Software development method and device based on large model, equipment and storage medium

Abstract

The invention discloses a software development method, device and equipment based on a large model and a storage medium, and relates to the technical field of automatic software development, and the method comprises the steps of obtaining software demand data and cue word data; processing the software demand data according to a preset processing strategy to obtain a function module function; and calling a corresponding function module function based on a preset large model and the cue word data to generate a target software framework so as to realize development of target software. Through large model-based visual data collection and software development, the problem of low manual development efficiency of middleware platform software codes is solved, automatic software development is realized, and the development efficiency and quality are improved.

Description

Technical Field

[0001] The present application relates to the technical field of automated software development, and in particular to a software development method, apparatus, device and storage medium based on a large model. Background Art

[0002] As the automotive electronic and electrical architecture evolves from distributed to centralized, domain controllers, as an important symbol of this change, play an increasingly critical role in modern smart cars. With its high-performance computing capabilities, high-bandwidth communication, reliable security and scalability, domain controllers have promoted their widespread application in the field of smart cars. In particular, the multifunctional integrated S32G domain controller can be used as a highly integrated solution to achieve centralized control and management of multiple vehicle systems, thereby reducing costs and improving efficiency. However, with the complexity of the domain controller architecture, the traditional software architecture also needs to be adjusted accordingly to adapt to new requirements. In this context, the domain controller middleware platform software came into being. Under the CP Autosar architecture, the middleware software platform interface is directly connected to the bottom software BSW layer, and the automatically generated code will be further processed for each system functional domain to call. However, existing in-vehicle software interface design tools, such as EA, mainly implement the establishment of interfaces between SWC and SWC applications, and the establishment of the middleware platform software interface and the bottom software BSW interface mostly relies on manual operations. This not only increases the workload of engineers, but also leads to a large amount of handwritten code in the middleware platform software, further affecting development efficiency and software quality.

[0003] Therefore, how to improve the development efficiency of the interfaces and functions of middleware software is a problem that needs to be solved urgently. Summary of the invention

[0004] The main purpose of this application is to provide a software development method, device, equipment and storage medium based on a large model, aiming to solve the technical problem of how to improve the development efficiency of the interface and functions of middleware software.

[0005] To achieve the above objectives, the present application proposes a software development method based on a large model, the method comprising:

[0006] Obtain software requirement data and prompt word data;

[0007] Processing the software requirement data according to a preset processing strategy to obtain a functional module function;

[0008] Based on the preset large model and the prompt word data, the corresponding functional module function is called to generate a target software framework to realize the development of the target software.

[0009] In one embodiment, the step of generating a target software framework by calling a corresponding functional module function based on a preset large model and the prompt word data includes:

[0010] Acquire internal data according to preset retrieval enhancement strategies;

[0011] Combining the internal data with the preset macro model to obtain a target macro model;

[0012] The target software framework is generated by calling the corresponding functional module function according to the prompt word data and the target large model.

[0013] In one embodiment, the functional module function includes a first interface function, a second interface function, a first functional function and a second functional function, and the step of calling the corresponding functional module function according to the prompt word data and the target large model to generate a target software framework includes:

[0014] Parsing the prompt word data according to the target macro model to obtain the demand type for developing software;

[0015] When the requirement type is the first type, calling the first function function and the first interface function to obtain a first function code and a first interface code, and generating a target software framework according to the first function code and the first interface code;

[0016] When the requirement type is the second type, the second function function and the second interface function are called to obtain a second function code and a second interface code, and a target software framework is generated according to the second function code and the second interface code.

[0017] In one embodiment, the step of processing the software requirement data according to a preset processing strategy to obtain a function module function includes:

[0018] Filtering the software requirement data according to a preset processing strategy to obtain requirement design data;

[0019] Generate corresponding functional module codes according to the required design data, and encapsulate the functional module codes to obtain functional module functions.

[0020] In one embodiment, the step of filtering the software requirement data according to a preset processing strategy to obtain requirement design data includes:

[0021] Obtaining diagnostic software requirement data and storage software requirement data according to the software requirement data;

[0022] The diagnostic software requirement data and the storage software requirement data are filtered according to a preset processing strategy to obtain requirement design data.

