Control method and device of delivery process and storage medium

Abstract

The present disclosure provides a control method and device of a delivery process and a storage medium, relates to the field of artificial intelligence, in particular to the technical field of large models, NLP, intelligent assistants and the like. The specific implementation scheme is: performing intent recognition according to input content to obtain a target stage that needs to be executed in the delivery process; collecting parameter information required by the target stage; triggering a target task corresponding to the target stage; calling a processing platform of the target task to execute the target task according to the parameter information required by the target stage.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of artificial intelligence, and in particular, to the technical field of large models, natural language processing (NLP), intelligent assistants, etc. BACKGROUND

[0002] Current project development delivery is mainly based on automated processes, converting human-driven delivery processes into machine execution, with software tools executing pre-set delivery processes. However, overly consistent delivery processes limit the personalization of each stage and service. Due to the concatenation of multiple delivery processes, a single delivery may need to be processed across multiple platforms. SUMMARY

[0003] The present disclosure provides a control method, device and storage medium of a delivery process.

[0004] According to an aspect of the present disclosure, a control method of a delivery process is provided, comprising:

[0005] performing intent recognition according to input content to obtain a target stage that needs to be executed in the delivery process;

[0006] collecting parameter information required by the target stage;

[0007] triggering a target task corresponding to the target stage;

[0008] calling a processing platform of the target task to execute the target task according to the parameter information required by the target stage.

[0009] According to another aspect of the present disclosure, a control device of a delivery process is provided, comprising:

[0010] a recognition module configured to perform intent recognition according to input content to obtain a target stage that needs to be executed in the delivery process;

[0011] a collection module configured to collect parameter information required by the target stage;

[0012] a triggering module configured to trigger a target task corresponding to the target stage;

[0013] a calling module configured to call a processing platform of the target task to execute the target task according to the parameter information required by the target stage.

[0014] According to another aspect of the present disclosure, an electronic device is provided, comprising:

[0015] at least one processor; and

[0016] a memory in communication with the at least one processor; wherein

[0017] The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method of any of the embodiments of the present disclosure.

[0018] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, where the computer instructions are used to enable the computer to perform the method according to any of the embodiments of the present disclosure.

[0019] According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the method according to any of the embodiments of the present disclosure.

[0020] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present disclosure, nor to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0021] The accompanying drawings are used to better understand the present scheme and do not limit the present disclosure. Among them:

[0022] Figure 1 is a flowchart of a control method of a delivery process according to an embodiment of the present disclosure;

[0023] Figure 2 is a flowchart of a control method of a delivery process according to another embodiment of the present disclosure;

[0024] Figure 3 is a flowchart of a control method of a delivery process according to another embodiment of the present disclosure;

[0025] Figure 4 is a flowchart of a control method of a delivery process according to another embodiment of the present disclosure;

[0026] Figure 5 is a schematic diagram of process management capability of an embodiment of the present disclosure;

[0027] Figure 6 is an example diagram of delivery process reply scripts;

[0028] Figure 7 is a schematic diagram of determining task status in task scheduling;

[0029] Figure 8 is a schematic diagram of giving repair suggestions by a large language model;

[0030] Figure 9 is a structural schematic diagram of a control device of a delivery process according to an embodiment of the present disclosure;

[0031] Figure 10 is a structural diagram of a control device for a delivery process according to another embodiment of the present disclosure;

[0032] Figure 11 It is a block diagram of an electronic device used to implement the control method of the delivery process of an embodiment of the present disclosure. DETAILED DESCRIPTION

[0033] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be appreciated by those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0034] Automated processes for project R&D delivery can shorten delivery cycles, improve test environment coverage during delivery, ensure consistent test results, successfully address the variability and inaccuracies introduced by manual operations, and reduce project delivery costs. However, overly uniform delivery processes limit the customization of each stage and service. Furthermore, due to the interconnected nature of multiple delivery processes, a single delivery may require processing across multiple platforms, such as the delivery process management platform, quality visualization platform, testing platform, launch platform, and monitoring platform. Such processes are complex, rigid, and provide a poor user experience.

[0035] The project R&D delivery process primarily relies on an assembly line, which links the different stages of the delivery process together to form a continuous delivery process. Each stage completes a specific task and, upon completion, passes the results to the next stage. These stages are numerous and span multiple platforms. This type of assembly line is typically set up by testers, while R&D personnel lack a comprehensive understanding of the entire process. Consequently, this approach is complex to operate, has a high learning curve, and features relatively fixed process tasks. Furthermore, R&D personnel's lack of familiarity with the process limits their ability to identify issues when they arise.

[0036] The pipeline delivery process has one or more of the following problems:

[0037] 1. Complex operations and high learning curve: The pipeline model involves multiple stages across multiple platforms, each requiring tester setup. This results in high costs and complex operations across multiple platforms. Furthermore, when developing multiple projects simultaneously, each at different stages, switching between projects and managing processes can easily lead to operational errors.

[0038] 2. Rigid delivery tasks: The delivery process usually adopts standardized tasks, which may cause waste of resources and inefficiency for projects that require specific or personalized delivery requirements. The lack of ability to flexibly adjust delivery tasks to meet project-specific needs limits the adaptability and effectiveness of the delivery process.

[0039] 3. Lack of problem positioning ability: When problems occur in the pipeline mode, for the R&D personnel: lack of understanding of the pipeline, do not know where the problem is and how to solve it, can only ask for help from the test personnel; for the test personnel: the test capabilities of each stage of the pipeline are relatively independent, and the responsible persons are also different, the problem positioning process is complex and the cost is high.

[0040] Although the pipeline model provides a certain structure and continuity, it has limitations in reducing learning costs, convenient operation, flexible delivery process, and problem positioning ability, which need to be further optimized and improved.

