Detection method and device based on computing power of intelligent computing center

Abstract

The invention provides a detection method and device based on computing power of an intelligent computing center, and relates to the technical field of intelligent computing centers, intelligent computing centers and computing power infrastructures, the method comprises the following steps: S1, receiving input static planning information, the static planning information being used for indicating business processing logic of a target business; and S2, pre-detecting the static planning information based on the computing power of the intelligent computing center to obtain pre-detection information, the pre-detection information being at least used for indicating whether the static planning information will generate execution abnormity in the pipeline stage. According to the method, the static planning information is pre-detected by using the computing power of the intelligent computing center, so that the condition of executing the static planning with defects is avoided, the execution reliability of the corresponding static planning is improved, the probability that the corresponding static planning passes through the assembly line stage at one time is further improved, and the efficiency is improved. And the calculation power waste and the time loss caused by repeatedly executing the static planning in the assembly line are avoided.

Description

Technical Field

[0001] The present invention relates to the technical fields of intelligent computing centers, intelligent computing centers and computing power infrastructure, and particularly relates to a detection method and device based on the computing power of an intelligent computing center. Background Art

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

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

[0004] The "intelligent computing center" includes but is not limited to the "intelligent computing center".

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

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

[0007] Currently, how to quickly generate an agent that matches the requirements is an urgent problem to be solved. A possible solution is as follows: By means of custom coding and / or external import of open-source code, a batch of general functions are pre-constructed in advance. This part of the pre-constructed functions can be called base functions. The base function is used to generate a base function schema and provide it to the large model so that the large model can master the uses and usages of the base function, and then form a static plan for a new task. Then, the static plan is used to quickly construct an agent that meets specific requirements; in an actual scenario, the base function may not cover all requirements. To address this problem, code can be temporarily written through a large language model (LLM) to generate an adapter function that can fill the gap of the base function to cooperate with the base function to achieve comprehensive coverage of the requirements.

[0008] In the application, the static plan output by the large model may have certain formal errors. Such formal errors will cause exceptions during the pipeline stage of the plan and result in relatively serious waste of computing power.

[0009] It can be seen that in the existing technology, there are problems of low reliability in the execution of the static plan and serious waste of computing power in the pipeline stage of the static plan. Summary of the Invention

[0010] The purpose of the present invention is to provide a detection method and device based on the computing power of an intelligent computing center to solve the problems of low reliability in the execution of the static plan and serious waste of computing power in the pipeline stage of the static plan in the related technology.

[0011] To solve the above problems, the present invention is implemented as follows:

[0012] In a first aspect, the present invention provides a detection method based on the computing power of an intelligent computing center, including:

[0013] Step S1: Receive the input static plan information, where the static plan information is used to indicate the business processing logic of the target business;

[0014] Step S2: Perform pre-detection on the static plan information based on the computing power of the intelligent computing center to obtain pre-detection information, where the pre-detection information is at least used to indicate whether an execution exception will occur in the pipeline stage of the static plan information.

[0015] In one embodiment, step S2 includes:

[0016] Step S21: Based on the computing power of the intelligent computing center, in the multiple step data bodies arranged in order included in the static planning information, check whether the data format of each step data body matches a set standard data format to obtain a format check result, where the pre-detection information includes the format check result, and the step data body is used to indicate one business processing step among the multiple business processing steps corresponding to the business processing logic, and the business processing steps corresponding to different step data bodies are different.

[0017] In one embodiment, step S2 further includes:

[0018] Step S22: In the case where the format check result indicates that the data format of each step data body matches the standard data format, based on the computing power of the intelligent computing center, check whether each step data body matches a set keyword to obtain a keyword check result, where the pre-detection information further includes the keyword check result.

[0019] In one embodiment, step S2 further includes:

[0020] Step S23: In the case where the keyword check result indicates that each step data body includes the keyword, based on the computing power of the intelligent computing center, check whether the placeholder identifier included in each step data body matches a set placeholder identification rule to obtain a placeholder identification check result, where the pre-detection information further includes the placeholder identification check result.

[0021] In one embodiment, step S2 further includes:

[0022] Step S24: In the case where the placeholder identification check result indicates that the placeholder identifier included in each step data body matches the placeholder identification rule, based on the computing power of the intelligent computing center, check whether the output data referenced by each step data body matches a set reference rule to obtain an output reference check result, where the pre-detection information further includes the output reference check result.

[0023] In one embodiment, in the case where the pre-detection information indicates that the static planning information will cause an execution exception in the pipeline stage, the pre-detection information is further used to indicate: among the multiple step data bodies arranged in order included in the static planning information, the defective step data body that causes the execution exception, and the defect information corresponding to the defective step data body.

[0024] In one embodiment, the static planning information includes a plurality of step data bodies arranged in an orderly manner and satisfying a set prompt rule, and the prompt rule includes at least one of the following:

[0025] The position of each step data body in the static planning information is indicated based on a set data marker;

[0026] The input information of each step data body is encapsulated as a placeholder identifier based on a set placeholder identification rule;

[0027] The business processing method corresponding to the step data body sorted last among the plurality of step data bodies is a set standard method, and the standard method includes an output method, an alarm method, and a large model method.

[0028] In one embodiment, after step S2, the method further includes:

[0029] Step S3: Based on the computing power of the intelligent computing center, when the pre-detection information indicates that the static planning information will not cause an execution exception in the pipeline stage, extract at least one planning object from the static planning information, where the planning object includes the business device and / or business data corresponding to the static planning information;

[0030] Step S4: Based on the computing power of the intelligent computing center, calculate the association degree between each planning object and the target business among the at least one planning object to form association prompt information, and the association prompt information is at least used to indicate whether the static planning information includes a planning object whose association degree with the target business is less than or equal to a set association threshold.

[0031] In one embodiment, after step S2, the method further includes:

[0032] Step S5: Based on the computing power of the intelligent computing center, perform a feasibility analysis on the business process indicated by the static planning information to obtain a feasibility analysis result, and the business process is generated based on the step description text of each step data body in the plurality of step data bodies arranged in an orderly manner included in the static planning information.

