Method for developing tool function for intelligent agent through computing power of intelligent computing center

Abstract

The invention provides a method for developing a tool function for an intelligent agent through computing power of an intelligent computing center, and relates to the technical field of intelligent computing centers, intelligent computing centers and computing power infrastructures, and the method comprises the following steps: S1, a manufacturing intelligent agent receives task information provided by a user through an interactive interface and used for developing the tool function for a target intelligent agent; s2, the manufacturing agent determines development requirements of the tool function according to the task information, and the development requirements at least comprise function requirements of the tool function; and S3, the manufacturing agent automatically generates a code of the tool function of the target agent according to the development requirement of the tool function. In the invention, through the manufacturing agent running in the intelligent computing center, the process of communicating with the user, confirming the development demand, automatically completing the development of the tool function of the target agent and replacing the manual development of the tool function code can be realized, so that the development cost is effectively saved, the intelligent computing center can provide sufficient computing power resources, and the development efficiency is improved. And the development efficiency is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical fields of intelligent computing centers, intelligent computing centers, and computing power infrastructure, and particularly relates to a method for developing tool functions for intelligent agents through the computing power of an intelligent computing center. Background Art

[0002] With the rapid development of artificial intelligence technology, "intelligent computing centers" and "intelligent computing centers" have emerged as the times require.

[0003] An "intelligent computing center" refers to a facility that uses large-scale heterogeneous computing power resources, including general computing power and intelligent computing power, to mainly provide the required computing power, data, and algorithms for artificial intelligence applications (such as scenarios for developing, training, and inferring artificial intelligence deep learning models). An intelligent computing center covers facilities, hardware, and software, and can provide full-stack capabilities from underlying computing power to top-level application enabling.

[0004] An "intelligent computing center" includes, but is not limited to, an "intelligent computing center".

[0005] An "intelligent computing center", that is, an artificial intelligence computing center, is a type of computing power infrastructure that is based on artificial intelligence theory, adopts an artificial intelligence computing architecture, and provides computing power services, data services, and algorithm services required for artificial intelligence applications.

[0006] "Computing power" is the core of "intelligent computing centers" and "intelligent computing centers". It is the ability of computer devices or computing / data centers to process information, the ability of computer hardware and software to cooperate to jointly execute a certain computing requirement, the computing ability to achieve the output of the target result by processing information data, and a new type of productive force that integrates information computing power, network carrying capacity, and data storage capacity. It mainly provides services to society through computing power infrastructure.

[0007] An "intelligent agent" is an agent that can perceive the environment and take actions to achieve specific goals. It can be software, hardware, or a system, and has autonomy, adaptability, and interaction capabilities. An intelligent agent perceives changes in the environment (such as through sensors or data input), makes judgments and decisions based on the knowledge and algorithms learned by itself, and then executes actions to affect the environment or achieve a predetermined goal. Intelligent agents are widely used in the field of artificial intelligence, commonly found in automation systems, robots, virtual assistants, and game characters, etc. The core lies in its ability to learn independently and evolve continuously to better complete tasks and adapt to complex environments.

[0008] The development of an agent is a complex process. Usually, it is necessary to first develop the framework of the agent, and then develop the tool functions of the agent. The development of the tool functions of the agent, especially the development of domain-specific tool functions, currently needs to be completed by personnel with both profound artificial intelligence expertise and domain knowledge, resulting in high development costs and low development efficiency. Summary of the Invention

[0009] The present invention provides a method for developing tool functions for an agent by using the computing power of an intelligent computing center, which is used to solve the problems of high development costs and low development efficiency of agents.

[0010] To solve the above technical problems, the present invention is implemented as follows:

[0011] In a first aspect, the present invention provides a method for developing tool functions for an agent by using the computing power of an intelligent computing center, including:

[0012] Step S1: The manufacturing agent receives task information for developing tool functions for the target agent provided by the user through the interaction interface;

[0013] Step S2: The manufacturing agent determines the development requirements for the tool functions according to the task information, and the development requirements at least include: the functional requirements of the tool functions;

[0014] Step S3: The manufacturing agent automatically generates the code of the tool functions of the target agent according to the development requirements of the tool functions.

[0015] Optionally, step S2 includes:

[0016] Step S21: The manufacturing agent communicates with the user through the interaction interface and confirms the development requirements according to the task information.

