A model migration method and system based on data classification
Through the model migration method based on data classification, the problem of lack of standardized storage and interpretation in model sharing is solved, and the rapid migration and sharing of model data is realized, cost and dependence are reduced, and the efficiency and feasibility of model application are improved.
Patent Information
- Application Number
- CN202111277206.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-10-29
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2041-10-29
AI Technical Summary
现有技术中模型共享过程中缺乏标准化的存储方法和解释文件,导致模型移植和复用困难,高成本和重复开发问题,且依赖于模型建设人员,难以实现多区域部署和人员离职后的模型 共享。
The model migration method based on data classification is adopted, by setting data classification and verification information, matching feature information, updating data and analyzing documents using processing templates, rapid migration and sharing of model data is achieved, and model understanding and data processing processes are reduced.
It realizes the rapid and efficient model sharing, reduces time and cost, simplifies the model sharing process, reduces dependence on model personnel, and improves the feasibility of model application scenarios.
Smart Images

Figure CN113837320B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of big data processing, and in particular to a model migration method and system based on data classification. Background Art
[0002] At present, each entity has its own unique model construction methods, model construction habits, and differentiated data content and data formats. However, when the model construction is completed and put into use, there is a lack of relatively standardized model storage methods and model interpretation files, and modeling work is a relatively high technical threshold. This leads to certain difficulties in data preprocessing and model application during model migration and model reuse, which leads to high costs in the model migration process and repeated model development. In addition, during the model sharing process, the integration and processing of raw data into input data that meets the model requirements requires a series of operations such as data mapping, data integration, data cleaning, data preprocessing, and feature engineering, which is complicated and time-consuming.
[0003] The model sharing method of the existing technology mainly adopts the method of exporting the model after modeling, and then in the application scenario, the model deployment personnel first understand the model, and then prepare the application scenario data according to the requirements of each model, and then substitute it into the model for calculation. Every time the model is shared, it is necessary to write code for data mapping, data preprocessing, feature engineering and other processing according to each application scenario, and then apply the model. The data processing work is complicated and redundant; when sharing the model, it is necessary to start with model understanding, programmatic data processing, interface organization and debugging, etc., which is a large workload and time-consuming. At the same time, due to the lack of detailed and easy-to-understand model explanation documents, the model sharing process is too dependent on its model construction personnel. Once there is a situation of simultaneous deployment in multiple regions and multiple points or the resignation of model construction personnel, it is easy to make the model sharing process difficult to achieve. Summary of the invention
[0004] In order to address the deficiencies of the prior art, the present invention proposes a model migration method and system based on data classification to achieve rapid model-related information transmission. The data sharing solution based on shared data, common data, and personalized data saves time and cost, achieves fast and efficient model sharing, reduces the processes of model understanding and data processing, and greatly reduces the time-consuming model sharing work. From the three aspects of model sharing environment, model parsing document, and data sharing system, the problems encountered in the model sharing process in terms of model understanding, data processing, and unified standards and platforms are comprehensively solved, which can greatly improve the application scenarios of model sharing. The entire model sharing process mainly involves operations such as template import and export, model import and export, data import and benchmarking, and data export. It is simple and easy to use, reducing the dependence of model sharing work on modeling personnel.
[0005] To achieve the above objectives, the technical solutions adopted by the present invention include:
[0006] A model migration method based on data classification, characterized by comprising:
[0007] S1. Setting data classification and first verification information for distinguishing different data classifications and processing templates corresponding to different data classifications;
[0008] S2. Obtain model data and extract feature information from each item of model data;
[0009] S3, matching the feature information with the first verification information, and marking the model data as a corresponding data classification according to the matching result;
[0010] S4, grouping the model data into a set of data to be processed according to data classification;
[0011] S5. Use the corresponding processing template to update the data set to be processed to obtain the migratable data set;
[0012] S6, combining various transferable data sets to form transferable model data;
[0013] S7. Export the model to be migrated and extract the model parsing document;
[0014] S8. Update the model parsing document according to the first verification information and the processing template;
[0015] S9. Migrate the migratable model data and the model combination with the updated model parsing document.