[0033] In a second aspect, the present invention further provides a detection device based on the computing power of an intelligent computing center, including:

[0034] A receiving module, configured to receive the input static planning information, where the static planning information is used to indicate the business processing logic of the target business;

[0035] A checking module is configured to check the static planning information based on the computing power of the intelligent computing center to obtain pre-detection information, where the pre-detection information is at least used to indicate whether an execution exception will occur in the pipeline stage of the static planning information.

[0036] In a third aspect, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, the steps in the detection method based on the computing power of the intelligent computing center as described in the first aspect above are implemented.

[0037] In a fourth aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the detection method based on the computing power of the intelligent computing center as described in the first aspect above are implemented.

[0038] In a fifth aspect, the present invention further provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the steps in the detection method based on the computing power of the intelligent computing center as described in the first aspect above are implemented.

[0039] In the present invention, after receiving the input static planning information, the computing power of the intelligent computing center is used to pre-detect the static planning information to check for information defects in the static planning information that may cause execution exceptions, avoiding the situation of executing a static planning with information defects, improving the reliability of the execution of the corresponding static planning, thereby increasing the probability of the corresponding static planning passing through the pipeline stage at one time, and avoiding the waste of computing power and time consumption caused by repeatedly executing the static planning in the pipeline. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a schematic flowchart of a detection method based on the computing power of an intelligent computing center provided by the present invention;

[0041] Figure 2 is an architecture diagram of a two-stage intelligent agent automation workflow provided by the present invention;

[0042] Figure 3 is a schematic structural diagram of a detection device based on the computing power of an intelligent computing center provided by the present invention;

[0043] Figure 4 is a schematic structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] The technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0045] The "computing power" as described in the present invention refers to the ability of a computer device or a computing / data center to process information, which is the ability of computer hardware and software to cooperate to jointly execute a certain computing requirement. It is the computing ability to achieve the output of the target result by processing information data. It is a new type of productive force integrating information computing power, network carrying capacity, and data storage capacity, and mainly provides services to society through computing power infrastructure.

[0046] The "computational power" (Computational Power, CP) as described in the present invention is an ability of a data center server to process data and achieve result output, and is a comprehensive index for measuring the computing ability of a data center, including general computing ability, supercomputing ability, and intelligent computing ability. The commonly used measurement unit is the number of floating-point operations per second (FLOPS: Floating Point Operations Per Second, 1EFLOPS = 10^18 FLOPS). The larger the value, the stronger the comprehensive computing ability. It is estimated that 1EFLOPS is approximately the computing power output of 5 Tianhe-2A or 500,000 mainstream server CPUs or 2 million mainstream laptops. The calculation formula is: CP = CP 通用 +CP 智能 +CP 超级 。

[0047] The "carrying capacity" (Network Power, NP) as described in the present invention is the manifestation of the data transmission ability of computing power facilities, including comprehensive capabilities such as network architecture, network bandwidth, transmission delay, intelligent management and scheduling, etc. The carrying capacity involves network transmission inside and between data centers and is a comprehensive index for measuring network transmission scheduling ability. In the present invention, the video memory bandwidth is used for the carrying capacity.

[0048] The "Storage Power (SP)" described in the present invention is the comprehensive ability of a data center in four aspects: data storage capacity, performance, security and reliability, and green and low-carbon, which is a comprehensive indicator for measuring the data storage capacity of a data center and includes external storage devices such as storage arrays and built-in storage devices of servers. The commonly used measurement unit for storage capacity is exabyte (EB, 1EB = 2^60 bytes), the commonly used measurement unit for performance is the number of read and write operations per second per unit capacity (Input / Output Operations Per Second / TB, IOPS / TB), and the disaster recovery ratio is an important manifestation of security and reliability.

[0049] The "computing power infrastructure" described in the present invention is a new type of information infrastructure that integrates information computing power, network carrying capacity, and data storage power, and can realize the centralized computing, storage, transmission, and application of information.

[0050] The "new type of information infrastructure" described in the present invention refers to: mainly including network infrastructures such as 5G networks, fiber broadband networks, backbone networks, international communication networks, and satellite Internet, computing power infrastructures such as data centers, general computing power centers, intelligent computing centers, and supercomputing centers, and new technology facilities such as artificial intelligence, blockchain, and quantum computing.

[0051] The "computing power" described in the present invention includes general computing power, intelligent computing power, and super computing power.

[0052] The "general computing power" described in the present invention is the computing power provided by servers based on central processing unit (CPU) chips and is used to support basic general computing such as cloud computing and edge computing.

[0053] The "intelligent computing power" described in the present invention is for various artificial intelligence innovation applications, and is a computing platform based on the large-scale deployment of dedicated chips such as graphics processing unit (GPU), field programmable gate array (FPGA), and application specific integrated circuit (ASIC), such as natural language processing and machine vision.

[0054] The "super computing power" described in the present invention is mainly the computing power provided by high-performance computing clusters such as supercomputers. It utilizes the centralized computing resources of multiple computer systems working in parallel and processes extremely complex or data-intensive problems through a dedicated operating system, and is mainly used for computing in cutting-edge scientific fields, such as planetary simulation, drug molecule design, and gene analysis.

[0055] The "Intelligent Computing Center" described in the present invention refers to a facility that uses large-scale heterogeneous computing power resources, including general computing power (CPU) and intelligent computing power (GPU, FPGA, etc.), and mainly provides the required computing power, data, and algorithms for artificial intelligence applications (such as scenarios like artificial intelligence deep learning model development, model training, and model inference). The intelligent computing center covers facilities, hardware, and software, and can provide full-stack capabilities from underlying computing power to top-level application enabling.

[0056] The "Intelligent Computing Center" described in the present invention includes, but is not limited to, the "Intelligent Computing Center".

[0057] The "Intelligent Computing Center" described in the present invention, namely the artificial intelligence computing center, is a type of computing power infrastructure that is based on artificial intelligence theory, adopts an artificial intelligence computing architecture, and provides computing power services, data services, and algorithm services required for artificial intelligence applications.