[0017] Optionally, step S3 includes:

[0018] Step S31: The manufacturing agent performs at least one of the following according to the development requirements of the tool functions to obtain the code of the tool functions: document reading, code reading, architecture design, data analysis, code generation, code checking.

[0019] Optionally, it further includes:

[0020] Step S4: The manufacturing agent debugs the code of the tool functions based on the development requirements.

[0021] Optionally, the development requirements further include at least one of the following: requirement background, applicable scenarios of the tool functions, test cases, performance indicators.

[0022] Optionally, it further includes:

[0023] Step S5: The manufacturing agent saves the code of each version of the tool function;

[0024] Step S6: When the manufacturing agent debugs the code of the tool function, if the code of the current version does not meet the functional requirements, and / or the code of the current version does not meet the performance metrics, and / or the performance metrics of the code of the current version are lower than those of the code of the previous version, it reverts to the previous version.

[0025] Optionally, it further includes:

[0026] Step S5: The manufacturing agent records the second information and generates a development feature document based on the second information. The second information includes at least one of the following: the functional requirements of the tool function, the architecture design, the code, the version information of the code, the debugging information, where the debugging information includes at least one of the following: the input and output of the debugging, the debugging log, the intermediate data of the debugging.

[0027] Optionally, the manufacturing agent is one agent or multiple agents, and the multiple agents include at least one of the following agents: a requirement confirmation agent, a design agent, a coding agent, a debugging agent, an archiving agent.

[0028] In a second aspect, the present invention provides a device for developing a tool function for an agent by the computing power of an intelligent computing center, including:

[0029] A receiving module, configured to receive task information for developing a tool function for a target agent provided by a user through an interaction interface;

[0030] A determining module, configured to determine the development requirements of the tool function according to the task information, where the development requirements at least include: the functional requirements of the tool function;

[0031] A generating module, configured to automatically generate the code of the tool function of the target agent according to the development requirements of the tool function.

[0032] In a third aspect, the present invention provides an electronic device, including: a processor, a memory, and a program stored on the memory and executable on the processor. When the program is executed by the processor, it implements the steps of the method for developing a tool function for an agent by the computing power of an intelligent computing center as described in the first aspect above.

[0033] Fourthly, the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for developing tool functions for an intelligent agent by the computing power of an intelligent computing center as described in the first aspect above are implemented.

[0034] Fifthly, the present invention provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the steps of the method for developing tool functions for an intelligent agent by the computing power of an intelligent computing center as described in the first aspect above are implemented.

[0035] In the present invention, through the manufacturing intelligent agent running in the intelligent computing center, it is possible to communicate with the user and confirm clear and detailed development requirements, automatically complete the development of the tool functions of the target intelligent agent, and replace the process of manually developing tool function codes, thereby effectively saving development costs. Moreover, the intelligent computing center can provide sufficient computing power resources, greatly improving the development efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0037] Figure 1 is a schematic flowchart of the method for developing tool functions for an intelligent agent by the computing power of an intelligent computing center of the present invention;

[0038] Figure 2 is a schematic structural diagram of the device for developing tool functions for an intelligent agent by the computing power of an intelligent computing center of the present invention;

[0039] Figure 3 is a schematic structural diagram of the electronic device of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The technical solutions in the present invention will be clearly and completely described below with reference to the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without making creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0041] First, the technical terms related to the present invention will be briefly described below.

[0042] The "computing power" described in the present invention refers to: the ability of a computer device or a computing / data center to process information, which is the ability of computer hardware and software to cooperate to jointly execute a certain computing requirement, and is the computing ability to achieve the output of the target result by processing information data. It is a new type of productive force integrating information computing power, network carrying capacity, and data storage capacity, and mainly provides services to society through computing power infrastructure.

[0043] The "computational power" (Computational Power, CP) described in the present invention refers to: the ability of a data center server to process data and achieve result output, which is a comprehensive indicator for measuring the computing ability of a data center and includes general computing ability, supercomputing ability, and intelligent computing ability. The commonly used measurement unit is the number of floating-point operations per second (FLOPS, 1EFLOPS = 10^18 FLOPS), and the larger the value, the stronger the comprehensive computing ability. It is estimated that 1EFLOPS is approximately the computing power output of 5 Tianhe 2A or 500,000 mainstream server CPUs or 2 million mainstream laptops. The calculation formula is: CP = CP_general + CP_intelligent + CP_super

[0044] The "network power" (Network Power, NP) described in the present invention refers to: the manifestation of the data transmission ability of computing power facilities, which is a comprehensive ability including network architecture, network bandwidth, transmission delay, intelligent management and scheduling, etc., and involves network transmission within and between data centers. It is a comprehensive indicator for measuring network transmission scheduling ability.