[0016] Furthermore, the data classification includes shared data, common data and individual data;
[0017] The shared data includes model data that can be directly obtained and used through public channels;
[0018] The common data include model data that cannot be directly obtained through public channels but can be used directly after being obtained;
[0019] The personality data includes model data that cannot be used directly and must be processed.
[0020] Furthermore, the first verification information includes data source, data level, data usage authority and data storage format.
[0021] Furthermore, the processing template includes:
[0022] Shared data processing template, including shared data features and shared data import path;
[0023] Common data processing template, including common data features and common data interfaces;
[0024] The personalized data processing template includes personalized data features, personalized data collection interface, personalized data import interface and personalized data mapping conversion template.
[0025] Furthermore, the step S3 comprises the following sub-steps:
[0026] S31, matching the feature information with the first verification information, verifying whether the model data belongs to the shared data, and marking the model data that passes the verification as the shared data;
[0027] S32, when the verification model data does not belong to the shared data, matching the feature information with the first verification information, verifying whether the model data belongs to the common data, and marking the model data that passes the verification as the common data;
[0028] S33. When the verification model data does not belong to common data, the model data is marked as personalized data.
[0029] Furthermore, the step S5 comprises the following sub-steps:
[0030] S51, recording the shared data features of the model data marked as shared data and the corresponding shared data import path according to the shared data processing template, and obtaining a shared migratable data set, wherein the shared migratable data set does not include the model data marked as shared data itself;
[0031] S52, recording the common data features of the model data marked as common data and the corresponding common data interfaces according to the common data processing template, and obtaining a common transferable data set, wherein the common transferable data set does not include the model data marked as common data itself;
[0032] S53. According to the personality data processing template, the personality data features of the model data marked as personality data and the corresponding personality data import interface are recorded, the model data is collected using the personality data collection interface, and the model data is converted into pre-processed data suitable for the personality data import interface using the personality data mapping conversion template to obtain a personality transferable data set, wherein the personality transferable data set includes the pre-processed data after the model data marked as personality data is converted.
[0033] Furthermore, the step S8 comprises:
[0034] Based on data classification, information on shared data features, shared data import paths, common data features, common data interfaces, personalized data features, personalized data collection interfaces, personalized data import interfaces and personalized data mapping conversion templates is added to the model parsing document.
[0035] The present invention also relates to a model migration system based on data classification, characterized in that it comprises:
[0036] A data classification module, used to classify and mark the model data using the first verification information;
[0037] A data processing module is used to process model data of different categories respectively using processing templates;
[0038] A document updating module, used for updating the model parsing document according to the first verification information and the processing template;
[0039] The migration module is used to migrate the processed model data and the model combination with the updated model parsing document.
[0040] The present invention also relates to a computer-readable storage medium, characterized in that a computer program is stored on the storage medium, and the computer program implements the above method when executed by a processor.
[0041] The present invention also relates to an electronic device, characterized in that it comprises a processor and a memory;
[0042] The memory is used to store model data and model parsing documents;
[0043] The processor is used to execute the above method by calling model data and model parsing documents.
[0044] The beneficial effects of the present invention are:
[0045] The model migration method and system based on data classification described in the present invention are used to realize rapid model-related information transmission. The data sharing solution based on shared data, common data, and personalized data saves time and cost, realizes fast and efficient model sharing, reduces the processes of model understanding and data processing, and greatly reduces the time-consuming model sharing work. From the three aspects of model sharing environment, model parsing document, and data sharing system, the problems encountered in the model sharing process in terms of model understanding, data processing, and unified standards and platforms are comprehensively solved, which can greatly improve the application scenarios of model sharing. The entire model sharing process mainly involves operations such as template import and export, model import and export, data import and benchmarking, and data export. It is simple and easy to use, reducing the dependence of model sharing work on modeling personnel. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] Figure 1 The figure is a flow chart of the model migration method based on data classification of the present invention.
[0047] Figure 2 This is a schematic diagram of the structure of the model migration system based on data classification of the present invention. DETAILED DESCRIPTION
[0048] In order to more clearly understand the content of the present invention, it will be described in detail with reference to the drawings and embodiments.