[0058] The "Computing Power Center" described in the present invention refers to a facility mainly composed of infrastructure such as wind, fire, water, and electricity, and IT software and hardware devices, and has computing power, carrying capacity, and storage capacity, including general data centers, intelligent computing centers, supercomputing centers, etc.

[0059] The "Supercomputing Center" described in the present invention, that is, the supercomputing data center, is a data center based on supercomputers or large-scale computing clusters, and can provide functions such as large-scale computing, storage, and network services, and is widely used in application scenarios such as aerospace, national defense, oil exploration, climate modeling, and genome sequencing.

[0060] The "Computing Power Resources" described in the present invention refer to technologies and facilities required for the development of the digital society with information computing, transmission, storage, and application capabilities, including but not limited to computing resources such as CPU and GPU, network resources such as switches and routers, storage resources such as storage arrays and distributed storage, security resources such as firewalls and intrusion detection systems, and support and guarantee resources such as wind, fire, water, and electricity.

[0061] The "models" and "large models" described in the present invention include, but are not limited to, "large language models" and "multi-modal large models".

[0062] The "Large Language Model" described in the present invention refers to a large language model (LLM), which is a language model with a relatively large number of parameters, aiming to understand and generate human language, trained through a large amount of text data, and can perform a wide range of tasks including text summarization, translation, sentiment analysis, etc.

[0063] The "multimodal large models" referred to in the present invention mean models trained by jointly using multimodal information such as text, images, videos, and audio, including but not limited to multimodal large language models.

[0064] The "planner" referred to in the present invention means a large model for task planning, which is used to generate a static plan in one go. The static plan contains all the business processing steps for fulfilling the business requirements indicated by the user. Each business processing step includes its explanation (also known as the step description text), the method name called, and the input parameter values (a dictionary with each parameter and its corresponding assignment).

[0065] The "static planning information" referred to in the present invention means the data representation of the static plan, which at least includes a plurality of step data bodies arranged in an orderly manner. The plurality of step data bodies correspond one-to-one to all the business processing steps of the business requirements indicated by the aforementioned user.

[0066] The "step data body" referred to in the present invention is the data representation of a corresponding business processing step, including but not limited to: the step description text of a corresponding business processing step (used to describe the business processing step to indicate the function, purpose, and / or execution method of the step, so as to facilitate understanding and implementation by users or developers), the step number of a corresponding business processing step among the plurality of business processing steps, the input data of a corresponding business processing step (such as the data name, data type requirements, data source, etc. of the input data), and the data processing method of a corresponding business processing step (calling a specific method through the method name).

[0067] The "pipeline" referred to in the present invention means the process of executing the static plan.

[0068] Please refer to Figure 1 , Figure 1 which is a detection method based on the computing power of an intelligent computing center provided by the present invention. As Figure 1 shown, it includes the following steps:

[0069] Step S1: Receive the input static planning information.

[0070] Among them, the static planning information is used to indicate the business processing logic of the target business.

[0071] The above-mentioned static planning information is generated by the "planner". The input of the planner is the business requirement information of the target business, and the output of the planner is the static plan indicating the business processing logic of the target business. The data representation of this static plan is the static planning information.

[0072] Exemplarily, the target service may be a service corresponding to computer vision (CV), such as: oil stain detection service, crack detection service, pedestrian tracking service, etc.

[0073] The service requirement information includes a requirement description text of the target service, and the requirement description text may include a scenario text indicating the service scenario of the target service, an object text indicating the data or device involved in the target service, a target text indicating the service purpose of the target service, a process text indicating the service processing flow of the target service, etc.

[0074] Step S2, pre-detect the static planning information based on the computing power of the intelligent computing center to obtain pre-detection information.

[0075] Among them, the pre-detection information is at least used to indicate whether an execution exception will occur in the static planning information in the pipeline stage.

[0076] It should be noted that the execution exception indicated by the pre-detection information should be understood as an execution exception caused by the information defect of the static planning information. For example: there is a data format mismatch of one or more step data bodies in the static planning information with the set standard data format, or there is a data type mismatch of the keywords included in one or more step data bodies in the static planning information with the set standard data type, etc.

[0077] In the present invention, after receiving the input static planning information, the computing power of the intelligent computing center is used to pre-detect the static planning information to check for information defects in the static planning information that may cause execution exceptions, avoid the situation of executing static planning with information defects, improve the reliability of the execution of the corresponding static planning, and further increase the probability of the corresponding static planning passing through the pipeline stage at one time, avoiding the waste of computing power and time consumption caused by repeatedly executing the static planning in the pipeline.

[0078] It should be specifically noted that the present invention specifically uses the computing power of the intelligent computing center to perform the actions in step S2. By utilizing the large-scale heterogeneous computing power resources (such as CPUs, GPUs, and FPGAs) equipped in the intelligent computing center, it is possible to support the rapid processing and analysis of a large amount of data, so as to complete the pre-detection processing of static planning information in an extremely short time; and by using parallel computing resources such as GPUs, it is possible to perform pre-detection processing on multiple static planning information simultaneously, which can further improve the pre-detection efficiency, especially supporting the batch pre-detection processing of a large amount of static planning information; and it is possible to utilize deep learning and machine learning algorithms to at least partially support the automatic correction of information defects identified in the static planning information, so as to improve the efficiency of the pipeline stage of the static planning information; in addition, by utilizing the heterogeneous computing resources and rich tool library expansion of the intelligent computing center, it is possible to support the inspection of multiple data formats (such as images, texts, audio, etc.), so as to expand the configurable data formats of the static planning information and support diverse application requirements.

[0079] Generally speaking, by utilizing the powerful computing power and flexible resource configuration of the intelligent computing center, during the pre-detection process of static planning information, it is possible to effectively improve the efficiency and quality of the pre-detection processing, provide reliable static planning information for subsequent applications, and thus support the intelligent computing center to realize the requirements of the intelligent agent automated workflow (which means that through the computing power of the intelligent computing center in an automated manner, training the task planning large model to master the functions and usage of the base function, and using the task planning large model as the planning component of the pre-generated intelligent agent, dynamically generating different static planning information for different application requirements received by the intelligent agent, and executing the static planning information to obtain the execution results corresponding to the application requirements) on this basis.

