A configuration file generation method, device and system

Abstract

This invention discloses a configuration file generation method, apparatus, and system, comprising: inputting user-sent requirement information and rule prompt information, as well as a preset correspondence between a first function and its description information, into a large-scale pre-trained model; receiving first generation information and the function name of a first function returned by the large-scale pre-trained model; the first generation information is generated by the large-scale pre-trained model based on the requirement information and rule prompt information, and the function name of the first function is determined by the large-scale pre-trained model based on the first generation information and the correspondence between the first function and its description information; calling the corresponding first function based on the determined function name of the first function to obtain second function description information; inputting the second function description information into the large-scale pre-trained model; receiving the second generation information returned by the large-scale pre-trained model; and generating a configuration file based on the first generation information and the second generation information.

Description

Technical Field

[0001] This invention relates to a configuration file generation method, apparatus, and system, belonging to the field of software technology. Background Technology

[0002] The latest GPT API introduces function call functionality, which means that you can write multiple functions in your program and tell GPT the function descriptions of each function. GPT will then call the functions when necessary to obtain text information to help with GPT generation.

[0003] The common practice when communicating with GPT is to tell GPT all the requirements at once. This means that all the requirements, regardless of whether they are useful for text generation, are told to GPT, which wastes tokens and occupies a lot of context dialogue space. Summary of the Invention

[0004] Objective: In view of at least one of the above technical problems, the present invention provides a configuration file generation method, apparatus and system, which uses configuration files to represent the logic in the process by defining certain rules, and generates the logic in each log by mapping strings to functions, which greatly saves the space of context dialogue.

[0005] Technical solution: To solve the above technical problems, the technical solution adopted by the present invention is as follows:

[0006] In a first aspect, the present invention provides a configuration file generation method, wherein the execution subject is a first program, the method comprising:

[0007] In response to receiving the user's request information and rule prompt information, the system inputs the correspondence between the request information, rule prompt information, and the preset first function and first function description information into the large-scale pre-trained model.

[0008] Receive first generation information and the function name of the first function returned by the large-scale pre-trained model; wherein, the first generation information is generated by the large-scale pre-trained model based on the requirement information and rule prompt information, and the function name of the first function is determined by the large-scale pre-trained model based on the correspondence between the first generation information and the first function and the first function description information;

[0009] Based on the determined function name of the first function, call the corresponding first function to obtain the description information of the second function;

[0010] The second function description information is input into the large-scale pre-trained model; wherein, the second function description information is used to instruct the large-scale pre-trained model to generate second generated information;

[0011] Receive the second generated information returned by a large-scale pre-trained model;

[0012] Based on the first and second generated information, a configuration file is generated.

[0013] Furthermore, the first program is configured to be a program that, in response to a user's instruction, invokes the service of a large-scale pre-trained model to generate a configuration file.

[0014] Furthermore, the demand information is represented as textual information that instructs the direction of business process generation by a large-scale pre-trained model.

[0015] Furthermore, the rule prompt information is a text message used to prompt the configuration file writer to write rules for large-scale pre-trained models.

[0016] In some embodiments, the rule prompt information includes at least:

[0017] A jump between nodes indicates whether a jump is possible between any two nodes;

[0018] Jump conditions are used to control jumps between specific nodes, representing the specific conditions designed to implement jumps between two nodes;

[0019] The changes in the attributes of the next node after the jump, and the function used to change the attributes;

[0020] Parameters that can be freely configured in the function.

[0021] In some embodiments, the first function is a function used to return function description information corresponding to the second function, and the first program is able to run the first function to obtain the second function description information.

[0022] Secondly, this invention provides a configuration file generation method, wherein the execution entity is a large-scale pre-trained model, the method comprising:

[0023] Based on the requirement information and rule prompts, first generation information is generated, and based on the correspondence between the first generation information, the first function, and the first function description information, the function name of the first function is determined; and the first generation information and the function name of the first function are sent to the first program.

[0024] Second generation information is generated based on the second function description information; and the second generation information is sent to the first program; wherein the second function description information is obtained by the first program by calling the corresponding first function based on the function name of the determined first function; wherein the first generation information and the second generation information are configured so that the first program can generate a configuration file based on the first generation information and the second generation information.

