Batch task processing method based on large model

Abstract

The invention provides a large-model-based batch task processing method and device, and the method comprises the steps: recognizing to-be-processed parameters in user input data through a large model, and determining the number of vehicles according to the to-be-processed parameters; under the condition that the number of the vehicles reaches the preset asynchronous order of magnitude, intercepting logic is inserted through a method interceptor before the offline processing method is executed so as to execute interception on the offline processing method; generating batch processing tasks according to the to-be-processed parameters; wherein the batch processing task comprises a plurality of subtasks; each subtask is used for processing data of one vehicle; and performing parallel processing on each sub-task in the batch processing task through an offline processing method, and returning processing information of the batch processing task to the user side through the large model. Compared with a mode that a large model funtioncall can only carry out callback processing on vehicles one by one in the processing process, the method has the advantages that the processing time is short and the processing mode is efficient by simultaneously executing a plurality of sub-tasks and processing the data of one vehicle by each sub-task.

Description

Technical Field

[0001] The present invention relates to the technical field of task processing, and in particular, to a method for batch task processing based on a large model. Background Art

[0002] More and more industries are using large models for task processing. For batch tasks, they are generally processed by batch calling of large model function calls. During the execution of large model function calls, callbacks can only be performed one by one. After processing license plate A, then process license plate B, and then process license plate C. In the serial processing method of one vehicle after another according to license plate A, license plate B, and license plate C, the overall processing time is relatively long. For scenarios with many queuing users and the need for real-time interaction, the waiting time for users is relatively long, and timeout errors often occur. Summary of the Invention

[0003] The present invention provides a method for batch task processing based on a large model, which is used to solve the defect that the waiting time of users is relatively long and timeout errors often occur in the prior art, and realizes efficient processing of batch tasks.

[0004] The present invention provides a method for batch task processing based on a large model, including the following steps: Identify the parameters to be processed in the user input data through the large model, and determine the number of vehicles according to the parameters to be processed; When the number of vehicles reaches the preset asynchronous order of magnitude, insert interception logic before the execution of the offline processing method through a method interceptor to intercept the offline processing method; Generate batch processing tasks according to the parameters to be processed; wherein, the batch processing tasks include multiple subtasks; each subtask is used to process the data of one vehicle; Parallelly process each subtask in the batch processing task through the offline processing method, and return the processing information of the batch processing task to the user side through the large model.

[0005] According to the method for batch task processing based on a large model provided by the present invention, identifying the parameters to be processed in the user input data through the large model includes: Identify the vehicle information in the user input data through the large model; wherein, the vehicle information includes at least one license plate number; When the vehicle information is a license plate number, use the license plate number as the parameter to be processed; When the vehicle information includes multiple license plate numbers, use the multiple license plate numbers separated by commas as the parameter to be processed.

[0006] A method for batch task processing based on a large model provided by the present invention. When the batch processing task is an asynchronous task, each subtask in the batch processing task is processed in parallel by an offline processing method, and the processing information of the batch processing task is returned to the user side through the large model, including: Each subtask in the batch processing task is processed in parallel by an offline processing method; The information that the batch processing task is being executed is returned to the user side through the large model; After all the batch processing tasks are executed, the information that the batch processing task has been executed is returned to the user side through the large model.

[0007] A method for batch task processing based on a large model provided by the present invention. A batch processing task is generated according to the number of vehicles in the parameters to be processed, including: Multiple subtasks in the batch processing task are generated according to the number of vehicles in the parameters to be processed; Record information of the batch processing task is created according to the parameters to be processed; wherein, the record information includes: user input data, the number of subtasks, the execution status of the subtasks, user information, and progress information.

[0008] A method for batch task processing based on a large model provided by the present invention. While each subtask in the batch processing task is processed in parallel by an offline processing method, the method further includes: The progress information is updated according to the execution progress of each subtask.

[0009] The present invention also provides a device for batch task processing based on a large model, including the following modules: A parameter identification module, configured to identify the parameters to be processed in the user input data through the large model, and determine the number of vehicles according to the parameters to be processed; An interception module, configured to insert interception logic before the execution of the offline processing method through a method interceptor to intercept the offline processing method when the number of vehicles reaches a preset asynchronous order of magnitude; A batch processing task generation module, configured to generate a batch processing task according to the parameters to be processed; wherein, the batch processing task includes multiple subtasks; each subtask is used to process the data of one vehicle; An execution module, configured to process each subtask in the batch processing task in parallel by an offline processing method, and return the processing information of the batch processing task to the user side through the large model.