[0080] Exemplarily, by utilizing the powerful computing power and flexible resource configuration of the intelligent computing center, the architecture for implementing the two-stage auto-pipeline of the intelligent agent can be as Figure 2 shown Figure 2 The relevant illustrations in

[0081] User query: Users input their request or demand content.

[0082] Planner: Used to parse user queries and formulate an execution plan. If the parsing fails, it may output an error that cannot be converted to the JSON format.

[0083] Plan list: A specific list of steps generated according to user queries.

[0084] Retrieve all-step function dependencies: Retrieve the function dependencies associated with each step to ensure that all required functions are available.

[0085] Get executable input values:

[0086] Given values: The values directly provided to the function.

[0087] User inputs: The dynamic input values provided by the user in the query.

[0088] Previous function outputs: The function output values from the previous steps.

[0089] Media download: This link can also support the download of additional required media content.

[0090] Execute one step: It refers to the indication of the execution details of a specific step.

[0091] Execute function: Used to actually call the function to perform data processing work based on the obtained input values.

[0092] Base function: Execute specific basic functions to support the agent to have the set basic business processing capabilities.

[0093] Adapter function (autoencoder): As a supplement to the base function.

[0094] Retrieve depended function schemas: Obtain the structures or schemas of other dependent functions.

[0095] Autoencoder coding: Automatically generate the code involved in the above process.

[0096] Exception code: If an error or exception occurs, the system will return a specific exception code for subsequent analysis and debugging. For example:

[0097] The exceptions can be classified into the following 6 categories:

[0098] 0: No exception, the pipeline execution is successful

[0099] 1: Other pipeline exceptions

[0100] 2: Automatic encoding execution failed

[0101] 3: Basic function execution failed

[0102] 4: Failure to obtain the plan (usually caused by failure to convert the plan to json)

[0103] 5: The variable placeholder in the plan is expressed incorrectly and parsing fails.

[0104] It should be noted that the application of a two-stage auto-pipeline has at least the following advantages:

[0105] 1. Higher accuracy: more cases where the given plan can be successfully executed;

[0106] 2. By decomposing the tasks, the total time of LLM inference can be shortened;

[0107] 3. By transferring core information between manual processing steps, the scope of effective information can be clarified and the transparency of the system can be increased.

[0108] The application of the detection method of the computing power of the intelligent computing center based on the present invention can avoid the occurrence of execution exceptions of exception types 4-5 as much as possible during the two-stage intelligent body automated workflow, so as to save the computing power and time waste caused by handling this type of execution exceptions.

[0109] The results of the two-stage auto-pipeline experiment based on multiple open source AI tools are shown in Table 1. The overall goal achievement rate is close to 54%, and errors mainly occur in the basic function execution stage and the variable placeholder expression stage.

[0110]

[0111]

[0112] Table 1

[0113] Among them, errors in the basic function execution phase include: incorrect parameter input, confused use of image lists and video paths, some parameters unrelated to the input (such as thresholds), etc.; errors in the variable placeholder expression phase include: importing external packages (lack of package dependencies), incorrect understanding and processing of complex parameters, etc.

[0114] In one embodiment, the static planning information includes a plurality of step data bodies arranged in an orderly manner and satisfying a set hint rule, and the hint rule includes at least one of the following:

[0115] The position of each step data body in the static planning information is indicated based on a set data marker;

[0116] The input information of each step data body is encapsulated as a placeholder identifier based on a set placeholder identification rule;

[0117] The business processing method corresponding to the step data body sorted last among the plurality of step data bodies is a set standard method, and the standard method includes an output method, an alarm method, and a large language model (llm) method.

[0118] Based on the above settings, the static planning information is made to approach the understanding habit of the large language model, thereby enhancing the business processing ability of the intelligent agent trained based on the static planning information.

[0119] Among them, indicating the position of each step data body in the static planning information based on a set data marker can ensure the accurate identification of the step data body by the intelligent computing center. The set data marker can be an md identifier, specifically including a head identifier "```json" and a tail identifier "```", and the step data body is located between the head identifier "```json" and the tail identifier "```".

[0120] Encapsulating the input information of each step data body as a placeholder identifier based on a set placeholder identification rule can reduce the situation of missing the "step", and ensure the logical feasibility of the static planning information.

[0121] Specifically, the set placeholder identification rule can be:

[0122] When the input information indicates that the user needs to input data, the character "input" is used in the placeholder identifier for substitution. For the situation involving input materials such as photos and images, it is considered that the user needs to input data;

[0123] When the input information indicates referring to the method output of the previous step data body, the character "output" is used in the placeholder identifier for substitution;

[0124] The placeholder identifier includes the serial number N of the business processing step corresponding to the step data body, specifically the character "step_N", and N is of the integer (int) type;

[0125] Set the variable name corresponding to the input information in the placeholder identifier.

[0126] For example, based on the placeholder identification rules set in the above example, when the input information of the step data body corresponding to the first business processing step indicates that the value of `video_path` in the input parameters needs to be input by the user, the corresponding placeholder identifier can be expressed as: `"<|input_step_1.video_path|>"`.

[0127] Setting the business processing method corresponding to the step data body sorted last among the multiple step data bodies as the set standard method can make the result finally output by the static planning information controllable in the pipeline stage, so as to improve the practicability of the static planning information.

[0128] In one embodiment, step S2 includes:

[0129] Step S21: Based on the computing power of the intelligent computing center, in the multiple step data bodies arranged in an orderly manner included in the static planning information, check whether the data format of each step data body matches the set standard data format to obtain a format check result, where the pre-detection information includes the format check result, and the step data body is used to indicate one business processing step among the multiple business processing steps corresponding to the business processing logic, and the business processing steps corresponding to different step data bodies are different.

[0130] Exemplarily, the standard data format can be the json format; in this example, if the data format of a step data body conforms to the format requirements of the json format, it can be determined that the data format of this step data body matches the set standard data format, and if the data format of a step data body does not conform to the format requirements of the json format, it can be determined that the data format of this step data body does not match the set standard data format. The above check operation can be completed based on the json verification tool externally connected to the intelligent computing center, or can be completed by the json verification tool built in the intelligent computing center.