[0041] Figure 1 is a flowchart of a control method 100 of a delivery process according to an embodiment of the present disclosure, which can include:

[0042] S110, performing intent recognition according to the input content to obtain a target stage that needs to be executed in the delivery process;

[0043] S120, collecting parameter information required by the target stage;

[0044] S130, triggering a target task corresponding to the target stage;

[0045] S140, calling a processing platform of the target task, and executing the target task according to the parameter information required by the target stage.

[0046] In the embodiments of the present disclosure, the delivery process can usually include multiple process stages. For example, the delivery process of a project can include one or more process stages such as compilation, pre-admission, admission, post-admission, test submission, release check, release, and online. Projects can include various research and development projects such as software development, system integration, and hardware research and development. The delivery process can need to be processed across multiple platforms. For example, one or more processing platforms such as a delivery process management platform, a platform quality visualization platform, a test submission platform, an online platform, and a monitoring platform. These processing platforms can be service platforms outside the process management system. The input content is determined according to the target stage to be executed. The process management system can be referred to as a project delivery assistant system, an intelligent delivery assistant system, etc.

[0047] In the embodiments of the present disclosure, which process stages of the execution delivery process are controlled according to input content. The input content can be text, voice, etc. input by a user such as a test personnel to a front end of a process management system. The front end can include an intelligent assistant, a service number, etc. The input content can include content related to process stages that need to be executed, such as “enter the compilation stage”, “execute the admission stage”, etc. The input content can also include execution commands to the front end, such as “construct the admission task”, “select the release task”, etc. The input content can also include queries to the process task stage, such as “query the progress of the test submission stage”, “is the release check stage completed?”. The input content can also include modifications to the task stage parameters, such as “modify the work order of task A in the admission stage”, “resubmit the form of task B in the compilation stage”, etc.

[0048] In the embodiments of the present disclosure, intent recognition is performed on the input content, and the target stage that needs to be executed in the current delivery process can be obtained. For example, intent recognition is performed on “enter the compilation stage”, and it is recognized that the target stage is the compilation stage. For another example, intent recognition is performed on “construct the admission task”, and it is recognized that the target stage is the admission stage.

[0049] In the embodiments of the present disclosure, various information related to the task corresponding to the execution process stage can be saved to a database. For example, project information, the name of the process stage, various parameters used, execution results, etc. are stored in the database. After the target stage is determined according to the input content, the parameter information required by the target stage can be collected in various ways. For example, various parameters that have been used by the target stage are searched in the database according to the name of the target stage. And these parameters are recommended to the user through the interactive interface. The user can select, delete, modify, etc. the recommended parameters, and upload them to the process management system. In this way, the project management system can collect the parameter information required by the target stage.

[0050] In the embodiments of the present disclosure, the user can dialogue with the process management system through the interactive interface. The process management system can obtain historical dialogue information of the input content, and extract relevant parameters of the target stage from the historical dialogue information. These parameters can be part of the parameter information required by the target stage. According to these parameters, part of the parameter information required by the target stage can also be searched from the database.

[0051] In the embodiments of the present disclosure, the target task corresponding to the target stage can be triggered automatically, and the collected parameter information required by the target stage can be sent to the processing platform of the target task. The processing platform can call the execution of the target task based on the collected parameter information. For example, after triggering the test submission task, the test submission platform can be called to execute the test submission task according to the parameters of the test submission task. For another example, after triggering the admission task, the admission platform can be called to execute the admission task according to the parameters of the test submission task.

[0052] According to the disclosed embodiments, after acquiring the target phase through intent recognition and collecting the parameter information of the target phase, the corresponding target task can be automatically triggered and the target task processing platform can be called to execute the target task. This can optimize the processing efficiency of the delivery process and further enhance the user experience.

[0053] Figure 2 This is a flow chart of a delivery process control method 200 according to another embodiment of the present disclosure. This method 200 can be used to implement step S110 in the delivery process control method 100. In one embodiment, the method 200 includes: performing intent recognition based on input content to obtain a target stage to be executed in the delivery process, and further includes:

[0054] S210: Perform natural language analysis on historical conversation information related to the input content to extract historical intent, and pre-fill parameter information of the process stage corresponding to the historical intent;

[0055] S220: Use the large model to understand the intent of the input content to extract the current intent and obtain the target stage that needs to be executed corresponding to the current intent.

[0056] In the disclosed embodiment, the intention layer of the process management system may include functions such as historical conversation analysis and large model analysis.

[0057] In an embodiment of the present disclosure, historical conversation information can be identified based on natural language processing (NLP) technology, and historical intent can be extracted from the historical conversation information of the user interacting with the system in the interactive interface. The historical intent may correspond to one or more process stages in the delivery process. For the parameters of these process stages, parameter information can be collected using a database or recommendation. For example, if the historical conversation information includes a conversation in which the submission of stage 1 and stage 2 is completed, it can be identified that stage 1 and stage 2 have been executed. The historical parameters of stage 1 and stage 2 can be searched from the database, and these historical parameters can be used to pre-fill the parameters of stage 1 and stage 2.

[0058] In an embodiment of the present disclosure, the input content may include current session information. The current session information is input into the large model, the current intent in the current session information is identified, and the target stage corresponding to the current intent is determined. For example, if the target stage includes stage 2, the parameters pre-filled in stage 2 can be used to perform the target task corresponding to the target stage. The large model can also identify some parameters of the target stage in the current session information. These parameters can also be used to perform the target task corresponding to the target stage. The pre-filled parameters and the parameters identified by the large model can be recommended to the user through the interactive interface, and some parameters of the target stage can be obtained based on the user's interactive operation. These parameters can also be used to perform the target task corresponding to the target stage.

[0059] According to the embodiments of the present disclosure, through historical conversation analysis and / or large model analysis, the intention of the conversation can be better understood, the accuracy of identifying the intention can be improved, and the target stage that needs to be executed can be accurately determined.