[0010] For a device for batch task processing based on a large model provided by the present invention, the parameter identification module includes: A vehicle information recognition sub-module, which is used to recognize vehicle information in the user input data through a large model; wherein, the vehicle information includes at least one license plate number; A first parameter generation sub-module, which is used to use the license plate number as the parameter to be processed when the vehicle information is a license plate number; A second parameter generation sub-module, which is used to use multiple license plate numbers separated by commas as the parameter to be processed when the vehicle information includes multiple license plate numbers.

[0011] According to an apparatus for batch task processing based on a large model provided by the present invention, in the case where the batch processing task is an asynchronous task, an execution module is used to parallelly process each sub-task in the batch processing task through an offline processing method; return the information that the batch processing task is being executed to the user side through the large model; after all the batch processing tasks are executed, return the information that the batch processing task has been executed to the user side through the large model.

[0012] According to an apparatus for batch task processing based on a large model provided by the present invention, a batch processing task generation module is used to generate multiple sub-tasks in the batch processing task according to the parameter to be processed; create record information of the batch processing task according to the parameter to be processed; wherein, the record information includes: user input data, the number of sub-tasks, the execution status of the sub-tasks, user information and progress information.

[0013] According to an apparatus for batch task processing based on a large model provided by the present invention, the execution module is further used to update the progress information according to the execution progress of the sub-tasks.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the method for batch task processing based on a large model as described in any one of the above.

[0015] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the method for batch task processing based on a large model as described in any one of the above.

[0016] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the method for batch task processing based on a large model as described in any one of the above.

[0017] The method for batch task processing based on a large model provided by the present invention identifies the parameters to be processed in the user input data through the large model, and determines the number of vehicles according to the parameters to be processed; when the number of vehicles reaches the preset asynchronous order of magnitude, an interception logic is inserted before the execution of the offline processing method through a method interceptor to intercept the offline processing method; a batch processing task is generated according to the parameters to be processed; wherein, the batch processing task includes a plurality of subtasks; each subtask is used to process the data of one vehicle; each subtask in the batch processing task is processed in parallel through the offline processing method, and the processing information of the batch processing task is returned to the user side through the large model. Compared with the way that the large model function call can only perform callback processing vehicle by vehicle during the processing, the embodiment of the present invention processes multiple subtasks simultaneously, with each subtask processing the data of one vehicle, resulting in a shorter processing time and an efficient processing method. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 It is a schematic flowchart of a method for batch task processing based on a large model provided by the present invention.

[0020] Figure 2 It is a schematic structural diagram of a device for batch task processing based on a large model provided by the present invention.

[0021] Figure 3 It is a schematic structural diagram of an electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention with reference to the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention belong to the scope of protection of the present invention.

[0023] The following will be described in conjunction with Figures 1 - 3 describe the present invention.

[0024] Figure 1 It is one of the schematic flowcharts of a method for batch task processing based on a large model provided by the present invention, as shown in Figure 1As shown, the method includes the following: Step 101: Identify the parameters to be processed in the user input data through a large model, and determine the number of vehicles based on the parameters to be processed.

[0025] In the above Step 101, the user input data refers to the content input by the user to the platform. For example, the user input data is "Please help me process the disconnection problems of Sichuan AB2123xx, Sichuan AB5123xx, and Sichuan AB8123xx".

[0026] The parameter to be processed refers to the parameter obtained through the large model based on the vehicle information in the user input data. For example, when the user input data is "Please help me process the disconnection problems of Sichuan AB2123xx, Sichuan AB5123xx, and Sichuan AB8123xx", the parameter to be processed can be multiple license plate numbers obtained from the user input data, separated by commas, that is, "AB2123xx, Sichuan AB5123xx, AB8123xx".

[0027] The method for determining the number of vehicles based on the parameter to be processed can be to calculate the number of vehicles according to the separation method of multiple license plate numbers. For example, if multiple license plate numbers are separated by commas, the number of vehicles is the number of commas plus 1.

[0028] Optionally, identifying the parameter to be processed in the user input data in the above Step 101 includes Steps A1 to A3: Step A1: Identify the vehicle information in the user input data through a large model; where the vehicle information includes at least one license plate number.