[0045] The "storage power" (Storage Power, SP) described in the present invention refers to: the comprehensive ability of a data center in four aspects: data storage capacity, performance, security and reliability, and green and low-carbon. It is a comprehensive indicator for measuring the data storage ability of a data center and includes external storage devices such as storage arrays and server internal storage devices. The commonly used measurement unit for storage capacity is exabyte (EB, 1EB = 2^60 bytes), the commonly used measurement unit for performance is the number of read and write operations per second per unit capacity (IOPS / TB, Input / Output Operations Per Second / TB), and the disaster recovery ratio is an important manifestation of security and reliability.

[0046] The "computing power infrastructure" described in the present invention refers to: a new type of information infrastructure integrating information computing power, network carrying capacity, and data storage capacity, which can realize the centralized computing, storage, transmission, and application of information, and presents characteristics such as multi - element omnipresence, intelligent agility, security and reliability, and green and low - carbon.

[0047] The "new information infrastructure" described in the present invention refers to: mainly including network infrastructures such as 5G networks, fiber broadband networks, backbone networks, international communication networks, and satellite Internet, computing power infrastructures such as data centers, general computing power centers, intelligent computing centers, and supercomputing centers, and new technology facilities such as artificial intelligence, blockchain, and quantum computing. With the emergence and popularization of new general technologies, the form of the new information infrastructure will be more diverse.

[0048] The "computing power" described in the present invention includes: general computing power, intelligent computing power, and super computing power.

[0049] The "general computing power" described in the present invention refers to: the computing power provided by servers based on CPU (Central Processing Unit) chips, which is used to support basic general computing such as cloud computing and edge computing.

[0050] The "intelligent computing power" described in the present invention refers to: for various artificial intelligence innovation applications, a computing platform deployed on a large scale based on dedicated chips such as GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), and ASIC (Application Specific Integrated Circuit), such as natural language processing, machine vision, and so on.

[0051] The "super computing power" described in the present invention refers to: mainly the computing power provided by high-performance computing clusters such as supercomputers. It utilizes the centralized computing resources of a multi-computer system working in parallel and processes extremely complex or data-intensive problems through a dedicated operating system. It is mainly used for computing in cutting-edge scientific fields, such as planetary simulation, drug molecule design, gene analysis, etc.

[0052] The "intelligent computing center" described in the present invention refers to: a facility that mainly provides the required computing power, data, and algorithms for artificial intelligence applications (such as scenarios like artificial intelligence deep learning model development, model training, and model inference) by using large-scale heterogeneous computing power resources, including general computing power (CPU) and intelligent computing power (GPU, FPGA, ASIC, etc.). The intelligent computing center covers facilities, hardware, and software, and can provide full-stack capabilities from underlying computing power to top-level application enabling.

[0053] The "intelligent computing center" described in the present invention includes but is not limited to the "intelligent computing center".

[0054] The "intelligent computing center" described in the present invention, that is, the artificial intelligence computing center, is a type of computing power infrastructure based on artificial intelligence theory, adopting an artificial intelligence computing architecture, and providing computing power services, data services, and algorithm services required for artificial intelligence applications.

[0055] The "computing power center" described in the present invention refers to: a facility mainly composed of infrastructure such as wind, fire, water, and electricity, and IT software and hardware devices, with computing power, carrying capacity, and storage capacity, including general data centers, intelligent computing centers, supercomputing centers, etc.

[0056] The "supercomputing center" described in the present invention refers to: that is, a supercomputing data center, which is a data center based on supercomputers or large-scale computing clusters, capable of providing functions such as large-scale computing, storage, and network services, and is widely used in application scenarios such as aerospace, national defense, oil exploration, climate modeling, and genome sequencing.