[0131] It should be noted that in the pipeline stage of the static planning information, it is usually necessary to encapsulate the multiple step data bodies arranged in an orderly manner included in the static planning information into multiple data objects (the data format of the data object is the standard data format) to facilitate subsequent data transmission and processing work.

[0132] If there is one or more step data bodies in the static planning information whose data formats do not match the set standard data format, then in the pipeline stage, it will cause the corresponding data object to fail to be generated normally, and further cause an execution exception.

[0133] Specifically, when the data format of each of the step data bodies matches the set standard data format, the format check result indicates that the static planning information passes the format check step; while when there is at least one step data body whose data format does not match the set standard data format, the format check result is at least used to indicate that the static planning information fails the format check step. In this case, the format check result can further indicate the serial number of the step data body that does not match the set standard data format, so as to facilitate the subsequent positioning and modification of the step data body that does not match the set standard data format.

[0134] In one embodiment, step S2 further includes:

[0135] Step S22: When the format check result indicates that the data format of each step data body matches the standard data format, check whether each step data body matches the set keywords based on the computing power of the intelligent computing center to obtain a keyword check result, where the pre-detection information further includes the keyword check result.

[0136] Exemplarily, the set keywords may include at least one of the following: "step", "function", "input". Among them, the keyword "step" is the key of the key-value pair indicating the step description text of the step data body, the keyword "function" is the key of the key-value pair indicating the business processing method of the step data body, and the keyword "input" is the key of the key-value pair indicating the input data of the step data body.

[0137] It should be noted that the step data body includes one or more key-value pairs. When the keys of the one or more key-value pairs included in the step data body form a key set and the key set has only the keywords, it is determined that the step data body matches the set keywords; otherwise, it is determined that the step data body does not match the set keywords.

[0138] Specifically, when the data format of each step data body matches the set keywords, the keyword check result indicates that the static planning information passes the keyword check step; while when there is at least one step data body whose data format does not match the set keywords, the keyword check result is at least used to indicate that the static planning information fails the keyword check step. In this case, the keyword check result can further indicate the serial number of the step data body that does not match the set keywords, so as to facilitate the subsequent positioning and modification of the step data body that does not match the set keywords.

[0139] Further, in this embodiment, further requirements can be imposed on the data types of the values corresponding to at least some of the set keywords, so as to improve the usability of the step data body by standardizing the data types of the values corresponding to the keywords.

[0140] For example, it can be set that the data types of the values corresponding to the keyword "step" in the step data body and the keyword "function" are both of the string type. In this case, it can be specified that when the key set has and only has the set keywords, and the data types of the values corresponding to the keyword "step" in the step data body and the keyword "function" are both of the string type, it is determined that the step data body matches the set keywords; otherwise, it is determined that the step data body does not match the set keywords.

[0141] In one embodiment, step S2 further includes:

[0142] Step S23: When the keyword check result indicates that each step data body includes the keyword, check whether the placeholder identifiers included in each step data body match the set placeholder identification rules based on the computing power of the intelligent computing center, so as to obtain a placeholder identification check result, where the pre-detection information further includes the placeholder identification check result.

[0143] Among them, for the set placeholder identification rules, refer to the foregoing descriptions and examples. In the application, when it is detected that the placeholder identifier of the step data body includes the character "input" referring to the user input (the character "output" output by the method of referring to the previous step data body), the character "step_N" referring to the serial number N of the corresponding business processing step, and the variable name character referring to the corresponding input variable name, it is determined that the currently checked placeholder identifier of the step data body matches the set placeholder identification rules; otherwise, it is determined that the currently checked placeholder identifier of the step data body does not match the set placeholder identification rules.

[0144] Specifically, when each placeholder identifier of each step data body matches the set placeholder identification rules, the placeholder identification check result indicates that the static planning information passes the placeholder identification check step; while when there is at least one placeholder identifier that does not match the set placeholder identification rules, the placeholder identification check result is at least used to indicate that the static planning information fails to pass the placeholder identification check step. In this case, the placeholder identification check result can further indicate the placeholder identifier that does not match the set placeholder identification rules and the serial number of the step data body where it is located, so as to facilitate the subsequent positioning and modification of the placeholder identifier that does not match the set placeholder identification rules.

[0145] In one embodiment, step S2 further includes:

[0146] Step S24, when the placeholder identification check result indicates that each placeholder identifier included in the step data body matches the placeholder identification rule, based on the computing power of the intelligent computing center, check whether the output data referenced by each step data body matches the set reference rule to obtain an output reference check result, where the pre-detection information further includes the output reference check result.

[0147] Wherein, when the placeholder identifier in the step data body includes the "output" character that references the method output of the previous step data body, by further checking whether the method output of the step data body before this step data body (i.e., the aforementioned referenced output data) includes the variable name character in the current placeholder identifier, to determine whether the output data referenced by the step data body matches the set reference rule;

[0148] If it does not include, it is determined that the output data referenced by the step data body does not match the set reference rule, and if it includes, it is determined that the output data referenced by the step data body matches the set reference rule.

[0149] Specifically, when the output data referenced by each placeholder identifier with a reference setting matches the set reference rule, the output reference check result indicates that the static planning information passes the output reference check step; and when there is at least one piece of output data referenced by the placeholder identifier that does not match the set reference rule, the placeholder identification check result is at least used to indicate that the static planning information fails the output reference check step. In this case, the output reference check result can further indicate the placeholder identifier that does not match the set reference rule and the serial number of the step data body where it is located, to facilitate the subsequent positioning and modification of the placeholder identifier that does not match the set reference rule.

[0150] In some embodiments, for the static planning information that passes the format check step, keyword check step, placeholder identification check step, and output reference check step, each placeholder identifier in the static planning information can be replaced with a corresponding temporary variable, and then the overall detection of the step data body can be performed according to the set format requirements, to check whether the data type of the value of each key-value pair in the step data body meets the type specified by the set format requirements.