[0023] In one embodiment, after the step of generating a target software framework by calling a corresponding functional module function based on the preset large model and the prompt word data to realize the development of the target software, the method further includes:

[0024] Obtain user feedback data;

[0025] The preset large model is updated according to the feedback data to obtain an optimized large model.

[0026] In one embodiment, after the step of updating the preset large model according to the feedback data to obtain the optimized large model, the method further includes:

[0027] Scoring the software code generated by the optimized large model according to preset test indicators to obtain an evaluation score;

[0028] When the evaluation score is greater than a preset optimization score threshold, the updating of the preset large model is stopped.

[0029] In addition, to achieve the above purpose, the present application also proposes a software development device based on a large model, the device comprising:

[0030] A data acquisition module is used to acquire software requirement data and prompt word data;

[0031] A data processing module, used to process the software requirement data according to a preset processing strategy to obtain a function module function;

[0032] The software development module is used to generate a target software framework by calling corresponding functional module functions based on a preset large model and the prompt word data to realize the development of the target software.

[0033] In addition, to achieve the above-mentioned objectives, the present application also proposes a big model-based software development device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the big model-based software development method as described above.

[0034] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the large model-based software development method described above are implemented.

[0035] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the large model-based software development method as described above.

[0036] The present application provides a software development method based on a large model, and the method of the present application includes: obtaining software requirement data and prompt word data; processing the software requirement data according to a preset processing strategy to obtain a functional module function; calling the corresponding functional module function based on the preset large model and the prompt word data to generate a target software framework to achieve the development of the target software. In summary, the present application solves the problem of low manual development efficiency of the software code of the middleware platform through visual data collection and software development based on a large model, realizes software automated development, and improves development efficiency and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0039] Figure 1 A flowchart diagram of the first embodiment of the software development method based on a large model of the present application is provided;

[0040] Figure 2 A schematic diagram of a software development framework of an embodiment of a software development method based on a large model of the present application;

[0041] Figure 3 A flowchart diagram of the second embodiment of the software development method based on a large model of the present application;

[0042] Figure 4 A schematic diagram of a prompt word encapsulation scenario of an embodiment of a software development method based on a large model of the present application;

[0043] Figure 5 A flowchart diagram of the third embodiment of the software development method based on a large model of the present application;

[0044] Figure 6 A schematic diagram of the module structure of a software development device based on a large model according to an embodiment of the present application;

[0045] Figure 7 Schematic diagram of the device structure of the hardware operating environment involved in the large model-based software development method in the embodiment of the present application.

[0046] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0047] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0048] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0049] The main solution of the embodiment of the present application is: obtaining software requirement data and prompt word data; processing the software requirement data according to a preset processing strategy to obtain functional module functions; calling the corresponding functional module functions based on the preset large model and the prompt word data to generate a target software framework to realize the development of the target software.

[0050] As the automotive electronic and electrical architecture evolves from distributed to centralized, domain controllers, as an important symbol of this change, play an increasingly critical role in modern smart cars. With its high-performance computing capabilities, high-bandwidth communication, reliable security and scalability, domain controllers have promoted their widespread application in the field of smart cars. In particular, the multifunctional integrated S32G domain controller can be used as a highly integrated solution to achieve centralized control and management of multiple vehicle systems, thereby reducing costs and improving efficiency. However, with the complexity of the domain controller architecture, the traditional software architecture also needs to be adjusted accordingly to adapt to new requirements. In this context, the domain controller middleware platform software came into being. Under the CP Autosar architecture, the middleware software platform interface is directly connected to the bottom software BSW layer, and the automatically generated code will be further processed for each system functional domain to call. However, existing in-vehicle software interface design tools, such as EA, mainly implement the establishment of interfaces between SWC and SWC applications, and the establishment of the middleware platform software interface and the bottom software BSW interface mostly relies on manual operations. This not only increases the workload of engineers, but also leads to a large amount of handwritten code in the middleware platform software, further affecting the development efficiency and software quality. Therefore, how to improve the development efficiency of the interfaces and functions of middleware software is a problem that needs to be solved urgently.