[0029] In the above Step A1, the user input data is generally not standardized and requires the large model to perform data preprocessing. The preprocessing includes text cleaning, word segmentation, and standardization.

[0030] Text cleaning refers to removing unnecessary characters, such as punctuation marks, special symbols, and extra spaces.

[0031] Word segmentation refers to splitting the text into words or phrases, which helps with subsequent feature extraction.

[0032] Standardization refers to converting the text to lowercase, handling abbreviations and synonyms to unify the format.

[0033] After the large model preprocesses the user input data, it then performs feature extraction on the user input data. Feature extraction includes: regular expressions, named entity recognition (NER), and keyword matching.

[0034] A regular expression refers to using a regular expression to match a specific pattern of vehicle information. For example, a license plate number usually has a certain format. For example, ABC-1234.

[0035] Named Entity Recognition (NER) refers to using pre-trained NLP models (such as spaCy, NLTK, BERT, etc.) to identify named entities in text, which may include brands, models, etc.

[0036] Keyword matching refers to defining a set of keywords, for example, brand names and model names, and then searching for these keywords in the text.

[0037] Step A2: When the vehicle information is a license plate number, use the license plate number as the parameter to be processed.

[0038] Step A3: When the vehicle information includes multiple license plate numbers, use the multiple license plate numbers separated by commas as the parameter to be processed.

[0039] In the above Step A2 and Step A3, for example, when the obtained vehicle information is "License Plate A", use "License Plate A" as the parameter to be processed. When the obtained vehicle information is "License Plate A License Plate B License Plate C", use "License Plate A, License Plate B, License Plate C" as the parameter to be processed.

[0040] Step 102: When the number of vehicles reaches the preset asynchronous order of magnitude, insert interception logic before the execution of the offline processing method through a method interceptor to intercept the offline processing method.

[0041] In the above Step 102, the preset asynchronous order of magnitude is set according to actual needs. For example, the preset asynchronous order of magnitude is 2.

[0042] A method interceptor refers to a technology for method calls or method executions that can insert additional logic before, after, or when an exception occurs in a method call.

[0043] Task processing is usually carried out on a single vehicle basis, and only one license plate number can be accepted each time a task is executed. For example, the method for offline problem processing can be handle_offline. In the embodiments of the present invention, when batch task processing is required, the handle_offline method is declared as a batch method, intercepted by a method interceptor for method callback, a batch processing task is generated, multiple license plate numbers are split into multiple subtasks, and the handle_offline method is called and processed in each subtask.

[0044] Step 103: Generate a batch processing task based on the parameter to be processed; among them, the batch processing task includes multiple subtasks; each subtask is used to process the data of one vehicle.

[0045] In the above Step 103, generating a batch processing task involves the definition of method parameters and the definition of batch configuration.

[0046] Exemplarily, the parameter definition of the handle_offline method supports multiple license plate numbers, separated by commas: truck_nos: str = Field(description="License plate number, if multiple license plate numbers are recognized, use commas to separate", title="truck_nos").

[0047] The batch configuration of the defined method includes: Batch field: batch_field = "truck_no"; Batch type: batch_type = "Offline problem handling"; Maximum quantity: batch_max_size = 100; Number of asynchronous tasks: batch_async_size = 11.

[0048] When this application executes a batch processing task, it only makes one callback for the large model function call. At the same time, it recognizes and processes multiple license plate numbers, and generates subtasks for each license plate number respectively. Each subtask is executed independently and does not affect each other.

[0049] However, the large model function call itself can only make callback processing for each license plate number one by one. After processing license plate A, it then processes license plate B, and then license plate C, which belongs to multiple callbacks.

[0050] Exemplarily, the batch processing task needs to process license plate A, license plate B, and license plate C. After an error occurs during the processing of license plate A, license plate B and license plate C can continue to execute. At the same time, the task progress will be continuously updated during the task execution. The user can check. However, after an error occurs during the processing of license plate A by the large model function call, the entire processing process stops. It cannot continue to process license plate B and license plate C. The task execution process cannot be tracked, and it does not support querying the processing situation and error situation of the vehicle, etc.

[0051] When this application detects that batch processing is required, it generates subtasks for each license plate number respectively and processes them in parallel. After the processing is completed, the results are summarized. If the large model takes 1 minute to process a single license plate number, the overall processing time of the batch processing task is the result of 1 minute plus a few seconds of result summarization time. However, the large model function call processes each vehicle serially according to license plate A, license plate B, and license plate C. The overall processing time is relatively long. If the large model takes 1 minute to process a single license plate number, the overall processing time is more than 3 minutes.