[0151] In one embodiment, when the pre-detection information indicates that the static planning information will cause an execution exception in the pipeline stage, the pre-detection information is further used to indicate: among the multiple step data bodies arranged in an orderly manner included in the static planning information, the defective step data body that causes the execution exception, and the defect information corresponding to the defective step data body.

[0152] Exemplarily, the defect information may be formed according to a set error code text, and the error code text is as follows:

[0153] Error code text(ERR0R_CODE_TEXT) = {

[0154] -1: No error.("No error.")

[0155] 0: The generated plan is not in a valid json format.("Planner generation is not validjson.")

[0156] 1: Some of the plan steps are missing the required keys (`step`, `function`, and `input`).("The plan step(s)[{steps_str}]do not contain all required keys:['step','function','input'].")

[0157] 2: The description or function name of some of the plan steps is not of string type.("The plan step(s)[{steps_str}]have non-string step description or function names.")

[0158] 3: The variable naming format of some of the plan steps is invalid. The correct format should be `<|output_step_n.var_name|>` or `<|input_step_n.var_name|>`.("The plan step(s)[{steps_str}]haveinvalid variable nomination format(s).Correct format should be<|output_step_n.var_name|>or<|input_step_n.var_name|>.")

[0159] 4: The input data type of some of the plan steps is invalid, and the error message will be provided via `{err_mess}`.("Theplan step[{steps_str}]has its input(s)being of invalid datatype(s):{err_mess}")

[0160] 5: The input variable `{var_name}` requested by the plan step was not found in the input or output of step `{step_num}`, with the specific direction specified by `{in_or_out}` ("The plan step[{steps_str}]has its inputasking for{var_name}from{step_num}'s{in_or_out}unfound.")

[0161] }

[0162] In one example, when the target business is oil pollution detection, the requirement description text can be:

[0163] The oil pollution detection agent produced by the power company needs to process the input images to detect whether there is oil pollution in 4 specific components (riser bushing, pressure relief valve, cooler, bushing) in the transformer and reactor. The task can be divided into the following steps: image preprocessing, object detection, segmentation to detect oil pollution, output the detection results and generate an alarm.

[0164] The static planning information generated according to the requirement description text can be:

[0165] <planlist>

[0167] {"step": "Preprocess the input image to ensure it is suitable for further processing.", "function": "preprocess", "input": {"image_path_list": "<|input_step_1.image_path_list|>", "resize_dims": [800, 600]}},

[0168] {"step": "Detect 4 key components of the oil-immersed transformer and high-voltage reactor in the image.", "function": "detection", "input": {"image_path_list": "<|output_step_1.preprocessed_path_list|>", "query_list": ["transformer_base", "pressure_relief_valve", "cooler", "bushing"], "method": "dino"}},

[0169] {"step": "Perform region segmentation on the detected components to obtain the surface oil spill area.", "function": "segmentation", "input": {"image_path_list": "<|output_step_2.predict_img_list|>", "query_list": ["oil_spill"], "method": "lan_sam"}},

[0170] {"step": "Post-process the detection and segmentation results, filter the oil spill areas and classify them.", "function": "postprocess", "input": {"image_path_list": "<|output_step_3.predict_img_list|>", "blur_type": "gaussian", "blur_ksize": [5, 5], "thresholding": [100, 255, "binary"]}},

[0171] ​{"step": "Output the detection result and initiate an alarm. If oil spill is detected, notify the relevant personnel.", "function": "alarm", "input": {"result_obj": {"image_path_list": "<|output_step_4.predict_img_list|>", "boxes_list": "<|output_step_4.boxes_list|>"}, "detect_class": ["oil_spill"], "detect_method": "appear", "alarm_threshold": 1}}

[0173] ​< / planlist>

[0174] The above static planning information includes 4 step data bodies arranged in order, corresponding to the 4 business processing steps of image preprocessing, object detection, segmentation to detect oil pollution, output the detection results and generate an alarm respectively. In the step data body corresponding to image preprocessing, the string "Preprocess the input images to ensure they are suitable for further processing." can be understood as the step description text, the string "preprocess" is the method name of the business processing method used, the character "<|input_step_1.image_path_list|>" indicates that this step requires user input, and the variable name of the input variable is "image_path_list", and the string "resize_dims":[800,600]} indicates the size requirement of the input image; the meanings of the step data bodies corresponding to steps such as object detection, segmentation to detect oil pollution, and output the detection results are similar, and for the sake of avoiding repetition, they will not be elaborated here.

[0175] In one embodiment, after step S2, the method further includes:

[0176] Step S3: Based on the computing power of the intelligent computing center, when the pre-detection information indicates that no execution exception will occur during the pipeline stage of the static planning information, extract at least one planning object from the static planning information, where the planning object includes the business device and / or business data corresponding to the static planning information;

[0177] Step S4: Based on the computing power of the intelligent computing center, calculate the association degree between each of the at least one planning object and the target service among the at least one planning object to form association prompt information, where the association prompt information is at least used to indicate whether the static planning information includes a planning object whose association degree with the target service is less than or equal to a set association threshold.

[0178] After the static planning information passes through the aforementioned pre-check processing, the association detection is completed by extracting at least one planning object from the static planning information and calculating the association degree between the target service and each planning object, and association prompt information is formed.

[0179] Exemplarily, the association degree between each of the planning objects and the target service can be obtained through a large model in the form of a question and answer.

[0180] Wherein, when the association prompt information indicates that the static planning information includes at least one planning object whose association degree with the target service is less than or equal to a set association threshold, the association prompt information may further include at least one planning object included in the static planning information whose association degree with the target service is less than or equal to the set association threshold and the corresponding association degree, so as to subsequently analyze whether to generate new static planning information based on this. The new static planning information does not include a planning object whose association degree with the target service is less than or equal to the set association threshold, and the static planning information is detected from a semantic perspective to avoid the static planning information including planning objects that are weakly associated or even not associated with the target service, so as to further improve the reliability of the static planning information.