[0051] This application solves the problem of low efficiency of manual development of software code on the middleware platform through visual data collection and software development based on large models, realizes software automated development, and improves development efficiency and quality.

[0052] It should be noted that the execution subject of this embodiment can be a software development system based on a large model, or a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device that can realize the software development function based on the large model, etc., and this embodiment does not specifically limit this. The following takes the software development system based on the large model as an example to illustrate this embodiment and the following embodiments.

[0053] Based on this, the embodiment of the present application provides a software development method based on a large model, referring to Figure 1 , Figure 1 This is a flowchart of the first embodiment of the software development method based on a large model of the present application.

[0054] In this embodiment, the software development method based on the large model includes steps S10 to S30:

[0055] Step S10: Acquire software requirement data and prompt word data.

[0056] It should be noted that in this step, the system will automatically identify and collect the required software requirement data from the cloud software requirement database through the data processing module. These data include but are not limited to the fault code data required for diagnostic software development, read and write DID (Data Identifier) ​​data, routine control RID (Routin Identifier) ​​data, etc., as well as memory address data information required for storage software development, memory allocation information data of each functional module, etc. At the same time, prompt word data is also obtained. These data are mainly used to standardize and simplify interface calls when calling large language models in subsequent steps, as well as to enrich the expression of user questions, so that the model can understand user intentions more comprehensively and accurately.

[0057] In addition, it should be noted that software requirement data refers to all relevant information required to develop specific software functions, including but not limited to functional descriptions, input and output requirements, performance indicators, etc., while prompt word data refers to a set of keywords and phrases used to guide large language models to perform specific task processing.

[0058] Step S20: Process the software requirement data according to a preset processing strategy to obtain a functional module function.

[0059] It should be noted that in this step, the preset processing strategy refers to the data processing rules and methods formulated according to the best practices of software development and the needs of specific projects. Functional module functions refer to code blocks that implement specific software functions, which are usually encapsulated in the form of functions to facilitate subsequent calls and tests. Specifically, in this step, the system will automatically process the collected software requirement data through the data processing module, including deleting the title information of table data, filtering and deleting duplicate data, etc., to obtain available requirement data that can be identified and processed by the application layer module. Subsequently, the application layer processing module automatically generates the functional module code corresponding to the software requirements based on these processed requirement data and the preset program script. Each software requirement corresponds to a piece of functional module code, and all of them are encapsulated in the form of functions to prepare for the subsequent processing calls of the service layer module.

[0060] In a feasible implementation manner, the step S20 specifically includes:

[0061] Step S201: filtering the software requirement data according to a preset processing strategy to obtain requirement design data.

[0062] It should be noted that the main purpose of this step is to pre-process the original software requirement data collected from the cloud software requirement database to obtain requirement design data that can be used for subsequent code generation. The preset processing strategy includes but is not limited to operations such as deleting the title information of table data and filtering and deleting duplicate data to ensure the accuracy and availability of the data. For example, after collecting the fault code data, read and write DID data, routine control RID data, etc. required for diagnostic software development, the system will first identify and delete the title information in these data to avoid confusion in subsequent processing. At the same time, the system will also filter and delete duplicate data to ensure that each data item is unique and valid. After these processes, the obtained requirement design data will be clearer and more accurate, which improves the accuracy of function generation of functional modules.

[0063] In addition, it should be noted that the demand design data (i.e., usable demand data) refers to demand data that can be directly used for code generation after preprocessing. These data have removed redundant information, and each data item is unique and valid, so they can be directly used by subsequent processing modules.

[0064] In a feasible implementation manner, the step S201 specifically includes:

[0065] Step A10: Obtain diagnostic software requirement data and storage software requirement data according to the software requirement data.

[0066] It should be noted that in this step, the system will automatically identify and collect the required software requirement data from the cloud software requirement database. These data include but are not limited to the fault code data, read and write DID data, routine control RID data, etc. required for diagnostic software development, as well as the memory address data information required for storage software development, memory allocation information data of each functional module, etc. Next, the system will classify and process the collected software requirement data, dividing it into diagnostic software requirement data and storage software requirement data. Diagnostic software requirement data is mainly used for the development of diagnostic application software, including but not limited to communication timeout fault codes, functional safety fault codes, and diagnostic interface codes, etc.; while storage software requirement data is mainly used for the development of storage software, including but not limited to underlying software and storage application interface codes, storage function codes, etc.