[0057] The "computing power resources" described in the present invention refer to: technologies and facilities with information computing, transmission, storage, and application capabilities required for the development of the digital society, including but not limited to computing resources such as CPUs and GPUs, network resources such as switches and routers, storage resources such as storage arrays and distributed storage, security resources such as firewalls and intrusion detection systems, and support and guarantee resources such as wind, fire, water, and electricity.

[0058] The "large language model" described in the present invention refers to a large language model (LLM), which is a language model with a relatively large number of parameters, aiming to understand and generate human language, trained through a large amount of text data, and can perform a wide range of tasks including text summarization, translation, sentiment analysis, etc.

[0059] The "agent" described in the present invention refers to an entity that can perceive the environment and take actions to achieve specific goals. It can be software, hardware, or a system, with autonomy, adaptability, and interaction capabilities. The agent perceives changes in the environment (such as through sensors or data input), makes judgments and decisions based on the knowledge and algorithms it has learned, and then executes actions to affect the environment or achieve a predetermined goal. Agents are widely used in the field of artificial intelligence, commonly found in automated systems, robots, virtual assistants, and game characters, and their core lies in the ability to learn independently and evolve continuously to better complete tasks and adapt to complex environments.

[0060] The "tool function" described in the present invention can also be called the "tool method", which is an executable function method called by the agent to complete a specific task; it is an executable function method selected and called by the agent according to the task context; it aims to intelligently construct parameters and dynamically call them using the large language model to complete the subtasks.

[0061] To solve the problems of high development cost and low development efficiency of agents in related technologies, please refer to Figure 1, the present invention provides a method for developing tool functions for an intelligent agent by the computing power of an intelligent computing center, which can also be referred to as a method for manufacturing an intelligent agent by the computing power of an intelligent computing center to develop tool functions for a target intelligent agent. The method includes:

[0062] Step S1: The manufacturing intelligent agent receives the task information for developing tool functions for the target intelligent agent provided by the user through the interaction interface;

[0063] The "manufacturing intelligent agent" described in the present invention is constructed based on a large language model, has the ability to develop tool functions for other intelligent agents, and can communicate with the user in the interaction interface through natural language processing.

[0064] In the present invention, the manufacturing intelligent agent can be one intelligent agent or multiple intelligent agents. The multiple intelligent agents form a multi-intelligent agent cooperation system. Each intelligent agent has specific capabilities. The multiple intelligent agents can cooperate in the form of dialogue communication in the interaction interface to complete the task of developing tool functions for the target intelligent agent. The interaction interface can display the communication process between the multiple intelligent agents during the execution of the task.

[0065] In some embodiments, optionally, the target intelligent agent can be an intelligent agent that completes the development of framework code. The target intelligent agent includes one or more tool functions. The framework code of the target intelligent agent may only include the framework of the tool function (such as name, description, and / or function declaration, etc.), and does not include the specific implementation code of the tool function.

[0066] For example, a tool function in the framework code of the target intelligent agent contains the following code:

[0067]

[0068] There is a "pass" in the tool function, and the user needs to implement the part of the "pass" with code.

[0069] In some embodiments, optionally, the target intelligent agent can also be a complete-code intelligent agent that needs to add additional tool functions to improve its functions.

[0070] In the present invention, the target intelligent agent can be an intelligent agent for completing a specified task, such as data analysis, large language model training (which can also be referred to as fine-tuning the large language model), computing power management of the intelligent computing center, financial reimbursement, travel planning, etc. The tool function of the target intelligent agent is a functional method for completing a specific task. For example, if the target intelligent agent is an intelligent agent for fine-tuning the large language model, the tool function is a data processing method for processing the data for fine-tuning the large language model.

[0071] In the present invention, the task information may include the functional requirements of the tool function of the target intelligent agent, and the functional requirements refer to the functions that the tool function needs to implement, such as being able to perform data processing on the input data.

[0072] Step S2: The manufacturing intelligent agent determines the development requirements of the tool function according to the task information, and the development requirements at least include: the functional requirements of the tool function;

[0073] In some embodiments, optionally, the development requirements further include at least one of the following: requirement background, applicable scenario of the tool function, test case, performance metric.

[0074] Among them, the requirement background refers to the background for developing the tool function. The applicable scenario refers to the scenario to which the tool function applies, such as a scenario applicable to the analysis of platform user growth. The test case is a case used to verify the function of the developed tool function. The performance metric is used to indicate the performance of the tool function, and may include, for example, at least one of the following: throughput of the tool function, concurrency number, resource utilization rate, etc.