[0181] Referring to the aforementioned example with the oil pollution detection service as the target service, in this example, the lifting seat, pressure relief valve, cooler, and casing can all be understood as planning objects.

[0182] In one embodiment, after step S2, the method further includes:

[0183] Step S5: Based on the computing power of the intelligent computing center, perform a feasibility analysis on the service process indicated by the static planning information to obtain a feasibility analysis result, where the service process is generated based on the step description text of each step data body in the multiple step data bodies arranged in an orderly manner included in the static planning information.

[0184] After the static planning information is processed by the foregoing pre-check, a feasibility analysis is performed through the indicated business process to complete the process feasibility analysis and form a feasibility analysis result, so as to subsequently analyze whether to delete or add the step data body included in the current static planning information based on this, in order to further improve the reliability of the static planning information.

[0185] Exemplarily, a feasibility analysis can be performed on the business process indicated by the static planning information in the form of a question and answer through a large model to obtain a feasibility analysis result.

[0186] It should be noted that steps S3 - S4 and step S5 can be executed synchronously or asynchronously (for example, first execute step S5, and then execute steps S3 - S4; or, first execute steps S3 - S4, and then execute step S5), and the present invention does not limit this.

[0187] In application, through the execution of steps S3 - S4 and step S5, semantic analysis of the static planning information can be achieved to identify, from a semantic perspective, parts of the static planning information that may cause waste of computing power, and corresponding risk warnings are made, so as to facilitate the generation of more accurate and reliable static planning information subsequently.

[0188] Please refer to Figure 3 , Figure 3 which is a schematic structural diagram of a detection device based on the computing power of an intelligent computing center provided by the present invention. As Figure 3 shown, the detection device 300 based on the computing power of the intelligent computing center includes:

[0189] A receiving module 301, configured to receive the input static planning information, where the static planning information is used to indicate the business processing logic of the target business;

[0190] An inspection module 302, configured to inspect the static planning information based on the computing power of the intelligent computing center to obtain pre-detection information, where the pre-detection information is at least used to indicate whether an execution exception will occur in the static planning information during the pipeline stage.

[0191] In one embodiment, the inspection module 302 includes:

[0192] A format inspection unit, configured to inspect, based on the computing power of the intelligent computing center, whether the data format of each step data body in the multiple step data bodies arranged in an orderly manner included in the static planning information matches a set standard data format to obtain a format inspection result, where the pre-detection information includes the format inspection result, and the step data body is used to indicate one business processing step among the multiple business processing steps corresponding to the business processing logic, and the business processing steps corresponding to different step data bodies are different.

[0193] In one embodiment, the checking module 302 includes:

[0194] A keyword checking unit, configured to check whether each of the step data bodies matches a set keyword based on the computing power of the intelligent computing center when the data format checking result indicates that the data format of each of the step data bodies matches the standard data format, so as to obtain a keyword checking result, where the pre-detection information further includes the keyword checking result.

[0195] In one embodiment, the checking module 302 includes:

[0196] A placeholder identifier checking unit, configured to check whether the placeholder identifiers included in each of the step data bodies match a set placeholder identifier rule based on the computing power of the intelligent computing center when the keyword checking result indicates that each of the step data bodies includes the keyword, so as to obtain a placeholder identifier checking result, where the pre-detection information further includes the placeholder identifier checking result.

[0197] In one embodiment, the checking module 302 includes:

[0198] An output reference checking unit, configured to check whether the output data referred to by each of the step data bodies matches a set reference rule based on the computing power of the intelligent computing center when the placeholder identifier checking result indicates that the placeholder identifiers included in each of the step data bodies match the placeholder identifier rule, so as to obtain an output reference checking result, where the pre-detection information further includes the output reference checking result.

[0199] In one embodiment, when the pre-detection information indicates that the static planning information will cause an execution exception in the pipeline stage, the pre-detection information is further used to indicate: among the multiple step data bodies arranged in an orderly manner included in the static planning information, the defective step data body that causes the execution exception, and the defect information corresponding to the defective step data body.

[0200] In one embodiment, the static planning information includes multiple step data bodies arranged in an orderly manner and satisfying a set hint rule, and the hint rule includes at least one of the following:

[0201] The position of each of the step data bodies in the static planning information is indicated based on a set data mark;

[0202] The input information of each of the step data bodies is encapsulated as a placeholder identifier based on a set placeholder identifier rule;

[0203] The business processing method corresponding to the last sorted step data body among the multiple step data bodies is a set standard method, and the standard method includes an output method, an alarm method, and a large model method.

[0204] In one embodiment, the detection device 300 based on the computing power of the intelligent computing center further includes:

[0205] An object extraction module, configured to extract at least one planning object from the static planning information based on the computing power of the intelligent computing center when the pre-detection information indicates that no execution exception will occur in the pipeline stage of the static planning information, where the planning object includes the business device and / or business data corresponding to the static planning information;

[0206] A correlation analysis module, configured to calculate the correlation degree between each of the at least one planning object and the target service based on the computing power of the intelligent computing center to form correlation prompt information, and the correlation prompt information is at least used to indicate whether the static planning information includes a planning object whose correlation degree with the target service is less than or equal to a set correlation threshold.

[0207] In one embodiment, the detection device 300 based on the computing power of the intelligent computing center further includes:

[0208] A feasibility analysis module, configured to perform a feasibility analysis on the service process indicated by the static planning information based on the computing power of the intelligent computing center to obtain a feasibility analysis result, where the service process is generated based on the step description text of each step data body in the multiple step data bodies arranged in an orderly manner included in the static planning information.

[0209] The detection device based on the computing power of the intelligent computing center provided by the present invention can implement each process of the above-mentioned detection method based on the computing power of the intelligent computing center, and the technical features correspond one by one and can achieve the same technical effects. To avoid repetition, details are not described here again.

[0210] It should be noted that the detection device based on the computing power of the intelligent computing center in the present invention can be a device, or a component, an integrated circuit, or a chip in an electronic device.