[0067] In addition, it should be noted that diagnostic software requirement data and storage software requirement data are two types of software requirement data. Diagnostic software requirement data refers to data required for developing diagnostic software, including but not limited to various codes, parameters and configuration information related to vehicle fault diagnosis; while storage software requirement data refers to data required for developing storage software, including but not limited to various codes, parameters and configuration information related to data storage.

[0068] Step A20: filtering the diagnostic software requirement data and the storage software requirement data according to a preset processing strategy to obtain requirement design data.

[0069] It should be noted that in this step, the preset processing strategies include but are not limited to data deduplication, data cleaning, data standardization, etc. Specifically, the system will deduplicate the diagnostic software requirement data and the storage software requirement data to ensure that each piece of data is unique, avoiding duplicate code and redundant information in the subsequent development process. Next, the data will be cleaned to remove invalid data and abnormal data. It is understandable that these invalid data and abnormal data are caused by errors in the data collection process or defects in the data itself, which will interfere with and affect the subsequent development work. Finally, the system will standardize the data to make it conform to the format and requirements in the subsequent development process. This includes but is not limited to the conversion of data types, adjustment of data formats, etc.

[0070] Step S202: Generate corresponding functional module codes according to the required design data, and encapsulate the functional module codes to obtain functional module functions.

[0071] It should be noted that after obtaining the demand design data, the system will automatically generate function module codes according to the demand design data, and encapsulate these codes to form function module functions that can be used for subsequent calls. In addition, it should be noted that the function module code refers to the code (such as interface function code, communication diagnosis code and storage function code) generated according to the demand design data to realize specific functions. The function module function refers to the function that can be called after encapsulating these codes. These functions will be applied to the corresponding software in the subsequent development process to realize different functional requirements. For example, in the development process of the general diagnostic application platform, the system will automatically generate the corresponding diagnostic function code according to the fault code data in the demand design data. These codes will include functions such as fault code identification, diagnostic logic implementation and fault information output. Similarly, in the development process of the storage application platform, the system will automatically generate storage function codes according to the memory address data information in the demand design data to realize data storage and reading operations. In addition, the system will also automatically generate interface function codes according to the interface requirements of the bottom software and the middleware platform software to realize communication and data exchange between the two.

[0072] Step S30: Based on the preset large model and the prompt word data, the corresponding functional module function is called to generate a target software framework to realize the development of the target software.

[0073] It should be noted that if Figure 2 As shown, in this step, the system will use the service layer processing module to use the interface call of the large language model to standardize and simplify the prompt word encapsulation technology, simplifying and standardizing the calling process of the large language model. At the same time, by calling different large model services and combining the previously generated functional module functions, the target software framework is automatically generated. For example, when developing a general diagnosis application, the system will call the large language model to generate functional module code and interface module code related to general diagnosis based on the processed diagnostic requirement data and the prompt word data entered by the user, thereby automatically generating the entire general diagnosis software framework.

[0074] In addition, it should be noted that the preset large model refers to a pre-trained large language model with powerful natural language processing and code generation capabilities. The target software framework refers to a software architecture generated according to software requirements and data that includes all necessary functional modules and structures.

[0075] This embodiment provides a software development method based on a large model, and the method of this embodiment includes: obtaining software requirement data and prompt word data; processing the software requirement data according to a preset processing strategy to obtain a functional module function; calling the corresponding functional module function based on the preset large model and the prompt word data to generate a target software framework to achieve the development of the target software. In summary, this embodiment solves the problem of low manual development efficiency of middleware platform software code through visual data collection and software development based on a large model, realizes software automated development, and improves development efficiency and quality.

[0076] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can refer to the above introduction, and will not be repeated later. Figure 3 , Figure 3 This is a flow chart of the second embodiment of the software development method based on a large model of the present application, wherein step S30 specifically includes:

[0077] Step S301: Acquire internal data according to a preset search enhancement strategy.