[0052] When the large model function call processes multiple vehicles, multiple callbacks are required. There will be a situation where only a part of the vehicles are processed. For example, when the large model function call processes three vehicles, it only makes one callback. In the embodiment of this application, the large model extracts the information of the batch of vehicles to be processed and converts it into a subtask mode, and only needs to make one callback to process all vehicles, which is more stable than multiple callbacks.

[0053] Step 104: Parallelly process each subtask in the batch processing task through an offline processing method, and return the processing information of the batch processing task to the user side through the large model.

[0054] In the above step 104, the processing information includes the information generated at the start, during the execution, and at the end of the execution of the batch processing task. The large model is used to summarize and generate the processing results of the batch task and notify the user. Specifically, the results of all subtasks and the notification template can be told to the large model, and the large model is allowed to polish and output.

[0055] Optionally, when the batch processing task is an asynchronous task, step 104 includes steps B1 to B3: Step B1: Parallelly process each subtask in the batch processing task through an offline processing method.

[0056] Step B2: Return the information that the batch processing task is being executed to the user side through the large model.

[0057] Step B3: After all the batch processing tasks are executed, return the information that the batch processing task has been executed to the user side through the large model.

[0058] In the above steps B2 to B3, the information that the batch processing task is being executed refers to the feedback information generated when the batch processing task starts to be executed, or the feedback information generated during the execution of the batch processing task. The feedback information is mainly used to represent the execution situation of the batch processing task.

[0059] Exemplarily, when the batch processing task starts to execute, the information that the batch processing task is being executed can be "It has been determined that you need to process multiple vehicles. Currently, entering the batch task processing mode. You will be notified after the processing is completed." During the execution of the batch processing task, the information that the batch processing task is being executed can be "Currently processing license plate A, license plate B, and license plate C, and the processing progress is 70%."

[0060] The difference between synchronous tasks and asynchronous tasks is that synchronous tasks need to wait for all subtasks to be processed before giving feedback to the user, and the user's waiting time is relatively long. Asynchronous tasks first give a reply to the user. Exemplarily, "It has been determined that you need to process multiple vehicles. Currently, entering the batch task processing mode. You will be notified after the processing is completed." After all subtasks are processed, the user will be notified again.

[0061] Optionally, the above step 103 includes steps C1 to C2: Step C1: Generate multiple subtasks in the batch processing task according to the number of vehicles in the parameters to be processed.

[0062] Step C2: Create record information for the batch processing task; wherein, the record information includes: user input data, the number of subtasks, the execution status of the subtasks, user information, and progress information.

[0063] In the above step C2, creating the record information for the batch processing task is to track and manage the status, progress, results, and any relevant metadata of these tasks. This helps to ensure the traceability of the tasks, facilitates debugging in case of problems, and provides a historical reference for future task executions.

[0064] The execution status of the subtasks refers to the total number of processed subtasks, the number of successfully processed subtasks, and the number of failed processed subtasks.

[0065] Optionally, when the above step 103 includes steps C1 to C2, while executing the above step 104, the method further includes step D: Step D: Update the progress information according to the execution progress of each subtask.

[0066] In the above step D2, the progress information provides real-time feedback on the task execution status for the user or system administrator. The progress information usually includes the proportion of the completed part of the subtasks to the total task volume, as well as any relevant status updates.

[0067] An embodiment of the present invention provides a method for batch task processing based on a large model. The method uses the large model to identify the parameters to be processed in the user input data, and determines the number of vehicles according to the parameters to be processed. When the number of vehicles reaches a preset asynchronous order of magnitude, an interception logic is inserted before the execution of the offline processing method through a method interceptor to intercept the offline processing method. A batch processing task is generated according to the parameters to be processed, where the batch processing task includes multiple subtasks, and each subtask is used to process the data of one vehicle. The offline processing method is used to process each subtask in the batch processing task in parallel, and the processing information of the batch processing task is returned to the user side through the large model. Compared with the way that the large model function call can only perform callback processing for each vehicle one by one during the processing, the embodiment of the present invention processes multiple subtasks simultaneously, with each subtask processing the data of one vehicle, resulting in a shorter processing time and an efficient processing method.