[0025] Thirdly, the present invention provides a configuration file generation apparatus, wherein the execution subject is a first program, and the apparatus includes:

[0026] The first input module is used to: in response to receiving the user's request information and rule prompt information, input the request information and rule prompt information, as well as the pre-set correspondence between the first function and the first function description information, into the large-scale pre-trained model;

[0027] The first receiving module is configured to: receive first generation information and the function name of the first function returned by the large-scale pre-trained model; wherein the first generation information is generated by the large-scale pre-trained model based on the requirement information and rule prompt information, and the function name of the first function is determined by the large-scale pre-trained model based on the correspondence between the first generation information and the first function and the first function description information;

[0028] The calling module is used to: based on the determined function name of the first function, call the corresponding first function to obtain the description information of the second function;

[0029] The second input module is used to: input the second function description information into the large-scale pre-trained model; wherein the second function description information is used to instruct the large-scale pre-trained model to generate second generated information;

[0030] The second receiving module is used to: receive the second generated information returned by the large-scale pre-trained model;

[0031] The generation module is used to generate a configuration file based on the first generation information and the second generation information.

[0032] Fourthly, the present invention provides a configuration file generation apparatus, the execution entity of which is a large-scale pre-trained model, the apparatus comprising:

[0033] The first generation module is used to: generate first generation information based on requirement information and rule prompt information, and determine the function name of the first function based on the correspondence between the first generation information and the first function and the first function description information; and send the first generation information and the function name of the first function to the first program;

[0034] The second generation module is used to: generate second generation information based on the second function description information; and send the second generation information to the first program; wherein the second function description information is obtained by the first program by calling the corresponding first function based on the function name of the determined first function; wherein the first generation information and the second generation information are configured so that the first program can generate a configuration file based on the first generation information and the second generation information.

[0035] Fifthly, the present invention provides a configuration file generation system, comprising:

[0036] In response to receiving the user's request information and rule prompt information, the first program inputs the request information and rule prompt information, as well as the pre-defined correspondence between the first function and the first function description information, into the large-scale pre-trained model;

[0037] The large-scale pre-trained model generates first generated information based on demand information and rule prompts, and determines the function name of the first function based on the correspondence between the first generated information and the first function and its description information; and sends the first generated information and the function name of the first function to the first program.

[0038] The first program receives first generation information and the function name of the first function returned by the large-scale pre-trained model; wherein, the first generation information is generated by the large-scale pre-trained model based on the requirement information and rule prompt information, and the function name of the first function is determined by the large-scale pre-trained model based on the correspondence between the first generation information and the first function and the first function description information;

[0039] The first program calls the corresponding first function based on the determined function name of the first function to obtain the description information of the second function;

[0040] The first program inputs the second function description information into the large-scale pre-trained model; wherein, the second function description information is used to instruct the large-scale pre-trained model to generate second generated information;

[0041] The large-scale pre-trained model generates second generated information based on the second function description information; and sends the second generated information to the first program;

[0042] The first program receives the second generated information returned by a large-scale pre-trained model;

[0043] The first program generates a configuration file based on the first and second generated information.

[0044] Sixthly, the present invention provides an apparatus comprising,

[0045] Memory;

[0046] processor;

[0047] as well as

[0048] Computer programs;

[0049] The computer program is stored in the memory and configured to be executed by the processor to implement the method described in the first aspect above.

[0050] In a seventh aspect, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect.

[0051] Beneficial Effects: The configuration file generation method, apparatus, and system provided by this invention have the following advantages: Considering the limited context space in large-scale pre-trained models, during the configuration file generation process, all function description information of the second functions is not provided to the large-scale pre-trained model at the same time. Providing it to the large-scale pre-trained model at the same time would waste tokens and result in insufficient context space. This invention, by setting a first program and utilizing the function call function of a large-scale pre-trained model, such as GPT, allows the large-scale pre-trained model to determine which first function should be called based on the generated first information (each first function corresponds to first function description information, the text length of which is shorter than that of the second function description information). Given the function description information, the large-scale pre-trained model can understand the first information and the first function description information based on its own understanding capabilities, and select the first function description information that is most suitable for the current first information (and further determine the function name of the first function suitable for the current first information). After determining the function name of the first function, the first program uses the first function to further obtain the second function description information, and then sends the second function description information to the large-scale pre-trained model. The large-scale pre-trained model can further generate second generated information based on the second function description information. At this point, the required information in the configuration file is complete, and the first program can generate the corresponding configuration file based on the first generated information and the second generated information. Attached Figure Description