[0078] It should be noted that the preset retrieval enhancement strategy is the RAG (Retrieve-and-Generate) technology. It is understandable that the purpose of this step is to efficiently retrieve information related to the current development task (ie, internal data) from a large amount of data for use in subsequent steps.

[0079] Specifically, the system will determine the specific requirements of the current development task, for example, which middleware platform software interface needs to be developed, or which functional module code needs to be optimized. Then, based on these requirements, RAG technology is used to search in the internal database. The internal database contains various types of data such as historical code, software architecture documents, and demand analysis reports. RAG technology will automatically retrieve the most relevant information from the database as internal data based on the requirements of the current task.

[0080] In addition, it should be noted that internal data refers to data that is directly related to the current development task and has been screened and sorted. These data may come from multiple different data sources, but after being processed by RAG technology, they are all converted into information that is helpful for the development task.

[0081] Step S302: combining the internal data with the preset macro model to obtain a target macro model.

[0082] It should be noted that the preset large model is a large language model with powerful learning and reasoning capabilities. It is understandable that the purpose of this step is to combine the internal data with the preset large model to obtain a target large model optimized for the current development task. Specifically, the system will input the obtained internal data into the preset large model, and then use the learning ability of the preset large model to process and analyze the internal data to extract key information and patterns. Based on this information and pattern, the system will adjust and optimize the preset large model to obtain a target large model optimized for the current development task.

[0083] Step S303: generating a target software framework by calling corresponding functional module functions according to the prompt word data and the target large model.

[0084] It should be noted that if Figure 4 As shown, the prompt word data is a specific word or phrase input by the user to guide the target large model to generate and call code. Specifically, in this step, the system will analyze the prompt word data according to the large model and construct corresponding instructions or query statements. The target large model will determine the requirements of the current development task based on the instructions or query statements, and call the corresponding functional module functions according to the requirements to generate the code of the target software framework.

[0085] In a feasible implementation manner, the step S303 specifically includes:

[0086] Step B10: parsing the prompt word data according to the target macro model to obtain the requirement type for developing software.

[0087] It should be noted that if Figure 4 As shown, in this step, the system will receive the prompt word data transmitted by the interaction layer, which contains information such as description of software functions, performance requirements, interface requirements, etc. Then the system will parse the received prompt word data through the target large model, and use the deep learning ability and internal data of the large model to understand and identify the type of requirements for the development software. The types of requirements include but are not limited to diagnostic functions, storage functions, communication functions, etc. It can be understood that the role of this step is to accurately identify the type of software requirements and provide data support for the subsequent calls to the corresponding functional functions and interface functions.

[0088] Step B20: When the requirement type is the first type, the first function function and the first interface function are called to obtain a first function code and a first interface code, and a target software framework is generated according to the first function code and the first interface code.

[0089] It should be noted that the first function and the first interface function are predefined. Assume that in this step, the first type of requirement type refers to the "diagnostic function type". Specifically, the system will search for the corresponding first function and the first interface function in the function function library and the interface function library according to the identified requirement type (i.e., the diagnostic function type). The first function is called, and the function generates corresponding diagnostic function codes according to the requirements, such as fault code identification, fault code clearing and other function codes. At the same time, the first interface function is called, and the function is responsible for generating the interface code required to interact with the underlying software, such as the communication protocol interface code related to diagnosis. Finally, the system will integrate the generated first function code and the first interface code, and generate the target software framework according to the preset software architecture and design specifications.

[0090] Step B30: When the requirement type is the second type, the second function function and the second interface function are called to obtain a second function code and a second interface code, and a target software framework is generated according to the second function code and the second interface code.

[0091] It should be noted that the specific process is similar to step B20, except that the function functions and interface functions called are different, and the generated codes and functions are also different. Assume that in this step, the second type of requirement type refers to the "storage function type". Specifically, the system will search for the corresponding second function function and second interface function in the function function library and the interface function library according to the identified requirement type (i.e., storage function type). Call the second function function, which generates corresponding storage function codes according to the requirements, such as data reading and writing, data backup and other function codes. At the same time, call the second interface function, which is responsible for generating the interface code required to interact with the underlying software, such as storage-related communication protocol interface code. Finally, the system will integrate the generated second function code and second interface code, and generate the target software framework according to the preset software architecture and design specifications.