[0068] The following describes a device for batch task processing based on a large model provided by the present invention. The device for batch task processing based on a large model described below can be correspondingly referred to the method for batch task processing based on a large model described above.

[0069] Figure 2 An example of a schematic diagram for batch task processing based on a large model is as Figure 2 As shown, the present invention also provides a device for batch task processing based on a large model, including the following modules: A parameter identification module 201, configured to identify the parameters to be processed in the user input data through the large model, and determine the number of vehicles according to the parameters to be processed; An interception module 202, configured to, when the number of vehicles reaches a preset asynchronous order of magnitude, insert an interception logic before the execution of the offline processing method through a method interceptor to intercept the offline processing method; A batch processing task generation module 203, configured to generate a batch processing task according to the parameters to be processed, where the batch processing task includes multiple subtasks, and each subtask is used to process the data of one vehicle; An execution module 204, configured to process each subtask in the batch processing task in parallel through the offline processing method, and return the processing information of the batch processing task to the user side through the large model.

[0070] Optionally, the parameter identification module 201 includes: A vehicle information identification sub-module, configured to identify the vehicle information in the user input data through the large model, where the vehicle information includes at least one license plate number; A first parameter generation sub-module, configured to, when the vehicle information is a license plate number, use the license plate number as the parameter to be processed; A second parameter generation sub-module, configured to use a comma to separate multiple license plate numbers as parameters to be processed when the vehicle information includes multiple license plate numbers.

[0071] Optionally, when the batch processing task is an asynchronous task, the execution module 204 is configured to parallelly process each sub-task in the batch processing task through an offline processing method, and return the information that the batch processing task is being executed to the client through a large model; after all the batch processing tasks are executed, return the information that the batch processing task has been executed to the client.

[0072] Optionally, the batch processing task generation module 203 is configured to generate multiple sub-tasks in the batch processing task according to the number of vehicles in the parameter to be processed; create record information for the batch processing task according to the parameter to be processed; wherein, the record information includes: user input data, the number of sub-tasks, the execution status of sub-tasks, user information, and progress information.

[0073] Optionally, the execution module 204 is configured to execute the batch processing task through an offline processing method; update the progress information according to the execution progress of each sub-task.

[0074] An embodiment of the present invention provides an apparatus for batch task processing based on a large model, which identifies parameters to be processed in user input data through the large model, determines the number of vehicles according to the parameters to be processed; when the number of vehicles reaches a preset asynchronous order of magnitude, inserts interception logic before the execution of the offline processing method through a method interceptor to perform interception on the offline processing method; generates a batch processing task according to the parameters to be processed; wherein, the batch processing task includes multiple sub-tasks; each sub-task is used to process the data of one vehicle; parallelly process each sub-task in the batch processing task through an offline processing method, and return the processing information of the batch processing task to the client through the large model. Compared with the way that the large model function call can only perform callback processing for each vehicle one by one during the processing, the embodiment of the present invention processes multiple sub-tasks simultaneously, with each sub-task processing the data of one vehicle, so the processing time is shorter and the processing method is more efficient.

[0075] Figure 3 Schematically shows an entity structure diagram of an electronic device, as Figure 3 shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 complete mutual communication through the communication bus 840. The processor 810 can call logical instructions in the memory 830 to execute a method for batch task processing based on a large model.

[0076] In addition, when the logical instructions in the above-mentioned memory 830 can be implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0077] A schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. By identifying the parameters to be processed in the user input data, the number of vehicles is determined according to the parameters to be processed; in the case where the number of vehicles reaches a preset asynchronous order of magnitude, an offline processing method is intercepted through a method interceptor; a batch processing task is generated according to the number of vehicles in the parameters to be processed; wherein, the batch processing task includes multiple subtasks; each subtask is used to process the data of one vehicle; the batch processing task is executed through the offline processing method, and the processing information of the batch processing task is returned to the user side. Compared with the way that the large model function call can only perform callback processing for each vehicle one by one during the processing, the embodiment of the present invention processes in a shorter time and with higher efficiency by simultaneously executing multiple subtasks, with each subtask processing the data of one vehicle.

[0078] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a method for batch task processing based on a large model provided by the above-mentioned various methods.