[0052] Figure 1 and Figure 2 This is a schematic diagram illustrating an application scenario of a configuration file generation method according to an embodiment of the present invention;

[0053] Figure 3 This is a schematic diagram of a configuration file generation method according to an embodiment of the present invention;

[0054] Figure 4 This is a schematic diagram of a data processing apparatus according to an embodiment of the present invention;

[0055] Figure 5 This is a schematic diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0056] The present invention will be further described below with reference to the accompanying drawings and embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention, and should not be used to limit the scope of protection of the present invention.

[0057] like Figure 1 As shown, before detailing the embodiments of the present invention, a specific scenario example is given to illustrate the scenarios in which the technical solution of the present invention can be used. The method of the present invention can be configured in... Figure 1 The server in the system can communicate with multiple clients.

[0058] The client application provides users with a visual interface for generating configuration files. It also includes an input interface where users can input their requirements and rule suggestions. For example, the requirement information could be "Please generate a configuration file about the reimbursement process." This requirement information is text that instructs the large-scale pre-trained model on the direction of the business process, which could be a reimbursement process, an approval process, or a process from procurement to payment, among others. The rule suggestions are text that guide the large model on how to write the rules for the configuration file. All of this text information is then transmitted to the server over the network.

[0059] The server, which has a built-in large-scale pre-trained model and a first program, is used to input requirement information and rule prompts, along with a pre-defined correspondence between the requirement information, rule prompts, and a first function and its description, into the large-scale pre-trained model. Based on the requirement information and rule prompts, the large-scale pre-trained model generates first generated information and determines the function name of the first function based on the correspondence between the first generated information and the first function and its description. The first program receives the first generated information and the function name of the first function returned by the large-scale pre-trained model; based on the determined function name, it calls the corresponding first function to obtain second function description information; and inputs the second function description information into the large-scale pre-trained model. The large-scale pre-trained model generates second generated information based on the second function description information and sends it to the first program. The first program generates a configuration file based on the first and second generated information. Finally, the configuration file can be returned to the user.

[0060] GPT function call functionality: This is a new feature introduced in the GPT API. We define multiple functions in our program and tell GPT the function description of each function. When calling the GPT API, GPT will automatically determine which function needs to be used to obtain text information.

[0061] token: GPT is a unit used to measure text length.

[0062] Contextual dialogue space: The maximum length of a complete session.

[0063] The aforementioned server can be an electronic device with a certain computing power. For example, a server can be a distributed system server, or a system with multiple processors, memory, network communication modules, etc., working together. A server can also be a cloud server, or an intelligent cloud computing server or intelligent cloud host with artificial intelligence technology. A server can also be a server cluster formed by several servers. Alternatively, with the development of science and technology, a server can also be a new technological means capable of realizing the corresponding functions of the implementation method described in the specification. For example, it can be a new form of "server" based on quantum computing.

[0064] The aforementioned client can be an electronic device with network access capabilities. Specifically, for example, the terminal can be a desktop computer, tablet computer, laptop computer, smartphone, etc. Alternatively, the terminal can also be software that can run on the electronic device.

[0065] The aforementioned network can be any type of network, which can use any of the various available protocols (including but not limited to TCP / IP, SNA, IPX, etc.) to support data communication. One or more networks can be a Local Area Network (LAN), an Ethernet-based network, a Token Ring network, a Wide Area Network (WAN), the Internet, a Virtual Network, a Virtual Private Network (VPN), an Intranet, an Extranet, a Public Switched Telephone Network (PSTN), an Infrared Network, a Wireless Network (e.g., Bluetooth, Wi-Fi), and / or any combination of these and / or other networks.

[0066] Example 1

[0067] Firstly, this embodiment provides a configuration file generation method, the execution subject of which is a first program, including:

[0068] In response to receiving the user's request information and rule prompt information, the system inputs the correspondence between the request information, rule prompt information, and the preset first function and first function description information into the large-scale pre-trained model.