[0092] In addition, it should be noted that the second function and the second interface function are also predefined, and they generate corresponding codes according to the requirement type. At the same time, the generation of the target software framework depends on the preset software architecture and design specifications. In addition, according to different requirement types, other function functions and interface functions can also be called to generate the corresponding software framework.

[0093] In this embodiment, by combining internal data and a preset large model, the prompt word data is intelligently parsed to determine the type of software requirement, and the corresponding functional functions and interface functions are called to generate the target software framework, thereby realizing automation and intelligence of software development and improving development efficiency and software quality.

[0094] Based on the first and second embodiments of the present application, in the third embodiment of the present application, the same or similar contents as those in the first and second embodiments can be referred to the above description, and will not be described in detail later. Figure 5 , Figure 5 This is a flowchart of the third embodiment of the software development method based on a large model of the present application. After step S30, steps S40-S70 are also included:

[0095] Step S40: Obtain user feedback data.

[0096] It is understandable that obtaining user feedback data is mainly used to continuously optimize and iterate the preset large model (i.e., large language model, LLM).

[0097] It should be noted that in this step, the system will establish a feedback channel between the user and the system through the interactive layer processing module. Specifically, when the user is developing or testing the software code generated by the system, if the code is found to have defects, does not meet the expected functions, or has other areas that can be optimized, the user can submit feedback data through the system's visual interface. Such feedback data includes but is not limited to code error information, performance bottleneck descriptions, user suggestions, and expected improvement directions.

[0098] Step S50: updating the preset large model according to the feedback data to obtain an optimized large model.

[0099] It should be noted that updating the preset large model according to user feedback data is the key to improving model performance, adapting to new requirements and changes in software architecture. It can be understood that this step mainly uses user feedback data, combined with preset update algorithms and training strategies, to iteratively train the preset large model. For example, after receiving user feedback on the problem of software code generation on the diagnostic middleware platform, the system will first analyze the error information, performance bottlenecks and other key information in the feedback data, and then use this information to conduct targeted training and adjustments to the preset large model. During the training process, the system will continuously try to generate new code samples, and verify and evaluate the generated code through automated testing tools. Based on the evaluation results, the system will adjust the parameters and structure of the model to improve the quality and efficiency of its generated code.

[0100] In addition, it should be noted that the update of the preset large model is an ongoing process, which requires continuous receipt of user feedback data and iterative training and optimization. Through this process, the model can gradually adapt to new requirements and software architecture changes, and maintain the timeliness and effectiveness of its generated code.

[0101] Step S60: scoring the software code generated by the optimized large model according to preset test indicators to obtain an evaluation score.

[0102] It should be noted that, as mentioned above Figure 2 As shown, in this step, the system will complete it through the service layer processing module and the interaction layer processing module. The service layer processing module is responsible for generating test samples and calling the automated testing tool to test and evaluate the generated code; the interaction layer processing module is responsible for displaying the test results and scoring information to the user. For example, when developing a storage application software for a vehicle domain controller, the system can use the optimized large model to generate the code for the storage application platform. Then, the system will score the generated code using preset test indicators (such as code correctness, execution efficiency, memory usage, etc.). The scoring results will be displayed in real time on the system's visual interface for user reference and evaluation.

[0103] In addition, it should be noted that the preset test indicators are an important basis for evaluating code quality, which needs to be customized and adjusted according to specific development scenarios and requirements. By selecting appropriate test indicators and evaluation methods, the quality and performance level of the model-generated code can be more accurately reflected.

[0104] Step S70: When the evaluation score is greater than a preset optimization score threshold, stop updating the preset large model.

[0105] It should be noted that when the evaluation score of the software code generated by the optimized large model is greater than the preset optimization score threshold, it means that the performance of the model has reached a relatively stable level, and further updates to the preset large model can be stopped at this time. It can be understood that this step is an important measure to prevent over-training and optimization from causing degradation of model performance or weakening of generalization ability. For example, when developing a general diagnostic application platform for an on-board domain controller, if the code generated by the system achieves high scores on multiple test indicators (such as an accuracy rate of more than 95%, an increase in execution efficiency of 20%, etc.), and these scores remain stable or slightly improved in multiple consecutive iterations, then the system can assume that the current large model has reached a relatively good performance level. At this point, the system can stop further updates to the preset large model.