[0049] The first aspect of the present invention relates to a process of steps as follows Figure 1 The model migration method based on data classification shown includes:
[0050] S1. Set data classification and first verification information for distinguishing different data classifications and processing templates corresponding to different data classifications. The data classification includes shared data, common data and individual data, wherein shared data includes model data that can be directly obtained and used through public channels, such as data placed on public network platforms by various companies or government departments at all levels, which can be downloaded and obtained, such as national statistical indicators, etc., generally referring to small-scale data; common data includes model data that cannot be directly obtained through public channels, but can be directly used after obtaining, such as external data that can be jointly searched or accessed by various companies or government departments at all levels. This part of data is provided by special government departments or companies and needs to be obtained by application, such as State Council data, market supervision data and central bank credit data, etc., generally referring to large-scale data or paid data; individual data includes model data that cannot be used directly and must be processed, such as unique data within various companies or government departments at all levels. This part of data comes from within each company or government department and represents the internal situation of each company or government department. Such as internal operating data, internal sales data, internal personnel data and regional data, etc. The first verification information includes data source, data level, data usage rights and data storage format. Through the first verification information, it is possible to fully identify which category the model data belongs to. The processing templates correspond to three data categories, including: shared data processing templates, including shared data features and shared data import paths. When sharing the model, this part of the data can be exported and used together with the model package and data mapping templates, data processing templates and other files to reduce the data access / data search process when the model is shared; common data processing templates, including common data features and common data interfaces. This part of the data should provide an available interface so that users can correctly obtain it according to their own permissions after migration; personalized data processing templates, including personalized data features, personalized data collection interfaces, personalized data import interfaces and personalized data mapping conversion templates. This part of the data has different data standards, different field definitions, and data types. There will be significant differences.
[0051] S2. Obtain model data and extract characteristic information from each item of model data. The items contained in the characteristic information should at least be the same as the items of the first verification information so as to perform matching verification.
[0052] S3. Match the feature information with the first verification information, and mark the model data as the corresponding data classification according to the matching result. Specifically, according to different data classifications, the following steps are preferably performed: S31. Match the feature information with the first verification information, verify whether the model data belongs to shared data, and mark the model data that has passed the verification as shared data; S32. When it is verified that the model data does not belong to shared data, match the feature information with the first verification information, verify whether the model data belongs to common data, and mark the model data that has passed the verification as common data; S33. When it is verified that the model data does not belong to common data, mark the model data as personalized data. That is, it is preferred to judge whether the data model belongs to shared data and common data, and for those that are determined not to belong to shared data and common data, they are directly processed and marked as personalized data.
[0053] S4. Group the model data according to data classification to form a data set to be processed.
[0054] S5. Use the corresponding processing template to update the data set to be processed to obtain the migratable data set. Specifically, according to different data classifications, the following steps are preferably performed: S51. Based on the shared data processing template, record the shared data features of the model data marked as shared data and the corresponding shared data import path to obtain a shared migratable data set, wherein the shared migratable data set does not contain the model data itself marked as shared data; S52. Based on the common data processing template, record the common data features of the model data marked as common data and the corresponding common data interface to obtain a common migratable data set, wherein the common migratable data set does not contain the model data itself marked as common data. The data interface is generally an API data interface, through which the database data can be imported in batches and continuously for subsequent data processing. Work; S53, according to the personality data processing template, record the personality data features of the model data marked as personality data and the corresponding personality data import interface, use the personality data collection interface to collect the model data and use the personality data mapping conversion template to convert the model data into pre-processed data suitable for the personality data import interface, and obtain a personality transferable data set, wherein the personality transferable data set includes the pre-processed data after the model data marked as personality data is converted. The personality data mapping conversion template mainly completes the personality data mapping task. Because the data field names, field types, etc. of each company or government department at all levels may be different, the personality data mapping conversion template helps to