[0075] Step S3: The manufacturing intelligent agent automatically generates the code of the tool function of the target intelligent agent according to the development requirements of the tool function.

[0076] In the present invention, the manufacturing intelligent agent running in the intelligent computing center can communicate with the user to confirm clear and detailed development requirements, and automatically complete the development of the domain-customized tool function of the target intelligent agent, replacing the process of manually developing the tool function code, thereby effectively saving development costs, and the intelligent computing center can provide sufficient computing power resources, greatly improving the development efficiency.

[0077] In some embodiments, optionally, step S2 includes:

[0078] Step S21: The manufacturing intelligent agent communicates with the user through the interaction interface and confirms the development requirements according to the task information.

[0079] In some embodiments, the task information given by the user may be relatively simple. For example, "Please develop a tool function for monthly user growth analysis for our data engineer intelligent agent.", that is, the task information only includes: the functional requirement of the functional method "analyze monthly user growth". In this case, the manufacturing intelligent agent can actively communicate with the user through the interaction interface one or more times to obtain more clear and detailed development requirements. For example, the manufacturing intelligent agent can ask questions on the user interface, such as "What scenario is the tool function you need to develop applicable to", or "What throughput does the tool function you need to develop require", etc.

[0080] The communication development requirements may include at least one of the functional requirements of the tool function (such as more detailed functional requirements), requirement background, applicable scenarios of the tool function, test cases, performance metrics, etc.

[0081] The following is an example to illustrate the communication process between the user and the manufacturing agent in the present invention.

[0082] In one example, the conversation between the user and the manufacturing agent in the interaction interface is as follows:

[0083] “

[0084] User: Help me add a special data processing method to the data processing agent. It can convert the fine-tuning corpus data of different types of samples into the data format supported by Firefly.

[0085] Manufacturing agent: Can you specifically explain what forms "different types of data" may include? Such as Alpaca format, plain text dialogue, or other structures?

[0086] User: Currently, it is mainly the Alpaca format (JSONL structure, including instruction, input, and output fields) and unstructured dialogue text (such as multi-turn Q&A, each line may be in the form of "User: xxx\nAssistant: xxx"). Other formats may be extended in the future.

[0087] Manufacturing agent: I see. Can you provide an example of the target data format required by Firefly? What I understand may be a structure similar to {"prompt":"...","response":"..."}, right?

[0088] User: No. The format requirement of Firefly is that each piece of data must contain conversation_id (dialogue id), conversation (dialogue process), for example:

[0089]

[0090]

[0091] Manufacturing agent: Okay. Does the input of this tool function only support a single file path? Does the output directly overwrite the original file or generate a new file?

[0092] User: The input is a single file path. After processing, a new file that conforms to the Firefly format is generated. The file name is appended with the _firefly suffix, and the new path is returned. The original file is retained.

[0093] Agent for manufacturing: Is it necessary to process large-scale data? For example, what are the performance requirements when a single file exceeds 100,000 records?

[0094] User: In the current scenario, the number of records in a single file does not exceed 10,000.

[0095] Agent for manufacturing: For unstructured conversation texts (such as "User:...\nAssistant:..."), is it necessary to automatically build a multi-turn conversation structure? For example, convert the following:

[0096] ```

[0097] User: Recommend a science fiction movie

[0098] Assistant: "Interstellar"

[0099] User: What's it about?

[0100] Assistant: It tells the story of humans traveling through a wormhole to find a new home.

[0101] ```

[0102] Convert to:

[0103]

[0104]

[0105] User: Correct. At the same time, it is necessary to generate unified category metadata (default "General") for the entire conversation and automatically generate an incrementing conversation_id.

[0106] Agent for manufacturing: Understood. Can you provide some data samples for verifying the function?

[0107] User: You can use / mnt / data / sample_dialogs.txt and / mnt / data / qa.jsonl to verify the conversion effect.

[0108] Agent for manufacturing: Summarize the requirements as follows:

[0109] Develop Python utility functions for the data processing agent. Given the path of a single file as input, output the path of the converted new file.

[0110] Support the conversion from Alpaca format to Firefly format (map instruction to the first-round human and output to the first-round assistant).

[0111] For unstructured multi-turn conversations, build a conversation array structure that conforms to the Firefly standard.