[0211] The present invention also provides an electronic device. Refer to Figure 4 , Figure 4 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. The electronic device includes a memory 401, a processor 402, and a program or instruction stored in the memory 401 and running on the processor 402. When the program or instruction is executed by the processor 402, it can implement Figure 1Any steps in the corresponding embodiments of the detection method based on the computing power of the intelligent computing center and the same beneficial effects achieved are not elaborated here.

[0212] Among them, the processor 402 can be a CPU, ASIC, FPGA or GPU.

[0213] Those of ordinary skill in the art can understand that all or part of the steps for implementing the above-mentioned embodiments of the detection method based on the computing power of the intelligent computing center can be completed by hardware related to program instructions, and the program can be stored in a readable medium.

[0214] The present invention also provides a readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it can implement any of the above Figure 1 corresponding steps in the embodiments of the detection method based on the computing power of the intelligent computing center and can achieve the same technical effects. To avoid repetition, it is not elaborated here. The storage medium includes, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disc, etc.

[0215] The terms "first", "second", etc. in the present invention are used to distinguish similar objects and do not necessarily describe a specific order or sequence. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices. In addition, in the present invention, "and / or" is used to represent at least one of the connected objects. For example, A and / or B and / or C represents seven cases including A alone, B alone, C alone, A and B both present, B and C both present, A and C both present, and A, B and C all present.

[0216] It should be noted that in this article, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not clearly listed, or also includes elements inherent to this process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or device including that element.

[0217] Through the description of the above embodiments, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal (which can be a mobile phone, computer, server, air conditioner, or a second terminal device, etc.) to execute the methods of various embodiments of the present invention.

[0218] The embodiments of the present invention are described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative rather than restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit of the present invention and the scope protected by the claims, and all of them belong to the protection scope of the present invention.

Claims

1. A detection method based on the computing power of an intelligent computing center, the method comprising: Step S1, receiving the input static planning information, where the static planning information is used to indicate the business processing logic of the target service; Step S2, pre-detecting the static planning information based on the computing power of the intelligent computing center to obtain pre-detection information, where the pre-detection information is at least used to indicate whether an execution exception will occur in the pipeline stage of the static planning information.

2. The method according to claim 1, characterized in that, The step S2 includes: Step S21, based on the computing power of the intelligent computing center, in the multiple step data bodies arranged in order included in the static planning information, checking whether the data format of each step data body matches the set standard data format to obtain a format check result, where the pre-detection information includes the format check result, and the step data body is used to indicate one business processing step among the multiple business processing steps corresponding to the business processing logic, and the business processing steps corresponding to different step data bodies are different.

3. The method according to claim 2, wherein The step S2 further includes: Step S22, when the format check result indicates that the data format of each step data body matches the standard data format, based on the computing power of the intelligent computing center, checking whether each step data body matches the set keywords to obtain a keyword check result, where the pre-detection information further includes the keyword check result.

4. The method according to claim 3, wherein The step S2 further includes: Step S23, when the keyword check result indicates that each step data body includes the keywords, based on the computing power of the intelligent computing center, checking whether the placeholder identifiers included in each step data body match the set placeholder identification rules to obtain a placeholder identification check result, where the pre-detection information further includes the placeholder identification check result.

5. The method according to claim 4, wherein The step S2 further includes: Step S24, when the placeholder identification check result indicates that the placeholder identifiers included in each step data body match the placeholder identification rules, based on the computing power of the intelligent computing center, checking whether the output data referenced by each step data body matches the set reference rules to obtain an output reference check result, where the pre-detection information further includes the output reference check result.

6. The method according to any one of claims 1 to 5, characterized in that, When the pre-detection information indicates that an execution exception will occur in the pipeline stage of the static planning information, the pre-detection information is further used to indicate: among the multiple step data bodies arranged in order included in the static planning information, the defective step data body that causes the execution exception, and the defect information corresponding to the defective step data body.

7. The method according to any one of claims 1-5, characterized in that, The static planning information includes multiple step data bodies arranged in order and satisfying the set prompt rules, and the prompt rules include at least one of the following: The position of each step data body in the static planning information is indicated based on the set data mark; The input information of each step data body is encapsulated as a placeholder identifier based on the set placeholder identification rules; The business processing method corresponding to the last sorted step data body among the multiple step data bodies is a set standard method, and the standard method includes an output method, an alarm method, and a large model method.

8. The method according to claim 1, characterized in that, After step S2, the method further includes: Step S3: Based on the computing power of the intelligent computing center, when the pre-detection information indicates that the static planning information will not cause execution exceptions in the pipeline stage, extract at least one planning object from the static planning information, where the planning object includes the business device and / or business data corresponding to the static planning information; Step S4: Based on the computing power of the intelligent computing center, calculate the association degree between each of the at least one planning object and the target business in the at least one planning object to form association prompt information, and the association prompt information is at least used to indicate whether the static planning information includes a planning object whose association degree with the target business is less than or equal to a set association threshold.

9. The method according to claim 1, characterized in that After step S2, the method further includes: Step S5: Based on the computing power of the intelligent computing center, perform a feasibility analysis on the business process indicated by the static planning information to obtain a feasibility analysis result, and the business process is generated based on the step description text of each step data body in the ordered multiple step data bodies included in the static planning information.

10. A detection device based on the computing power of an intelligent computing center, characterized in that, The device includes: A receiving module, configured to receive the input static planning information, where the static planning information is used to indicate the business processing logic of the target business; An inspection module, configured to inspect the static planning information based on the computing power of the intelligent computing center to obtain pre-detection information, and the pre-detection information is at least used to indicate whether the static planning information will cause execution exceptions in the pipeline stage.

11. An electronic device, characterized in that, Including: A processor, a memory, and a program stored on the memory and executable on the processor, and when the program is executed by the processor, the steps of the detection method based on the computing power of the intelligent computing center according to any one of claims 1 to 9 are implemented.

12. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is executed by the processor, the steps of the detection method based on the computing power of the intelligent computing center according to any one of claims 1 to 9 are implemented.

13. A computer program product, characterized in that, Including computer instructions, and when the computer instructions are executed by the processor, the steps of the detection method based on the computing power of the intelligent computing center according to any one of claims 1 to 9 are implemented.

Images