[0060] In one embodiment, the parameter information required for the target stage includes one or more of the following:

[0061] Parameters pre-populated based on the recognition results of the historical conversation information;

[0062] Pre-populated parameters based on the recognition results of the input content;

[0063] The interaction parameters inputted into the interaction interface in response to the interaction command of the target stage.

[0064] In the disclosed embodiments, the parameters required for different process stages may be different. When executing the target task corresponding to the target stage, many parameters need to be obtained. The parameters can come from various sources. For example, step S210 pre-fills the parameters based on historical conversation information. For another example, step S220 pre-fills the parameters based on the recognition results of the model. For another example, the intelligent delivery assistant of the process management system enters the interactive parameters in the front-end interactive interface according to the interactive commands at this process stage.

[0065] According to the embodiments of the present disclosure, personalized parameter recommendations can be made based on pre-populated parameters, reducing the number of parameters that require user input and improving the efficiency of the automatic execution of the delivery process. By obtaining parameter information required for the target stage from different sources, more comprehensive parameter information required for the target stage can be automatically obtained.

[0066] In one embodiment, collecting parameter information required for the target stage includes one or more of the following:

[0067] Displaying recommended parameters for the target stage based on pre-populated parameters for the target stage, and obtaining parameter information required for the target stage in response to a modification command and / or a confirmation command for the recommended parameters;

[0068] In response to the interaction parameters of the target stage, parameter information required for the target stage is obtained.

[0069] In an embodiment of the present disclosure, based on the pre-filled parameters of the target stage, the recommended parameters of the target stage can be displayed on the interactive interface. If the user modifies the recommended parameters, a modification command can be generated. If the user confirms the recommended parameters, a confirmation command can be generated. The modified or confirmed recommended parameters can be used as part of the parameter information required for the target stage. For example, "Please confirm whether to use the parameters P of stage 2" and "Please select the parameters P1, P2 and P3 of stage 2". The user can choose to confirm the use of the parameters P of stage 2. The user can also select the parameters P1 and P3 of stage 2.

[0070] In the disclosed embodiment, the interactive interface displays parameter input items for the target stage. If the user enters interaction parameters in the parameter input items, these interaction parameters can be collected as part of the parameter information required for the target stage. For example, the interactive interface displays "Please enter the parameters required for stage 3."

[0071] According to the embodiments of the present disclosure, information of the target stage can be collected more conveniently through recommendation parameters, and more abundant and demand-oriented information of the target stage can be collected through interaction parameters.

[0072] Figure 3 FIG. 3 is a flow chart of a method 300 for controlling a delivery process according to another embodiment of the present disclosure. The method may include one or more features of the above-mentioned method embodiments. In one embodiment, the method further includes:

[0073] S310: Perform parameter verification on the parameter information required for the target stage.

[0074] In the disclosed embodiment, it is necessary to verify the various parameters required for the target stage, such as necessary parameters and optional parameters. For example, the intelligent delivery assistant of the process management system can verify the parameters pre-filled based on historical conversation information in step S210. For another example, the parameters pre-filled based on the model recognition results in step S220 can also be verified. For another example, the interaction parameters obtained from the interaction interface in this process stage can also be verified. The number of parameter verifications can be divided into multiple times, or a unified verification can be performed once. The success rate of the task of executing the delivery process can be improved through parameter verification.

[0075] In one embodiment, triggering the target task corresponding to the target stage includes:

[0076] S320: In response to the parameter verification result of the target stage passing, trigger the target task corresponding to the target stage.

[0077] In the disclosed embodiment, the parameter verification result of the target stage may include pass or fail. It is possible to verify whether the parameters are legal, whether there are errors, whether the necessary parameters are complete, etc. If the parameters are illegal, erroneous or incomplete, the parameter verification result may be fail. If the parameter verification fails, the intelligent delivery assistant can generate a reply content about the parameter verification failure in the interactive interface, and can also provide recommended parameters or parameter collection instructions, etc. If the parameter verification passes, the intelligent delivery assistant can trigger the target task corresponding to the target stage. By automatically triggering the target tasks corresponding to various target stages, the tasks of each process stage of the delivery process can be flexibly executed, thereby improving the efficiency of task execution.

[0078] In one embodiment, the method 300 further includes: S330, checking whether the target stage has one or more preceding stages that fail to meet set conditions.

[0079] In the embodiment of the present disclosure, some stages may not be completed before the target stage determined based on the input content. If the task execution result of a certain stage does not pass the set conditions, then the stage is not completed. If the task execution result of a certain stage passes the set conditions, then the stage is completed. The set conditions can be set according to the characteristics of the process stage. For example, the compilation result setting conditions corresponding to the compilation stage. The admission result setting conditions corresponding to the admission stage. The other stages are similar.

[0080] If there are unfinished prerequisites, the corresponding prerequisite tasks can be executed first, followed by the target tasks for the target phase. For example, the Intelligent Delivery Assistant checks whether the compilation phase preceding the approval phase is complete. Another example is the Intelligent Delivery Assistant checks whether the compilation phase and the approval phase preceding the test submission phase are complete. By checking whether the prerequisites for the target phase are complete, tasks corresponding to multiple unfinished process phases can be automatically executed, thereby improving the completion efficiency and pass rate of the delivery process.

[0081] In one embodiment, collecting parameter information required for the target stage includes:

[0082] In the case that the one or more preceding stages exist, parameter information required by the one or more preceding stages and the target stage is collected.

[0083] In the disclosed embodiments, parameter information for a specific pre-stage can be collected first, and the pre-tasks corresponding to that pre-stage can be executed. After the pre-stage is completed, parameter information for the next stage can be collected, and the tasks corresponding to that next stage can be executed until the target stage is completed. By collecting parameter information required for one or more pre-stages and the target stage, the target tasks corresponding to multiple unfinished pre-stages and target stages can be automatically executed, thereby improving the completion efficiency of the delivery process and increasing the delivery process's pass rate.