[0069] Receive first generation information and the function name of the first function returned by the large-scale pre-trained model; wherein, the first generation information is generated by the large-scale pre-trained model based on the requirement information and rule prompt information, and the function name of the first function is determined by the large-scale pre-trained model based on the correspondence between the first generation information and the first function and the first function description information;

[0070] Based on the determined function name of the first function, call the corresponding first function to obtain the description information of the second function;

[0071] The second function description information is input into the large-scale pre-trained model; wherein, the second function description information is used to instruct the large-scale pre-trained model to generate second generated information;

[0072] Receive the second generated information returned by a large-scale pre-trained model;

[0073] Based on the first and second generated information, a configuration file is generated.

[0074] This embodiment also provides a configuration file generation method, wherein the execution subject is a large-scale pre-trained model, and the method includes:

[0075] Based on the requirement information and rule prompts, first generation information is generated, and based on the correspondence between the first generation information, the first function, and the first function description information, the function name of the first function is determined; and the first generation information and the function name of the first function are sent to the first program.

[0076] Second generation information is generated based on the second function description information; and the second generation information is sent to the first program; wherein the second function description information is obtained by the first program by calling the corresponding first function based on the function name of the determined first function; wherein the first generation information and the second generation information are configured so that the first program can generate a configuration file based on the first generation information and the second generation information.

[0077] User: Used to input requirement information and rule prompts into the first program. For example, the first program can provide an interactive interface to obtain the user's input requirement information and rule prompts, and send the final generated configuration file to the user.

[0078] Requirement information, such as "Please generate a configuration file about the reimbursement process," is a type of text information used to instruct large-scale pre-trained models to generate business process directions. These business process directions can be various processes such as reimbursement processes, approval processes, or processes from procurement to payment.

[0079] Rule hints are text messages used to suggest rules to large models in their configuration files.

[0080] The first procedure: the procedure that interacts with large-scale pre-trained models to generate configuration files, can be defined as a configuration file generation procedure.

[0081] In this first program, multiple first functions are pre-configured. These first functions are used to return function description text information corresponding to the second functions (the function description text can be understood as the API documentation of the second function, as well as the inputs, outputs, and examples of the second function). The first functions are shown below:

[0082]

[0083]

[0084] The first function can be written manually or can utilize a function that has already implemented the corresponding functionality; there are no specific limitations.

[0085] In this embodiment, the function description information of the corresponding second function in the second program is obtained using the first function configured in the first program.

[0086] The return statement mentioned above corresponds to the function description information of the second function.

[0087] The description mentioned above is the function description information for the first function. This first function description information is textual information that describes the function's purpose, type, or function.

[0088] In other words, the first program pre-configures the correspondence between the first function and its description, as well as the mapping relationship between the first function and its description. The description of the second function is the output of the first function. The description of the first function is written within the first function.

[0089] The first program can be configured with multiple first functions, and different first functions are used to output different description information of second functions.

[0090] Therefore, after the first function is determined, the first program can call the first function to obtain the function description text of the second function.

[0091] The second program is a program that generates business process logs based on configuration files, and can be defined as a business process generation program.

[0092] In some embodiments, such as Figure 2 As shown, a configuration file generation method includes:

[0093] Step S101: Receive the user's request information and rule prompt information.

[0094] In this embodiment, the aforementioned receiving action is performed by a first program, which is a program that responds to the user's instruction and calls the service of a large-scale pre-trained model to generate a configuration file, and can be defined as a configuration file generation program.

[0095] In this embodiment, the aforementioned requirement information is represented as textual information instructing the large-scale pre-trained model to generate a business process. This business process can be a business process in various application scenarios, such as a reimbursement process, an approval process, or a process from procurement to payment. For example, the requirement information could be "Please generate a configuration file about the reimbursement process."

[0096] In this embodiment, the rule suggestion information mentioned above refers to text information used to suggest rules for writing configuration files for large-scale pre-trained models. For example, the rule suggestion information could be:

[0097] What is the current node:[

[0098] {

[0099] "condition": "What is the condition for moving to the next node?"

[0100] "next": "Next node",

[0101] "attribute":{

[0102] "Attributes of the next node":{

[0103] "function": "What function is used to manipulate the current node's attribute and make it the value of the next node's attribute",

[0104] "parameter":{

[0105] What parameters does this function have?: Note the data types of the parameters.