[0112] Automatically generate necessary fields: conversation_id (auto-increment), category (configurable default value), dataset (inferred from the source);

[0113] The number of items in a single file does not exceed 10,000;

[0114] Sample data for verification, / mnt / data / sample_dialogs.txt and / mnt / data / qa.jsonl.

[0115] Partial thinking

[0116] - It is necessary to refer to the requirements document, design document, and code of the data processing agent.

[0117] - Consider different data format inputs. It may be possible to use the method of dynamically generating code by the large language model according to the input data format samples to implement data conversion for different input datasets, or even directly use the large language model for data format conversion.

[0118] - It is possible to consider using a small model to implement category tagging for the corpus.

[0119] User: Add two points: 1. Each conversation array should contain at least 1 group of human-assistant pairs. 2. Consider the fault tolerance ability to ensure the stability of the tool function and avoid frequent errors or processing the converted dataset with incorrect data.

[0120] Manufacturing agent: No problem, looking forward to the delivery.

[0121] ”.

[0122] Among them, the first to fourth items in the above “summarize requirements” (i.e., development requirements) are the functional requirements of the tool function, the fifth item “the number of items in a single file does not exceed 10,000” is the performance indicator, and the sixth item “sample data for verification, / mnt / data / sample_dialogs.txt and / mnt / data / qa.jsonl” is the test case.

[0123] From the above example, it can be seen that the manufacturing agent can be determined by communicating with the user based on at least one of the functional requirements of the tool function, the requirement background, the applicable scenario of the tool function, the test case, the performance indicator, etc.

[0124] Optionally, the step S3 includes:

[0125] Step S31: According to the development requirements of the tool function, the manufacturing agent performs at least one of the following to obtain the code of the tool function: document reading, code reading, architecture design, data analysis, code generation, and code inspection.

[0126] In some embodiments, the task information given by the user may be relatively simple, or it may be an atomic method. In one example, "Please add a bubble sort tool function to the agent." The manufacturing agent can directly generate code that conforms to the current agent's tool function specification.

[0127] In some embodiments, the task information given by the user is a tool function for domain-specific business. For example, "Please help me add a special data processing method to the data processing agent. It can convert the existing unstructured corpus data of the company into the corpus format of the A fine-tuning framework." The input data in this task information contains the corpus format specific to the company, and the output contains the data format specific to the A fine-tuning framework.

[0128] The manufacturing agent will complete this task through optional steps such as document reading, code reading, architecture design, data analysis, code generation, and code inspection. The document reading includes the project documents of the data processing agent project itself. Optionally, it can also read the archived log documents during debugging. The code reading means that the manufacturing agent reads the code of the data processing agent, including the overall framework code, existing tool function code, some basic tool class codes, etc. The architecture design includes at least one of the following optional design options, including module division, component interaction, class relationship design, interface definition, data structure design, etc. The data analysis refers to viewing the existing data sampling to understand the data format. Optionally, exploring and analyzing the data through code to fully understand the characteristics of the data to ensure the accuracy when generating development code. The code generation means that the manufacturing agent, according to the requirements confirmed by communicating with the user, combines the relevant information sorted out from the reading documents, code reading, architecture design, data analysis, etc., and adds the code of the newly generated tool function to the appropriate code file of the existing agent. Optionally, the manufacturing agent creates a new code file and adds the code of the newly generated tool function to it. The code inspection means that the manufacturing agent inspects the generated tool function code to ensure that the code meets the requirements. Optionally, it checks the scalability, compatibility, and readability of the code and gives a code inspection report and modification suggestions. The manufacturing agent regenerates the code according to the code inspection report and modification suggestions and then performs code inspection again until the code inspection passes.

[0129] In some embodiments, optionally, the method further includes:

[0130] Step S4: The manufacturing agent debugs the code of the tool function based on the development requirements.

[0131] In some embodiments, the manufacturing agent debugs the code of the tool function by running the tool function. Optionally, it runs a complete test case while monitoring the process log, output result, etc. of the tool function. The manufacturing agent determines whether the requirements are met by checking the process log and output result. If the requirements are not met, it gives an error cause analysis and archives it in a temporary document, then returns to step S3 to regenerate the code and enters step S4 for debugging again until the debugging step is passed.