[0084] In one embodiment, triggering the target task corresponding to the target stage includes:

[0085] In response to parameter verification results of the target stage and the one or more preceding stages being passed, a target task corresponding to the target stage and preceding tasks corresponding to the one or more preceding stages are triggered.

[0086] In an embodiment of the present disclosure, after collecting parameter information for a certain preceding stage and passing parameter verification, the preceding task corresponding to the preceding stage can be triggered and executed. After the preceding stage is completed, parameter information for the next stage can be collected. After the parameter verification passes, the task corresponding to the next stage can be triggered and executed until the target stage is completed. By collecting parameter information required for one or more preceding stages and the target stage, target tasks corresponding to multiple unfinished preceding stages and the target stage can be automatically executed, thereby improving the completion efficiency of the delivery process and the pass rate of the delivery process.

[0087] Figure 4 FIG. 4 is a flow chart of a method 400 for controlling a delivery process according to another embodiment of the present disclosure. The method may include one or more features of the above-mentioned method embodiments. In one embodiment, the method 400 further includes:

[0088] S410: Send tasks corresponding to one or more process stages to a task scheduling system for hosting.

[0089] In one embodiment, the tasks corresponding to the one or more process stages include the target task corresponding to the target stage and / or the predecessor tasks corresponding to one or more predecessor stages of the target stage.

[0090] In the disclosed embodiment, the task scheduling system can schedule the tasks that need to be executed in the delivery process. If it is necessary to execute a process stage, such as the target task corresponding to the target stage, the task scheduling system can call the processing platform of the target task corresponding to the target stage to execute the target task corresponding to the target stage. If it is necessary to execute multiple process stages, such as the tasks corresponding to the pre-stage and the target stage. The task scheduling system can call the processing platform of the pre-task corresponding to the pre-stage to execute the pre-task corresponding to the pre-stage. After the pre-stage is completed, the task scheduling system can call the processing platform of the target task corresponding to the target stage to execute the target task corresponding to the target stage. Until the target stage is completed. By hosting the tasks corresponding to each stage through the task scheduling system, cross-platform scheduling can be automatically performed, the operation is simple, the switching between different development stages of the project is convenient, and the scheduling accuracy is high.

[0091] In one embodiment, the method 400 further includes:

[0092] S420, saving information of the managed task through the task status storage module of the task scheduling system;

[0093] S430: Control the execution process of the tasks corresponding to the one or more process stages through the task scheduling service of the task scheduling system.

[0094] In an embodiment of the present disclosure, the task scheduling system may include a task status storage module and a task scheduling service. The task status storage module may store information about tasks hosted by the intelligent delivery assistant. For example, basic information about the current task execution, execution parameters, task execution status, current execution stage, target stage, etc., may be stored. Furthermore, the collected information about these tasks may be used by the intent layer. Saving the hosted tasks through the task scheduling system and automatically controlling the execution process of the hosted tasks through cross-platform scheduling is not only simple to operate but also highly efficient.

[0095] In one embodiment, controlling the execution of the tasks corresponding to the one or more process stages through the task scheduling service of the task scheduling system includes:

[0096] The task scheduling service is used to execute multiple managed tasks in the order of process stages.

[0097] According to the current stage of the current task, the processing platform corresponding to the current stage is called to execute the current task.

[0098] In an embodiment of the present disclosure, if the intelligent delivery assistant hosts tasks corresponding to multiple process stages to the task scheduling system, the task scheduling service of the task scheduling system can execute these tasks in sequence according to the order of the process stages. The tasks of the next stage can be executed after the tasks of one stage are completed. For tasks that do not have sequence requirements, they can also be executed in parallel or in the order of hosting. According to the current stage of the task currently scheduled by the task scheduling service (referred to as the current task), the corresponding processing platform can be called to execute the current task. By saving multiple hosted tasks through the task scheduling system, the processing platform of each task can be scheduled separately, and the execution process of the multiple hosted tasks can be automatically controlled, which is not only simple to operate but also highly efficient.

[0099] In one embodiment, the method 400 further includes one or more of the following steps:

[0100] If the current stage is not the target stage and the current stage meets the set conditions, the task corresponding to the next process stage is triggered to execute;

[0101] If the current stage is the target stage and the current stage passes the set conditions, a reply content of the delivery result is generated.

[0102] If the status of the current task is execution failure, the current task is terminated and the problem is located.

[0103] In the disclosed embodiment, the task scheduling service can execute tasks sequentially according to the process stage that the current task is in. If there are no problems during the execution process, the task can be scheduled until the target stage is completed. If there are problems during the execution process, the problem can be located.

[0104] For example, managed tasks include access tasks, testing tasks, and release tasks. You can first execute the access task and determine whether it is in the target phase (assuming the target phase is the release phase). If the access task is not in the target phase, call the access platform to execute the access task and then check whether the access phase is complete based on the execution result. If the access phase is complete, you can execute the next testing task and determine whether it is in the target phase. If the testing task is not in the target phase, call the testing platform to execute the testing task and then check whether the testing phase is complete based on the execution result. If the testing phase is complete, you can execute the next release task and determine whether it is in the target phase. If the release task is in the target phase, call the release platform to execute the release task and then check whether the release phase is complete based on the execution result. If the release phase is complete, this task scheduling is complete. If a problem occurs in any of these phases, the current task in that phase can be terminated and the problem can be located.

[0105] By scheduling the processing platforms of multiple hosted tasks in sequence through the task scheduling system, multiple tasks can be efficiently scheduled, which is not only simple to operate but also highly efficient.

[0106] In one embodiment, problem location includes:

[0107] Input the parameter information and execution data of the current task into the big model to determine the problem type of the current task; wherein the problem type includes one or more of the following: code problem, self-operation and maintenance tool problem, and expected failure problem;

[0108] The large model is used to output failure cause analysis and / or processing suggestions for the current task based on the problem type.