[0106] }

[0107] }

[0108] }

[0109] }

[0110] }

[0111] You need to write the corresponding configuration information based on the business process direction generated from the requirements information. The above is a template for one node in the configuration information. Each business process configuration information includes multiple nodes. Nodes represent the stages involved in the business process. There are jumps between nodes, specifically controlled by the value field corresponding to the key field "condition". The "What is the current node?" field is the name of the current node to be generated. The value field corresponding to the key field "next" is the next node to jump to from the current node. The value field corresponding to the key field "attribute" is the attribute information of the next node. It can have multiple attribute information, and each attribute information can correspond to a function field. The function field is used to adjust the attribute information. Correspondingly, the function field can further correspond to parameter fields. The function field's function is to ensure that the node's attribute information changes as the business process progresses, used to fit the formal business process. The specific text format is as shown above, and will not be elaborated further.

[0112] Step S102: Input the correspondence between the requirement information and rule prompt information, as well as the preset first function and first function description information, into the large-scale pre-trained model.

[0113] In this embodiment, the input action is performed by the first program, and the first function is a function used to return function description information corresponding to the second function. That is, the first program can run the first function to obtain the second function description information.

[0114] After receiving the user's input requirements and rule prompts, the first program inputs the requirements, rule prompts, and the pre-defined first function and its description into the large-scale pre-trained model.

[0115] Step S103: The large-scale pre-trained model generates first generated information based on the demand information and rule prompt information, and determines the function name of the first function based on the correspondence between the first generated information and the first function and the first function description information.

[0116] Step S104: Receive the first generation information and the function name of the first function returned by the large-scale pre-trained model; wherein, the first generation information is generated by the large-scale pre-trained model based on the requirement information and rule prompt information, and the function name of the first function is determined by the large-scale pre-trained model based on the correspondence between the first generation information and the first function and the first function description information.

[0117] In this embodiment, the action of the first program is performed by the first program, the first generated information is information that the large-scale pre-trained model can directly generate based on the requirement information and rule prompt information without calling the API, and the second generated information is information generated after the large-scale pre-trained model uses the function call function. The first generated information may include at least key information and value information corresponding to some of the key information, and the second generated information is value information corresponding to other parts of the key information.

[0118] For example, the second generated information includes the value information for the key "function," which is the function name of the second function, and the value information for the key "parameter," which is the parameter of the second function. All other information is the first generated information.

[0119] Step S105: Based on the determined function name of the first function, call the corresponding first function to obtain the description information of the second function.

[0120] In this embodiment, when the first program determines the function name of the first function to be called, it calls the corresponding first function to obtain the description information of the second function.

[0121] Step S106: Input the second function description information into the large-scale pre-trained model; wherein, the second function description information is used to instruct the large-scale pre-trained model to generate second generated information.

[0122] Step S107: The large-scale pre-trained model generates second generated information based on the second function description information.

[0123] Step S108: Receive the second generated information returned by the large-scale pre-trained model.

[0124] Step S109: Generate a configuration file based on the first generation information and the second generation information returned by the large-scale pre-trained model.

[0125] Step S110: Return the configuration file to the user.

[0126] In summary, this invention provides a configuration file generation method that considers the limited context space in large-scale pre-trained models. Therefore, during the configuration file generation process, the function description information of all second functions is not provided to the large-scale pre-trained model at the same time. Providing it to the large-scale pre-trained model at the same time would waste tokens and result in insufficient context space. Therefore, this solution sets up a first program and utilizes the function call function of a large-scale pre-trained model, such as GPT, to allow the large-scale pre-trained model to determine which first function should be called based on the generated first information (each first function corresponds to first function description information, the text length of which is shorter than that of the second function description information). If the first information and the first function description information are obtained, the large-scale pre-trained model can understand them based on its own understanding capabilities, select the first function description information most suitable for the current first information, and further determine the function name of the first function suitable for the current first information. After determining the function name of the first function, the first program uses the first function to further obtain the second function description information. The first program then sends the second function description information to the large-scale pre-trained model. The large-scale pre-trained model can further generate second generated information based on the second function description information. At this point, the required information in the configuration file is complete, and the first program can generate the corresponding configuration file based on the first generated information and the second generated information.

