well as At least one memory, the at least one memory being coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions causing the electronic device to perform the method according to any one of claims 1 to 12 when executed by the at least one processing unit.
15. A computer-readable storage medium having a computer program stored thereon, wherein the computer program can be executed by a processor to implement the method according to any one of claims 1 to 12.
16. A computer program product, comprising a computer program, wherein the computer program implements the method according to any one of claims 1 to 12 when executed by a processor.
Citation Information
Patent Citations
Model application scheme arrangement method and device, electronic equipment and storage medium
CN117271729A
Database query instruction generation method and device, computer equipment and storage medium
CN117909355A
Method and device for managing workflow, equipment and medium
CN118409681A
Application program construction method based on low-code platform and low-code platform
CN118444900A