[0109] In an embodiment of the present disclosure, if an exception occurs during the execution of the current task, part or all of the parameter information of the current task, as well as execution data such as the execution process and / or execution results, can be collected. The collected content is input into the big model to locate the problem. The big model can be a big model that is fine-tuned based on the characteristics of the project delivery process management on the basis of the benchmark big model. The fine-tuned samples can be determined according to the type of problem that needs to be located. For example, code problem samples, self-operation and maintenance tool problem samples, and expected failure problem samples, etc. Through the big model, not only can the problem type of the current task be located, but also accurate failure causes and / or handling suggestions can be given for the problem type, which improves the problem location capability, facilitates the rapid resolution of problems, and thereby improves the processing speed of the delivery process.

[0110] The control method of the delivery process of the embodiment of the present disclosure includes a method of combining natural language processing technology with project R&D delivery process management, which can enable intelligent delivery using large models. The core role of this method includes an intelligent delivery assistant. This assistant can accurately analyze the intentions expressed in natural language by R&D, analyze and process the requirements and code changes to be delivered by R&D, control the entire testing process, and at the same time understand the problems encountered by users during the delivery process and provide timely assistance. Using this assistant will make the work of the R&D team more convenient. The R&D team only needs to have a concise conversation with this intelligent assistant to promote the delivery process, generate customized delivery tasks, and arrange delivery tasks and automatically execute them. With such assistance, the R&D team can focus more on R&D work.

[0111] Projects in the disclosed embodiments can include various R&D projects, including software development, system integration, and hardware R&D. Each project can have its own Project Delivery Assistant system controlling the delivery process, or multiple projects can use the same Project Delivery Assistant system. In the disclosed embodiments, multiple platforms within an enterprise's internal or outsourced R&D projects can be integrated.

[0112] See also Figure 5 The process management of the embodiment of the present disclosure mainly includes the following capabilities: intent recognition, process management interaction, task scheduling system, and problem location system:

[0113] 1. Intent Recognition: The main function of intent recognition is to analyze the intention expressed in natural language by R&D personnel and then execute the corresponding delivery phase (process phase). Figure 5 ,The intent layer mainly includes functions such as historical conversation ,analysis and large model analysis.

[0114] (1) Historical conversation analysis: During each interaction, the R&D team will first analyze the historical conversation information, taking the user's natural language expressions and recognized intentions in the recent period as well as the parameter information of the user's execution process stage for pre-filling.

[0115] (2) Big model analysis: After the historical conversation is filled, the intention is understood and the parameters of the conversation are extracted using the big model, and then transferred to the process that R&D wants to deliver for further processing.

[0116] 2. Process Management System: The main function of this part is to perform some pre-processing and stage triggering for this delivery request. It includes the following steps: parameter verification, pre-task check, information collection, task triggering, and splicing and generating reply scripts. Figure 5 .

[0117] (1) Parameter Verification: First, during a conversation, the intent layer performs a first round of parameter extraction, primarily based on the user's query and historical conversation information. If the intent layer fails to extract the necessary process execution information, multiple rounds of interactive dialogue with R&D can be conducted, using methods such as forms to accurately extract the necessary parameter information.

[0118] (2) Pre-task check: Before the current process (target stage) is advanced, a pre-task check will be conducted to ensure that the conditions for advancing the current process are met. For example, before submitting the test, it is necessary to determine whether the access has been passed, and before releasing the test, it is necessary to confirm whether the test has been passed. Only when the conditions are met can the current process be officially entered.

[0119] (3) Information Collection: After the pre-task check, the agent will call the personalized recommendation module to recommend parameters for the execution of this task (the current task). Based on the analysis of the code, the recommended parameters for this task are submitted to the user for confirmation and modification. After confirmation, the process is officially executed using these parameters.

[0120] (4) Task triggering: After the parameter verification is passed, the assistant will trigger the task and hand it over to the task scheduling system for hosting.

[0121] (5) Response script: Each time the conversation assistant is in progress, it will personalize the script based on the execution result of this process, guide R&D to provide information, correct information, synchronize delivery progress, and guide the completion of the entire delivery process.

[0122] See also Figure 6 This example shows how responses are generated based on input conversations. This allows for flexible input and response generation for each stage of the delivery process. For example, if the conversation is "Help me construct a query, modify the vocabulary, module: XXX, configuration file link XXX," the intent is to construct a query, and a response can be generated that reads, "Hey! The vocabulary construction task has begun! Please stay tuned for push notifications." For another example, if the conversation is "I want access," the access platform can be called to execute performance, automation, and stability tasks, and then a response can be generated that reads, "We've identified performance and stability risks in the code you submitted. These tasks, automation, and stability tasks are required. We've recommended tasks and parameters for you. Please confirm your submission and the assistant will automatically execute them." Flexible input and response generation is also possible for stages like testing, release, registration, and launch. The specific content of the input conversations and responses is not limited and can be adjusted flexibly based on needs.

[0123] 3. Task Scheduling System: If the triggering phase involves multiple preceding phases, how can we ensure that the specified phase can be reached and that the specified phase can be run with the specified parameters? For example, if the triggering phase is the execution access phase, but the access phase is preceded by the compilation phase and the preceding task check phase, task scheduling services are essential. The disclosed embodiment uses a task state storage module + task scheduling service approach.

[0124] (1) Task status storage module: This module stores the task information managed by the assistant. This module contains basic information about the current task execution, execution parameters, task execution status, current execution stage, target stage, etc. This module also collects this task execution information for use by the intent layer. Tasks at different stages may require different external service platforms, such as a delivery process management platform (or assembly line platform), a quality visualization platform, a testing platform, an online platform, a monitoring platform, etc. for processing.

[0125] (2) Task scheduling service: manage all tasks in execution, determine whether the task is executed to the target stage. If the task is executed to the target stage, the task state is determined and the assistant information is pushed. If the task is not executed to the target stage, the task state is determined, the operation in the execution state is not needed, the problem positioning is performed and the push is performed through the assistant, and the state of the execution stage can be triggered. When the execution time is too long, the task will be killed and the problem positioning will be performed.