[0127] Example 2

[0128] Secondly, based on Example 1, such as Figure 3 As shown, this embodiment provides a configuration file generation device, the execution body of which is a first program, including:

[0129] The first input module is used to: in response to receiving the user's request information and rule prompt information, input the request information and rule prompt information, as well as the pre-set correspondence between the first function and the first function description information, into the large-scale pre-trained model;

[0130] The first receiving module is configured to: receive first generation information and the function name of the first function returned by the large-scale pre-trained model; wherein the first generation information is generated by the large-scale pre-trained model based on the requirement information and rule prompt information, and the function name of the first function is determined by the large-scale pre-trained model based on the correspondence between the first generation information and the first function and the first function description information;

[0131] The calling module is used to: based on the determined function name of the first function, call the corresponding first function to obtain the description information of the second function;

[0132] The second input module is used to: input the second function description information into the large-scale pre-trained model; wherein the second function description information is used to instruct the large-scale pre-trained model to generate second generated information;

[0133] The second receiving module is used to: receive the second generated information returned by the large-scale pre-trained model;

[0134] The generation module is used to generate a configuration file based on the first generation information and the second generation information.

[0135] The present invention also provides a configuration file generation device, wherein the execution entity is a large-scale pre-trained model, the device comprising:

[0136] The first generation module is used to: generate first generation information based on requirement information and rule prompt information, and determine the function name of the first function based on the correspondence between the first generation information and the first function and the first function description information; and send the first generation information and the function name of the first function to the first program.

[0137] The second generation module is used to: generate second generation information based on the second function description information; and send the second generation information to the first program; wherein the second function description information is obtained by the first program by calling the corresponding first function based on the function name of the determined first function; wherein the first generation information and the second generation information are configured so that the first program can generate a configuration file based on the first generation information and the second generation information.

[0138] Example 3

[0139] Based on Embodiments 1 and 2, the present invention provides a configuration file generation system, comprising:

[0140] In response to receiving the user's request information and rule prompt information, the first program inputs the request information and rule prompt information, as well as the pre-defined correspondence between the first function and the first function description information, into the large-scale pre-trained model;

[0141] The large-scale pre-trained model generates first generated information based on demand information and rule prompts, and determines the function name of the first function based on the correspondence between the first generated information and the first function and its description information; and sends the first generated information and the function name of the first function to the first program.

[0142] The first program receives first generation information and the function name of the first function returned by the large-scale pre-trained model; wherein, the first generation information is generated by the large-scale pre-trained model based on the requirement information and rule prompt information, and the function name of the first function is determined by the large-scale pre-trained model based on the correspondence between the first generation information and the first function and the first function description information;

[0143] The first program calls the corresponding first function based on the determined function name of the first function to obtain the description information of the second function;

[0144] The first program inputs the second function description information into the large-scale pre-trained model; wherein, the second function description information is used to instruct the large-scale pre-trained model to generate second generated information;

[0145] The large-scale pre-trained model generates second generated information based on the second function description information; and sends the second generated information to the first program;

[0146] The first program receives the second generated information returned by a large-scale pre-trained model;

[0147] The first program generates a configuration file based on the first and second generated information.

[0148] Example 3

[0149] Thirdly, based on Example 1, such as Figure 4 As shown, this embodiment provides a device, including,

[0150] Memory;

[0151] processor;

[0152] as well as

[0153] Computer programs;

[0154] The computer program is stored in the memory and configured to be executed by the processor to implement the method described in Embodiment 1.

[0155] Example 4

[0156] Fourthly, based on Embodiment 1, this embodiment provides a storage medium on which a computer program is stored, and when the computer program is executed by a processor, it implements the method described in Embodiment 1.

[0157] Example 5

[0158] Fifthly, based on Embodiment 1, this embodiment proposes a hardware system that applies a configuration file generation method from Embodiment 1. This method is applied to the server side.

[0159] A terminal can be an electronic device with network access capabilities. For example, a terminal can be a desktop computer, tablet computer, laptop computer, smartphone, digital assistant, shopping guide terminal, television set, etc.