[0106] In addition, it should be noted that the preset optimization score threshold is a parameter that is customized and adjusted according to specific development scenarios and requirements. By selecting an appropriate threshold level, you can avoid unnecessary training and optimization costs while ensuring stable model performance.

[0107] In this embodiment, by obtaining user feedback data and iteratively optimizing the large model based on the test indicator scores, the model can gradually adapt to new requirements and software architecture changes, maintain the timeliness and effectiveness of its generated code, ensure the quality of the software code generated by the model, and improve software development efficiency.

[0108] This application also provides a software development device based on a large model, please refer to Figure 6 , the software development device based on the large model includes:

[0109] The data acquisition module 10 is used to acquire software requirement data and prompt word data;

[0110] The data processing module 20 is used to process the software requirement data according to a preset processing strategy to obtain a function module function;

[0111] The software development module 30 is used to generate a target software framework by calling corresponding functional module functions based on the preset large model and the prompt word data to realize the development of the target software.

[0112] The software development device based on a large model provided by the present application adopts the software development method based on a large model in the above-mentioned embodiment, and can solve the technical problem of how to improve the development efficiency of the interface and function of the middleware software. Compared with the prior art, the beneficial effects of the software development device based on a large model provided by the present application are the same as the beneficial effects of the software development method based on a large model provided by the above-mentioned embodiment, and the other technical features in the software development device based on a large model are the same as the features disclosed in the above-mentioned embodiment method, which will not be repeated here.

[0113] In one embodiment, the data processing module 20 is further used to filter the software requirement data according to a preset processing strategy to obtain requirement design data; generate corresponding function module codes according to the requirement design data, and encapsulate the function module codes to obtain function module functions.

[0114] In one embodiment, the data processing module 20 is further used to obtain diagnostic software requirement data and storage software requirement data based on the software requirement data; and filter the diagnostic software requirement data and the storage software requirement data according to a preset processing strategy to obtain requirement design data.

[0115] In one embodiment, the software development module 30 is also used to obtain internal data according to a preset retrieval enhancement strategy; combine the internal data with a preset large model to obtain a target large model; and call corresponding functional module functions according to the prompt word data and the target large model to generate a target software framework.

[0116] In one embodiment, the software development module 30 is further used to parse the prompt word data according to the target large model to obtain the requirement type of the development software; when the requirement type is the first type, the first function function and the first interface function are called to obtain the first function code and the first interface code, and a target software framework is generated according to the first function code and the first interface code; when the requirement type is the second type, the second function function and the second interface function are called to obtain the second function code and the second interface code, and a target software framework is generated according to the second function code and the second interface code.

[0117] In one embodiment, the software development device based on the big model also includes an optimization verification module, and the optimization verification module is used to obtain user feedback data; according to the feedback data, the preset big model is updated to obtain an optimized big model.

[0118] In one embodiment, the optimization verification module is also used to score the software code generated by the optimized large model according to preset test indicators to obtain an evaluation score; when the evaluation score is greater than a preset optimization score threshold, stop updating the preset large model.

[0119] The present application provides a software development device based on a big model, and the software development device based on the big model includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the software development method based on the big model in the above-mentioned embodiment one.

[0120] Reference below Figure 7 , which shows a schematic diagram of the structure of a software development device based on a large model suitable for implementing the embodiment of the present application. The software development device based on the large model in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 7 The large model-based software development device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0121] like Figure 7As shown, the software development device based on the big model may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the software development device based on the big model are also stored. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the software development device based on the big model to communicate with other devices wirelessly or by wire to exchange data. Although the big model-based software development device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have instead.

[0122] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0123] The software development device based on a large model provided by the present application adopts the software development method based on a large model in the above-mentioned embodiment, which can solve the technical problem of how to improve the development efficiency of the interface and function of the middleware software. Compared with the prior art, the beneficial effects of the software development device based on a large model provided by the present application are the same as the beneficial effects of the software development method based on a large model provided by the above-mentioned embodiment, and the other technical features in the software development device based on a large model are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0124] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0125] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0126] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the large model-based software development method in the above-mentioned embodiment.