[0126] Referring to Figure 7 , all task states in the task execution table are obtained, and the data indicating the to-be-executed task such as task_status=1 is obtained. The task state of the to-be-executed task is determined, if the determination is timed out, the problem positioning is performed, and the task_status is set to a value indicating the timeout problem such as 3. If the determination is not timed out, the state of the task pipeline is obtained.

[0127] When the task pipeline is not running to the target stage, the task state is determined. For example, the task state generally has running (RUNNING), failure (FAIL), waiting for user (PENDING FOR USER), cancellation (CANCEL), and the like. When the task state is running, no operation is needed, and the task is waited to run; when the task state is failure, the problem positioning is performed, and the task_status is set to a value indicating the failure problem such as 2; when the task state is waiting for user and cancellation, the task triggering is performed.

[0128] When the task pipeline is running to the target stage, the task state is determined, and the task state generally has running (RUNNING), failure (FAIL), waiting for user (PENDING FOR USER), cancellation (CANCEL), success (SUCC), and the like. When the task state is success, the task_status is set to a value indicating the success such as 0; when the task state is running, no operation is needed, and the task is waited to run; when the task state is failure, the problem positioning is performed, and the task_status is set to a value indicating the failure problem such as 2; when the task state is waiting for user and cancellation, the task triggering is performed.

[0129] 4, problem positioning system: the problem positioning plays a crucial role in the perception of code risks and the dredging of process bottlenecks in the entire delivery process. For example, the current problems mainly exist in three aspects: code problems, self-operation and maintenance tool problems, and expected failure problems. These problems may be inevitable and very common for research and development (RD) personnel. Most of the task exception cause positioning logic is relatively fixed, and the repetitive labor consumes manpower, interrupts the work, and affects the overall efficiency. Referring to Figure 8, the assistant can give reliable positioning and repair suggestions for one or more types of problems. The task problem type can be determined first, and task logs, task labels, error codes, and other information can be obtained. Then, these information is sent to the client of the R&D, tool manager or positioning manager, and the organization's reply tactics can be displayed on the interactive interface of the client.

[0130] (1) Code problem: For process task failure caused by code problem, the problem code is sliced to obtain the problem code segment. The problem code segment and error are analyzed by the large model. The large model gives the reason analysis and repair suggestion.

[0131] (2) Self-operation and maintenance tool: For self-operation and maintenance tool problems in the project delivery process, error information is entered into the problem positioning library when an error occurs in each test task and test tool. When an error occurs, the content in the problem positioning library is read and the error information is extracted.

[0132] (3) Result index fluctuation: For the expected delivery result index fluctuation, the assistant will expose the fluctuation information to the R&D personnel for judgment and analysis. If it meets the expectation, it can continue to pass, and if it does not meet the expectation, it can be repaired and retested.

[0133] The delivery process control scheme of the embodiments of the present disclosure combines natural language processing technology with project R&D delivery process management, which can pass through all project delivery processes without bottlenecks. It can be adapted to special scenarios (such as multi-service access, real-time parameter inspection before access, etc.), continuously optimize and upgrade the delivery process, and optimize user experience. Problem positioning can accurately locate code problems, self-operation and maintenance problems, and result index fluctuation problems, and give problem analysis and repair suggestions combined with the large language model.

[0134] In some examples, using the scheme of the embodiments of the present disclosure, the learning cost of new people in the delivery process can be reduced from 7 days to 1 day. The total time consumption of the delivery process can be reduced by 17%. Using the assistant for delivery can quickly increase the online projects.

[0135] Figure 9 is a structural schematic diagram of a delivery process control device 900 according to an embodiment of the present disclosure. The device 900 can include:

[0136] The recognition module 901 is configured to perform intent recognition according to the input content to obtain a target stage that needs to be executed in the delivery process.

[0137] The collection module 902 is configured to collect parameter information required by the target stage.

[0138] The triggering module 903 is configured to trigger a target task corresponding to the target stage.

[0139] The calling module 904 is used to call the processing platform of the target task and execute the target task according to the parameter information required by the target stage.

[0140] Figure 10 This is a schematic diagram of a delivery process control device 1000 according to another embodiment of the present disclosure. The device 1000 includes: an identification module 1001, a collection module 1002, a trigger module 1003, and a call module 1004. The functions of the above modules can refer to the functions of the modules of the delivery process control device in the above embodiment. In one embodiment, the identification module 1001 may include:

[0141] The language analysis submodule 10011 is used to perform natural language analysis on the historical conversation information related to the input content to extract the historical intent and pre-fill the parameter information of the process stage corresponding to the historical intent;

[0142] The intention understanding submodule 10012 is used to use the large model to understand the intention of the input content, so as to extract the current intention and obtain the target stage that needs to be executed corresponding to the current intention.

[0143] In one embodiment, the parameter information required for the target stage includes one or more of the following:

[0144] Parameters pre-populated based on the recognition results of the historical conversation information;

[0145] Pre-populated parameters based on the recognition results of the input content;

[0146] The interaction parameters inputted into the interaction interface in response to the interaction command of the target stage.

[0147] In one embodiment, collecting parameter information required for the target stage includes one or more of the following:

[0148] Displaying recommended parameters for the target stage based on pre-populated parameters for the target stage, and obtaining parameter information required for the target stage in response to a modification command and / or a confirmation command for the recommended parameters;

[0149] In response to the interaction parameters of the target stage, parameter information required for the target stage is obtained.

[0150] In one embodiment, the device further comprises:

[0151] The verification module 1005 is used to perform parameter verification on the parameter information required for the target stage.

[0152] In an embodiment, the triggering module 1003 is configured to trigger the target task corresponding to the target stage in response to the parameter verification pass result of the target stage.

[0153] In an embodiment, the apparatus further comprises:

[0154] The checking module 1006 is configured to check whether the target stage has one or more preceding stages that fail to meet the set conditions.