[0160] A server can be an electronic device with a certain computing power. For example, a server can be a distributed system server, or a system with multiple processors, memory, network communication modules, etc., working together. A server can also be a cloud server, or an intelligent cloud computing server or intelligent cloud host with artificial intelligence technology. A server can also be a server cluster formed by several servers. Alternatively, with the development of science and technology, a server can also be a new technological means capable of realizing the corresponding functions of the implementation method described in the manual. For example, it can be a new form of "server" based on quantum computing.

[0161] Example 6:

[0162] Sixthly, based on Embodiment 1, this embodiment also provides a computer program product containing instructions that, when executed by a computer, cause the computer to perform a configuration file generation method as described in Embodiment 1.

[0163] It is understood that the specific examples in this document are only intended to help those skilled in the art better understand the embodiments described herein, and are not intended to limit the scope of the invention.

[0164] It is understood that in the various embodiments described in this specification, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments described in this specification.

[0165] It is understood that the various implementation methods described in this specification can be implemented individually or in combination, and the implementation methods in this specification are not limited in this respect.

[0166] Unless otherwise stated, all technical and scientific terms used in the embodiments of this specification have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of this specification. The term "and / or" as used in this specification includes any and all combinations of one or more of the associated listed items. The singular forms "a," "the," and "the" as used in the embodiments of this specification and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0167] It is understood that the processor in the embodiments of this specification can be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method embodiments can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this specification. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this specification can be directly implemented by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above methods.

[0168] It is understood that the memory in the embodiments of this specification may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory may be random access memory (RAM). It should be noted that the memory in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0169] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this specification.

[0170] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the aforementioned method implementations, and will not be repeated here.

[0171] In the several embodiments provided in this specification, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0172] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0173] In addition, the functional units in the various embodiments of this specification can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0174] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of this specification, in essence, or the parts that contribute to the prior art, or parts of the technical solutions, can be embodied in the form of software products. These computer software products are stored in a storage medium and include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this specification. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0175] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A configuration file generation method, an execution subject of which is a first program, characterized by comprising: The method includes: In response to receiving user-sent requirement information and rule prompts, the system inputs the requirement information, rule prompts, and the pre-defined correspondence between a first function and its description into a large-scale pre-trained model. The rule prompts include at least: node transitions, indicating whether a transition between any two nodes is possible; transition conditions, used to specifically control node transitions, representing the specific conditions for implementing the transition between two nodes; the attribute change of the next node after the transition, and the function used to change the attribute; freely configurable parameters in the function; and the first function is a function that returns a function description corresponding to a second function, and the first program can run the first function to obtain the second function description. Receive first generation information and the function name of the first function returned by the large-scale pre-trained model; wherein, the first generation information is generated by the large-scale pre-trained model based on the requirement information and rule prompt information, and the function name of the first function is determined by the large-scale pre-trained model based on the correspondence between the first generation information and the first function and the first function description information; Based on the determined function name of the first function, call the corresponding first function to obtain the description information of the second function; The second function description information is input into the large-scale pre-trained model; wherein, the second function description information is used to instruct the large-scale pre-trained model to generate second generated information; Receive the second generated information returned by a large-scale pre-trained model; Based on the first and second generated information, a configuration file is generated.

2. The configuration file generation method according to claim 1, characterized in that, The first program is configured to generate a configuration file in response to a user's instruction and by invoking a service of a large-scale pre-trained model.

3. The configuration file generation method according to claim 1, characterized in that, The requirement information is represented as textual information that instructs the direction of business process generation by a large-scale pre-trained model.

4. The configuration file generation method according to claim 1, characterized in that, The rule suggestion information refers to text information used to suggest rules for writing configuration files for large-scale pre-trained models.

5. A configuration file generation method, wherein the execution entity is a large-scale pre-trained model, characterized in that, The method includes: Based on the requirement information and rule prompts, first generation information is generated, and based on the correspondence between the first generation information, the first function, and the first function description information, the function name of the first function is determined; the first generation information and the function name of the first function are then sent to the first program; wherein, the rule prompts include at least: jumps between nodes, indicating whether a jump is possible between any two nodes; jump conditions, used to specifically control jumps between nodes, indicating the specific conditions designed to implement jumps between two nodes; the changes in the attributes of the next node after the jump, and the function used to change the attributes; freely configurable parameters in the function; the first function is a function used to return function description information corresponding to the second function, and the first program can run the first function to obtain the second function description information; Second generation information is generated based on the second function description information; and the second generation information is sent to the first program; wherein the second function description information is obtained by the first program by calling the corresponding first function based on the function name of the determined first function; wherein the first generation information and the second generation information are configured so that the first program can generate a configuration file based on the first generation information and the second generation information.