[0132] In some embodiments, when the user has clear performance index requirements, in step S4, the manufacturing agent simulates the corresponding stress test call and statistics the performance index. If the performance index does not reach the performance index required by the user, it gives performance optimization suggestions, then returns to step S3 to regenerate the code and enters step S4 for debugging again until the debugging step is passed.

[0133] In some embodiments, the manufacturing agent can automatically debug the code of the tool function after generating the code of the tool function.

[0134] In some embodiments, the manufacturing agent can also debug the code of the tool function after the user gives a debugging task. At this time, the user can give the above debugging requirements in the given debugging task.

[0135] In some embodiments, optionally, the method further includes:

[0136] Step S5: The manufacturing agent saves the code of each version of the tool function;

[0137] Step S6: When the manufacturing agent debugs the code of the tool function, if the current version of the code does not meet the functional requirements, and / or the current version of the code does not meet the performance index, and / or the performance index of the current version of the code is lower than that of the previous version of the code, it rolls back to the previous version.

[0138] In some embodiments, optionally, the method further includes:

[0139] Step S7: The manufacturing agent records the second information and generates a development characteristics document based on the second information. The second information includes at least one of the following: the functional requirements of the tool function, the architecture design, the code, the version information of the code, the debugging information, where the debugging information includes at least one of the following: the input and output of the debugging, the debugging log, the intermediate data of the debugging.

[0140] Recording the above second information can facilitate the reference and optimization of similar tasks. Among them, recording different versions of the code can also avoid duplicate development during code development. For example, if the code of a certain version does not meet the functional requirements, record that version, and when rewriting the code next time, the previous mistakes can be avoided.

[0141] As mentioned in the above embodiments, the manufacturing agent may include multiple agents, and the multiple agents include at least one of the following agents: a requirement confirmation agent, a design agent, a coding agent, a debugging agent, and an archiving agent. Among them, the requirement confirmation agent is used to confirm the development requirements with the user, the design agent is used to determine the design scheme of the tool function, the coding agent is used to generate the code of the tool function, the debugging agent is used to debug the code of the tool function, and the archiving agent is used to generate development feature documents. Through the division of labor and cooperation of the above multiple agents, the development of the code of the tool function of the target agent can be completed quickly.

[0142] It should be noted that the tool function of the target agent that needs to be developed by the intelligent agent in the present invention can be one or multiple. For example, if three tool functions are required to complete a specific function of the target agent, the manufacturing agent can be required to develop three tool functions at this time.

[0143] In some embodiments, the method further includes:

[0144] Step S8: The manufacturing agent feeds back the development result to the user through the interaction interface, and the development result may include at least one of the following: notifying the user that the development of the tool function has been completed, the storage location of the code of the tool function (such as a download link), etc.

[0145] Please refer to Figure 2 , the present invention also provides a device 10 for developing a tool function for an intelligent agent by the computing power of an intelligent computing center, including:

[0146] A receiving module 11, configured to receive task information for developing a tool function for a target intelligent agent provided by a user through an interaction interface;

[0147] A determining module 12, configured to determine the development requirements of the tool function according to the task information, and the development requirements at least include: the functional requirements of the tool function;

[0148] A generating module 13, configured to automatically generate the code of the tool function of the target intelligent agent according to the development requirements of the tool function.

[0149] In some embodiments, optionally, the determining module 12 is configured to communicate with the user through the interaction interface according to the task information and confirm the development requirements.

[0150] In some embodiments, optionally, the generating module 13 includes: performing at least one of the following according to the development requirements of the tool function to obtain the code of the tool function: document reading, code reading, architecture design, data analysis, code generation, code checking.

[0151] In some embodiments, optionally, the apparatus 10 for developing a tool function for an intelligent agent by means of the computing power of an intelligent computing center further includes:

[0152] A debugging module, configured to debug the code of the tool function based on the development requirements.

[0153] In some embodiments, optionally, the development requirements further include at least one of the following: requirement background, functional requirements of the tool function, applicable scenarios of the tool function, test cases, performance metrics.

[0154] In some embodiments, optionally, the apparatus 10 for developing a tool function for an intelligent agent by means of the computing power of an intelligent computing center further includes:

[0155] A saving module, configured to save the code of each version of the tool function;

[0156] A fallback module, configured to fallback to the previous version when debugging the code of the tool function if the code of the current version does not meet the functional requirements, and / or the code of the current version does not meet the performance metrics, and / or the performance metrics of the code of the current version are lower than those of the code of the previous version.