[0155] In an embodiment, the collecting module 1002 is configured to collect the parameter information required by the one or more preceding stages and the target stage in the presence of the one or more preceding stages.

[0156] In an embodiment, the triggering module 1003 is configured to trigger the target task corresponding to the target stage and the preceding task corresponding to the one or more preceding stages in response to the parameter verification pass result of the target stage and the one or more preceding stages.

[0157] In an embodiment, the apparatus further comprises:

[0158] The sending module 1007 is configured to send the task corresponding to one or more process stages to a task scheduling system for hosting; wherein the task corresponding to the one or more process stages comprises the target task corresponding to the target stage and / or the preceding task corresponding to the one or more preceding stages of the target stage.

[0159] In an embodiment, the apparatus further comprises:

[0160] The saving module 1008 is configured to save the information of the hosted task by a task state storage module of the task scheduling system.

[0161] The control module 1009 is configured to control the execution process of the task corresponding to the one or more process stages by a task scheduling service of the task scheduling system.

[0162] In an embodiment, the control module 1009 comprises:

[0163] The execution submodule 10091 is configured to execute the hosted tasks in the order of the process stages by the task scheduling service.

[0164] The calling submodule 10092 is configured to call a processing platform corresponding to a current stage to execute a current task according to the current stage in which the current task is located.

[0165] In an embodiment, the modules of the apparatus further have one or more of the following functions:

[0166] The trigger module 1003 is further configured to trigger execution of a task corresponding to a next process stage in a case where the current stage is not the target stage and the current stage passes the set condition.

[0167] The generation module 1010 is configured to generate reply content of a delivery result in a case where the current stage is the target stage and the current stage passes the set condition.

[0168] The problem positioning module 1011 is configured to terminate the current task and perform problem positioning in a case where a state of the current task is execution failure.

[0169] In an embodiment, the problem positioning module 1011 is configured to input parameter information and execution data of the current task into a large model to determine a problem type of the current task, wherein the problem type includes one or more of the following: a code problem, a self-operations and maintenance tool problem, and an expected internal failure problem; and the large model is used to output failure cause analysis and / or processing suggestions of the current task according to the problem type.

[0170] The specific functions and examples of the modules and sub-modules of the apparatuses in the embodiments of the present disclosure are described above in the corresponding steps of the method embodiments, and will not be described here.

[0171] In the technical solutions of the present disclosure, the acquisition, storage, and application of user personal information comply with relevant laws and regulations and do not violate public order and good customs.

[0172] According to the embodiments of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium, and a computer program product.

[0173] Figure 11 A schematic block diagram of an example electronic device 1100 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present disclosure described and / or claimed in this document.

[0174] As Figure 11As shown, the device 1100 includes a computing unit 1101, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1102 or a computer program loaded from a storage unit 1108 into a random access memory (RAM) 1103. Various programs and data required for the operation of the device 1100 can also be stored in the RAM 1103. The computing unit 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0175] Various components in device 1100 are connected to I / O interface 1105, including an input unit 1106, such as a keyboard and mouse; an output unit 1107, such as various types of displays and speakers; a storage unit 1108, such as a magnetic disk and optical disk; and a communication unit 1109, such as a network card, a modem, a wireless communication transceiver, etc. Communication unit 1109 allows device 1100 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0176] Computing unit 1101 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of computing unit 1101 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Computing unit 1101 executes the various methods described above, such as the delivery process control method. For example, in some embodiments, the delivery process control method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 1108. In some embodiments, part or all of the computer program can be loaded and / or installed onto device 1100 via ROM 1102 and / or communication unit 1109. When the computer program is loaded into RAM 1103 and executed by computing unit 1101, one or more steps of the delivery process control method described above can be performed. Alternatively, in other embodiments, computing unit 1101 can be configured to execute the delivery process control method in any other suitable manner (e.g., via firmware).

[0177] The various embodiments of the systems and techniques described above can be implemented in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a load programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0178] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or the block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0179] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0180] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0181] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0182] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0183] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not limited herein.

[0184] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A method for controlling a delivery process, comprising: Identify intent based on input content and obtain the target stage to be executed in the delivery process; Checking whether the target stage has one or more preceding stages that fail set conditions; In the case where the one or more preceding stages exist, collecting parameter information required by the one or more preceding stages and the target stage; Triggering the one or more preceding tasks corresponding to the preceding stages and the target tasks corresponding to the target stage; Invoking a processing platform of a predecessor task corresponding to the one or more predecessor stages, and executing the predecessor task corresponding to the one or more predecessor stages according to parameter information required by the one or more predecessor stages; After all the preceding stages are completed, the processing platform of the target task is called to execute the target task according to the parameter information required by the target stage; Here, parameter information of a pre-stage is first collected, and a pre-task corresponding to the pre-stage is executed; After the preceding stage is completed, parameter information for the next stage is collected and the tasks corresponding to the next stage are executed until the target stage is completed.

2. The method according to claim 1, wherein Intent recognition is performed based on the input content to determine the target stages that need to be executed in the delivery process, including: Performing natural language analysis on historical conversation information related to the input content to extract historical intent, and pre-filling parameter information of the process stage corresponding to the historical intent; The large model is used to understand the intent of the input content to extract the current intent and obtain the target stage that needs to be executed corresponding to the current intent.

3. The method according to claim 2, wherein: The parameter information required for the target stage includes one or more of the following: Parameters pre-populated based on the recognition results of the historical conversation information; pre-populated parameters based on the recognition results of the input content; The interactive parameters are inputted in the interactive interface in response to the interactive command of the target stage.

4. The method according to claim 3, wherein: Collect parameter information required for the target phase, including one or more of the following: displaying recommended parameters for the target stage based on pre-populated parameters for the target stage, and obtaining parameter information required for the target stage in response to a modification command and / or a confirmation command for the recommended parameters; In response to the interaction parameters of the target stage, parameter information required by the target stage is obtained.