6. A configuration file generation apparatus, wherein the execution subject is a first program, characterized in that, The device includes: The first input module is configured to: respond to receiving user-sent requirement information and rule prompt information, input the requirement information and rule prompt information, as well as the pre-defined correspondence between a first function and its description, into a large-scale pre-trained model; wherein, the rule prompt information includes at least: a jump between nodes, indicating whether a jump is possible between any two nodes; a jump condition, used to specifically control the jump between nodes, indicating the specific conditions for implementing the jump design between two nodes; the change in the attributes of the next node after the jump, and the function used for the attribute change; freely configurable parameters in the function; the first function is a function used to return function description information corresponding to a second function, and the first program can run the first function to obtain the second function description information; The first receiving module is configured to: receive first generation information and the function name of the first function returned by the large-scale pre-trained model; wherein the first generation information is generated by the large-scale pre-trained model based on the requirement information and rule prompt information, and the function name of the first function is determined by the large-scale pre-trained model based on the correspondence between the first generation information and the first function and the first function description information; The calling module is used to: based on the determined function name of the first function, call the corresponding first function to obtain the description information of the second function; The second input module is used to: input the second function description information into the large-scale pre-trained model; wherein the second function description information is used to instruct the large-scale pre-trained model to generate second generated information; The second receiving module is used to: receive the second generated information returned by the large-scale pre-trained model; The generation module is used to generate a configuration file based on the first generation information and the second generation information.

7. A configuration file generation device, wherein the execution entity is a large-scale pre-trained model, characterized in that, The device includes: The first generation module is used to: generate first generation information based on requirement information and rule prompt information; determine the function name of the first function based on the correspondence between the first generation information, the first function, and the first function description information; and send the first generation information and the function name of the first function to the first program; wherein, the rule prompt information includes at least: jumps between nodes, indicating whether a jump is possible between any two nodes; jump conditions, used to specifically control jumps between nodes, indicating specific conditions for implementing the jump design between two nodes; the change in the attribute of the next node after the jump, and the function used for the attribute change; freely configurable parameters in the function; the first function is a function used to return function description information corresponding to the second function, and the first program can run the first function to obtain the second function description information; The second generation module is used to: generate second generation information based on the second function description information; and send the second generation information to the first program; wherein the second function description information is obtained by the first program by calling the corresponding first function based on the function name of the determined first function; wherein the first generation information and the second generation information are configured so that the first program can generate a configuration file based on the first generation information and the second generation information.

8. A configuration file generation system, characterized in that, include: In response to receiving user-sent requirement information and rule prompts, the first program inputs the requirement information, rule prompts, and the pre-defined correspondence between a first function and its description into a large-scale pre-trained model. The rule prompts include at least: node transitions, indicating whether a transition between any two nodes is possible; transition conditions, used to specifically control node transitions, representing the specific conditions designed to implement the transition between two nodes; the attribute change of the next node after the transition, and the function used to change the attribute; freely configurable parameters in the function; the first function is a function used to return a function description corresponding to a second function, and the first program can run the first function to obtain the second function description. The large-scale pre-trained model generates first generated information based on demand information and rule prompts, and determines the function name of the first function based on the correspondence between the first generated information and the first function and its description information; and sends the first generated information and the function name of the first function to the first program. The first program receives first generation information and the function name of the first function returned by the large-scale pre-trained model; wherein, the first generation information is generated by the large-scale pre-trained model based on the requirement information and rule prompt information, and the function name of the first function is determined by the large-scale pre-trained model based on the correspondence between the first generation information and the first function and the first function description information; The first program calls the corresponding first function based on the determined function name of the first function to obtain the description information of the second function; The first program inputs the second function description information into the large-scale pre-trained model; wherein, the second function description information is used to instruct the large-scale pre-trained model to generate second generated information; The large-scale pre-trained model generates second generated information based on the second function description information; and sends the second generated information to the first program; The first program receives the second generated information returned by a large-scale pre-trained model; The first program generates a configuration file based on the first and second generated information.

Images