[0127] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0128] The computer-readable storage medium may be included in a software development device based on a large model; or may exist independently without being assembled into a software development device based on a large model.

[0129] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by a software development device based on a large model, the software development device based on the large model: obtains software requirement data and prompt word data; processes the software requirement data according to a preset processing strategy to obtain a functional module function; and calls the corresponding functional module function based on the preset large model and the prompt word data to generate a target software framework to realize the development of the target software.

[0130] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0131] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0132] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0133] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned software development method based on a large model, and can solve the technical problem of how to improve the development efficiency of the interfaces and functions of the middleware software. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the software development method based on a large model provided in the above-mentioned embodiment, and will not be repeated here.

[0134] The present application also provides a computer program product, including a computer program, which implements the steps of the large model-based software development method as described above when executed by a processor.

[0135] The computer program product provided by this application can solve the technical problem of how to improve the development efficiency of the interface and function of the middleware software. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the software development method based on the large model provided by the above embodiment, which will not be repeated here.

[0136] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A software development method based on a large model, characterized in that: The method comprises: Obtain software requirement data and prompt word data; Processing the software requirement data according to a preset processing strategy to obtain a functional module function; Based on the preset large model and the prompt word data, the corresponding functional module function is called to generate a target software framework to realize the development of the target software.

2. The method according to claim 1, characterized in that The step of calling the corresponding functional module function based on the preset large model and the prompt word data to generate the target software framework includes: Acquire internal data according to preset retrieval enhancement strategies; Combining the internal data with the preset macro model to obtain a target macro model; The target software framework is generated by calling the corresponding functional module function according to the prompt word data and the target large model.

3. The method according to claim 2, characterized in that The functional module function includes a first interface function, a second interface function, a first functional function and a second functional function. The step of calling the corresponding functional module function according to the prompt word data and the target large model to generate a target software framework includes: Parsing the prompt word data according to the target macro model to obtain the demand type for developing software; When the requirement type is the first type, calling the first function function and the first interface function to obtain a first function code and a first interface code, and generating a target software framework according to the first function code and the first interface code; When the requirement type is the second type, the second function function and the second interface function are called to obtain a second function code and a second interface code, and a target software framework is generated according to the second function code and the second interface code.

4. The method according to claim 1, characterized in that The step of processing the software requirement data according to a preset processing strategy to obtain a function module function comprises: Filtering the software requirement data according to a preset processing strategy to obtain requirement design data; Generate corresponding functional module codes according to the required design data, and encapsulate the functional module codes to obtain functional module functions.

5. The method according to claim 4, characterized in that The step of filtering the software requirement data according to the preset processing strategy to obtain the requirement design data comprises: Obtaining diagnostic software requirement data and storage software requirement data according to the software requirement data; The diagnostic software requirement data and the storage software requirement data are filtered according to a preset processing strategy to obtain requirement design data.

6. The method according to claim 1, characterized in that After the step of generating a target software framework based on the preset large model and the prompt word data and calling the corresponding function module function to realize the development of the target software, the method further includes: Obtain user feedback data; The preset large model is updated according to the feedback data to obtain an optimized large model.

7. The method according to claim 6, characterized in that After the step of updating the preset large model according to the feedback data to obtain the optimized large model, the method further includes: Scoring the software code generated by the optimized large model according to preset test indicators to obtain an evaluation score; When the evaluation score is greater than a preset optimization score threshold, the updating of the preset large model is stopped.

8. A software development device based on a large model, characterized in that: The device comprises: A data acquisition module is used to acquire software requirement data and prompt word data; A data processing module, used to process the software requirement data according to a preset processing strategy to obtain a function module function; The software development module is used to generate a target software framework by calling corresponding functional module functions based on a preset large model and the prompt word data to realize the development of the target software.

9. A software development device based on a large model, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the large model-based software development method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the large model-based software development method as described in any one of claims 1 to 7 are implemented.

Images