[0157] In some embodiments, optionally, the apparatus 10 for developing a tool function for an intelligent agent by means of the computing power of an intelligent computing center further includes:

[0158] A recording module, configured to record second information and generate a development characteristics document based on the second information, where the second information includes at least one of the following: functional requirements of the tool function, architecture design, code, version information of the code, debugging information, where the debugging information includes at least one of the following: input and output of debugging, debugging log, intermediate data of debugging.

[0159] Please refer to Figure 3, the present invention further provides an electronic device 20, including a processor 21, a memory 22, and a computer program stored on the memory 22 and executable on the processor 21. When the computer program is executed by the processor 21, it implements each process of the above method embodiment of developing tool functions for an intelligent agent through the computing power of an intelligent computing center, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0160] The present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements each process of the above method embodiment of developing tool functions for an intelligent agent through the computing power of an intelligent computing center, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here. Among them, the computer-readable storage medium can be, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc.

[0161] The embodiment of the present application further provides a computer program product, including computer instructions, which when executed by a processor implement the above Figure 1 shown each process of the method embodiment of developing tool functions for an intelligent agent through the computing power of an intelligent computing center, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0162] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or device including that element.

[0163] Through the description of the above embodiments, those skilled in the art can clearly understand that the above method embodiments can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), and includes several instructions for causing a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0164] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit of the present invention and the scope protected by the claims, and all of them fall within the protection scope of the present invention.

Claims

1. A method for developing tool functions for intelligent agents using the computing power of an intelligent computing center, characterized in that: include: Step S1: The manufacturing agent receives task information of developing a tool function for a target agent provided by a user through an interactive interface; Step S2: the manufacturing agent determines the development requirements of the tool function according to the task information, and the development requirements at least include: functional requirements of the tool function; Step S3: The manufacturing agent automatically generates the code of the tool function of the target agent according to the development requirements of the tool function.

2. The method according to claim 1, characterized in that The step S2 comprises: Step S21: The manufacturing agent communicates with the user through the interactive interface based on the task information and confirms the development requirements.

3. The method according to claim 1, characterized in that The step S3 comprises: Step S31: The manufacturing agent executes at least one of the following according to the development requirements of the tool function to obtain the code of the tool function: document reading, code reading, architecture design, data analysis, code generation, and code checking.

4. The method according to claim 1, characterized in that: Also includes: Step S4: The manufacturing agent debugs the code of the tool function based on the development requirements.

5. The method according to claim 1 or 4, characterized in that: The development requirements also include at least one of the following: requirement background, applicable scenarios of tool functions, test cases, and performance indicators.

6. The method according to claim 4, characterized in that Also includes: Step S5: the manufacturing agent saves the code of each version of the tool function; Step S6: When the manufacturing intelligent agent is debugging the code of the tool function, if the current version of the code does not meet the functional requirements, and / or the current version of the code does not meet the performance indicators, and / or the performance indicators of the current version of the code are lower than the performance indicators of the previous version of the code, it will roll back to the previous version.

7. The method according to claim 1, characterized in that Also includes: Step S7: The manufacturing agent records the second information and generates a development feature document based on the second information, wherein the second information includes at least one of the following: functional requirements of the tool function, architecture design, code, version information of the code, and debugging information, wherein the debugging information includes at least one of the following: debugging input and output, debugging log, and debugging intermediate data.

8. A device for developing tool functions for intelligent agents through the computing power of an intelligent computing center, characterized in that: include: A receiving module, used for receiving task information for developing a tool function for a target agent provided by a user through an interactive interface; A determination module, used to determine the development requirements of the tool function according to the task information, wherein the development requirements at least include: functional requirements of the tool function; A generation module is used to automatically generate the code of the tool function of the target intelligent agent according to the development requirements of the tool function.

9. An electronic device, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, the steps of the method for developing tool functions for an intelligent agent through the computing power of an intelligent computing center are implemented as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for developing tool functions for an intelligent agent through the computing power of an intelligent computing center as described in any one of claims 1 to 7.

11. A computer program product, characterized in that It comprises computer instructions, which, when executed by a processor, implement the steps of the method for developing tool functions for an intelligent agent through the computing power of an intelligent computing center as described in any one of claims 1 to 7.

Images