[0079] A computer program product provided by an embodiment of the present invention identifies processing parameters to be processed in user input data through a large model, determines the number of vehicles according to the processing parameters to be processed; in the case where the number of vehicles reaches a preset asynchronous order of magnitude, an interception logic is inserted before the execution of the offline processing method through a method interceptor to intercept the offline processing method; generates a batch processing task according to the processing parameters to be processed; wherein, the batch processing task includes multiple subtasks; each subtask is used to process the data of one vehicle; the offline processing method is used to process each subtask in the batch processing task in parallel, and the processing information of the batch processing task is returned to the user side through the large model. Compared with the method in which the large model function call can only perform callback processing vehicle by vehicle during the processing process, the embodiment of the present invention processes multiple subtasks simultaneously, and each subtask processes the data of one vehicle, so the processing time is shorter and the processing method is efficient.

[0080] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it is configured to execute a method for batch task processing based on a large model provided by the above-mentioned various methods.

[0081] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0082] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solutions, or the part that contributes to the prior art, can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for batch task processing based on a large model, characterized in that: include: Identify the parameters to be processed in the user input data through the large model, and determine the number of vehicles according to the parameters to be processed; When the number of vehicles reaches a preset asynchronous order of magnitude, an interception logic is inserted before the offline processing method is executed by a method interceptor to intercept the offline processing method; Generate a batch processing task according to the parameters to be processed; wherein the batch processing task includes a plurality of subtasks; each of the subtasks is used to process data of one vehicle; Each of the subtasks in the batch processing task is processed in parallel by the offline processing method, and the processing information of the batch processing task is returned to the user end through the large model.

2. The method for batch task processing based on a large model according to claim 1, characterized in that: The step of identifying the parameters to be processed in the user input data by using the large model includes: Recognize vehicle information in the user input data through the large model; wherein the vehicle information includes at least one license plate number; In the case where the vehicle information is a license plate number, taking the license plate number as a parameter to be processed; In the case that the vehicle information includes a plurality of license plate numbers, the plurality of license plate numbers are separated by commas and used as parameters to be processed.

3. The method for batch task processing based on a large model according to claim 1, characterized in that: In the case where the batch processing task is an asynchronous task, the processing of each subtask in the batch processing task in parallel by the offline processing method, and returning the processing information of the batch processing task to the user end by the large model, includes: Process each of the subtasks in the batch processing task in parallel by using the offline processing method; The information that the batch processing task is being executed is returned to the user end through the large model; After all the batch processing tasks are executed, the information that the batch processing tasks are executed is returned to the user end through the large model.

4. The method for batch task processing based on a large model according to claim 1, characterized in that: The generating of batch processing tasks according to the parameters to be processed includes: Generate multiple subtasks in the batch processing task according to the number of vehicles in the parameters to be processed; The record information of the batch processing task is created according to the parameters to be processed; wherein the record information includes: user input data, the number of subtasks, the execution status of subtasks, user information and progress information.

5. The method for batch task processing based on a large model according to claim 4, characterized in that: While processing each of the subtasks in the batch processing task in parallel by the offline processing method, the method further includes: The progress information is updated according to the execution progress of each of the subtasks.

6. A device for batch task processing based on a large model, characterized in that: include: A parameter identification module, used to identify the parameters to be processed in the user input data through a large model, and determine the number of vehicles according to the parameters to be processed; An interception module, used for inserting interception logic before the execution of the offline processing method through a method interceptor to intercept the offline processing method when the number of vehicles reaches a preset asynchronous order of magnitude; A batch processing task generation module, used to generate a batch processing task according to the parameters to be processed; wherein the batch processing task includes a plurality of subtasks; each of the subtasks is used to process the data of one vehicle; The execution module is used to process each of the subtasks in the batch processing task in parallel through the offline processing method, and return the processing information of the batch processing task to the user end through the large model.

7. The device for batch task processing based on a large model according to claim 6, characterized in that: The parameter identification module comprises: The vehicle information recognition submodule is used to recognize the vehicle information in the user input data through the large model; wherein the vehicle information includes at least one license plate number; A first parameter generating submodule, for, when the vehicle information is a license plate number, using the license plate number as a parameter to be processed; The second parameter generating submodule is used to separate the multiple license plate numbers with commas as parameters to be processed when the vehicle information includes multiple license plate numbers.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, a method for batch task processing based on a large model as described in any one of claims 1 to 5 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a method for batch task processing based on a large model as described in any one of claims 1 to 5 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, a method for batch task processing based on a large model as described in any one of claims 1 to 5 is implemented.

Images