5. The method according to any one of claims 1 to 4, further comprising: Parameter verification is performed on parameter information required by the one or more preceding stages and the target stage.

6. The method according to claim 5, wherein: Triggering the one or more preceding tasks corresponding to the preceding stages and the target tasks corresponding to the target stage includes: In response to parameter verification results of the target stage and the one or more preceding stages being passed, a target task corresponding to the target stage and preceding tasks corresponding to the one or more preceding stages are triggered.

7. The method according to any one of claims 1 to 4, further comprising: Send the tasks corresponding to one or more process stages to the task scheduling system for hosting; wherein the tasks corresponding to the one or more process stages include the target tasks corresponding to the target stage and / or the predecessor tasks corresponding to one or more predecessor stages of the target stage.

8. The method according to claim 7, further comprising: The task status storage module of the task scheduling system is used to store information about the managed tasks; The execution process of the tasks corresponding to the one or more process stages is controlled by the task scheduling service of the task scheduling system.

9. The method according to claim 8, wherein Controlling the execution of the tasks corresponding to the one or more process stages through the task scheduling service of the task scheduling system includes: Executing the managed multiple tasks in the order of process stages through the task scheduling service; According to the current stage of the current task, a processing platform corresponding to the current stage is called to execute the current task.

10. The method according to claim 9, further comprising one or more of the following steps: If the current stage is not the target stage and the current stage meets the set conditions, trigger the execution of the task corresponding to the next process stage; If the current stage is the target stage and the current stage passes the set conditions, generating a reply content of the delivery result; When the status of the current task is execution failure, the current task is terminated and the problem is located.

11. The method according to claim 10, wherein: Perform problem location, including: Input the parameter information and execution data of the current task into the big model to determine the problem type of the current task; wherein the problem type includes one or more of the following: code problems, self-operation and maintenance tool problems, and expected failure problems; use the big model to output the failure cause analysis and / or processing suggestions of the current task according to the problem type.

12. A control device for a delivery process, comprising: The recognition module is used to identify the intent based on the input content and obtain the target stage to be executed in the delivery process; A checking module, configured to check whether the target stage has one or more preceding stages that fail to meet set conditions; a collecting module, configured to collect parameter information required by the one or more preceding stages and the target stage when the one or more preceding stages exist; A triggering module, configured to trigger the preceding tasks corresponding to the one or more preceding stages and the target tasks corresponding to the target stage; A calling module, configured to call a processing platform of a preceding task corresponding to the one or more preceding stages, and execute the preceding task corresponding to the one or more preceding stages according to parameter information required by the one or more preceding stages; After all the preceding stages are completed, the processing platform of the target task is called to execute the target task according to the parameter information required by the target stage; The collecting module, the triggering module and the calling module first collect parameter information of a pre-stage and execute the pre-task corresponding to the pre-stage; After the preceding stage is completed, parameter information for the next stage is collected and the tasks corresponding to the next stage are executed until the target stage is completed.

13. The device according to claim 12, wherein The identification module includes: A language analysis submodule is used to perform natural language analysis on historical conversation information related to the input content to extract historical intent and pre-fill parameter information of the process stage corresponding to the historical intent; The intention understanding submodule is used to use the large model to understand the intention of the input content, so as to extract the current intention and obtain the target stage that needs to be executed corresponding to the current intention.

14. The device according to claim 13, wherein The parameter information required for the target stage includes one or more of the following: Parameters pre-populated based on the recognition results of the historical conversation information; pre-populated parameters based on the recognition results of the input content; The interactive parameters are inputted in the interactive interface in response to the interactive command of the target stage.

15. The device according to claim 14, wherein Collect parameter information required for the target phase, including one or more of the following: displaying recommended parameters for the target stage based on pre-populated parameters for the target stage, and obtaining parameter information required for the target stage in response to a modification command and / or a confirmation command for the recommended parameters; In response to the interaction parameters of the target stage, parameter information required for the target stage is obtained.

16. The device according to any one of claims 12 to 15, further comprising: The verification module is used to perform parameter verification on the parameter information required for the target stage.

17. The device according to claim 16, wherein The trigger module is configured to trigger a target task corresponding to the target stage and a predecessor task corresponding to the one or more predecessor stages in response to a parameter verification pass result of the target stage and the one or more predecessor stages.

18. The apparatus according to any one of claims 12 to 15, further comprising: A sending module is used to send tasks corresponding to one or more process stages to a task scheduling system for hosting; wherein the tasks corresponding to the one or more process stages include target tasks corresponding to the target stage and / or predecessor tasks corresponding to one or more predecessor stages of the target stage.

19. The apparatus according to claim 18, further comprising: A saving module, configured to save information of the managed task via the task status storage module of the task scheduling system; The control module is used to control the execution process of the tasks corresponding to the one or more process stages through the task scheduling service of the task scheduling system.

20. The device according to claim 19, wherein The control module includes: An execution submodule, configured to execute the multiple tasks managed by the task scheduling service in the order of the process stages; The calling submodule is used to call the processing platform corresponding to the current stage to execute the current task according to the current stage of the current task.

21. The apparatus according to claim 20, further comprising one or more of the following steps: The trigger module is further configured to trigger the execution of tasks corresponding to the next process stage if the current stage is not the target stage and the current stage meets the set conditions; A generating module, configured to generate a reply content of a delivery result if the current stage is the target stage and the current stage passes a set condition; The problem locating module is used to terminate the current task and locate the problem when the status of the current task is execution failure.

22. The device according to claim 21, wherein The problem location module is used to input the parameter information and execution data of the current task into the big model to determine the problem type of the current task; wherein the problem type includes one or more of the following: code problems, self-operation and maintenance tool problems, and expected failure problems; use the big model to output the failure cause analysis and / or processing suggestions of the current task according to the problem type.

23. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 11.

24. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-11.

25. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 11.

Images