map this part of the personality data into standardized data that can be imported into the model. For example, when establishing model M, the feature used contains the employee education label, whose field name is 'education', field type is 'char', and field value is {A-uneducated, B-primary school, C-junior high school, D-senior high school, E-technical secondary school, F-junior college, G-bachelor's degree, H-master's degree, I-doctorate}, but the employee education field name provided by company T is 'education', field type is 'char', and field value is {1-uneducated, 2-primary school, 3-junior high school, 4-senior high school, 5-technical secondary school, 6-junior college, 7-bachelor's degree, 8-master's degree, 9-doctorate}. When model M needs to be applied to company T, the field name and field value of company T's employee education field 'education' need to be mapped to the 'education' field in model M. Using the personalized data mapping conversion template, you can clearly and conveniently visually select fields with the same meaning, and realize the comparison and mapping of fields, field types and field values. Another function of the personalized data mapping conversion template is to complete tasks such as data wide table integration, data preprocessing, feature engineering, and variable custom derivation, to assist in the rapid integration of data from different sources, and to process and derive them into model input data.Taking any model M and company T as an example, before the completed model input data is formed, the shared, common and individual data need to be processed, including merging into one table, and performing missing value filling, normalization, independent hot code, and derivation of new feature variables and other data-related processing operations. The individual data mapping transformation template saves the primary key information related to the whole table merger when building model M, as well as the missing value filling, normalization, independent hot code, and derivation of new feature variables in data preprocessing and feature engineering, and other operation information and operation processes to help realize data integration and data processing. When the data calls the individual data mapping transformation template, it will execute a series of data processing operations saved in it, and finally output the complete model application data (that is, the model input data).
[0055] S6. Combine various migratable data sets to form migratable model data.
[0056] S7. Export the model that needs to be migrated and extract the model parsing document.
[0057] S8. Update the model parsing document based on the first verification information and the processing template. Specifically, based on the data classification, add the information of shared data features, shared data import paths, common data features, common data interfaces, individual data features, individual data collection interfaces, individual data import interfaces, and individual data mapping conversion templates to the model parsing document. In order to achieve rapid model sharing, a detailed model interpretation mechanism needs to be established. This requires the model builder to provide detailed model parsing and model application guidance documents after the model is built, including a series of model construction-related contents including model data input format, data fields and field types, data processing methods, data feature engineering, model algorithms, algorithm parameters, model applicable scenarios, model usage methods, and model construction languages (C / Java / Python, etc.), as well as model application-related information such as model application interface introduction, and package the description document with the model for sharing, so as to facilitate the model user to understand and use the model.
[0058] S9. Migrate the model combination of the migratable model data and the updated model analysis document. To speed up model sharing, an operational data sharing mechanism can be established. By establishing templates, it helps save the process of data understanding and data processing, realizes the rapid processing of model input data, and promotes rapid model sharing and reuse.
[0059] Of course, in order to achieve rapid sharing of models, a unified model construction and model sharing platform can be established to help users quickly understand and use models through the use of platform services. Make full use of information technology means such as big data and artificial intelligence to establish a unified model construction platform to achieve full-process modeling services from data input, data processing, feature engineering, modeling, model reporting, model library, model application and model sharing. By using a unified model construction platform, the rapid sharing of models in the platform can be achieved. At the same time, the unified modeling platform service can also help unify users' modeling methods and modeling habits, thereby improving the understanding and application of shared models by companies and government departments at all levels. At this point, the overall business process can be summarized as follows: a. The entire model sharing process can be carried out in a unified modeling environment - the modeling platform; b. The required data of the model is divided into three categories (shared data, common data, and individual data) and processed separately; c. Export the model that needs to be shared (including model code and model parsing documents); d. Overall packaging and encapsulation to achieve maximum model sharing (direct sharing of part of the data, sharing of data processing logic, and sharing of compiled models), avoiding data sovereignty issues and achieving rapid migration of models in similar environments; e. In the new company / province / business scenario, access the data, and use the exported data mapping template and data processing template to perform data mapping, data preprocessing, feature engineering and other data processing work to form qualified model input data; f. Import the model and apply the processed model input data to the model. The model sharing document can help users understand the model and set up model access.
[0060] Another aspect of the present invention also relates to a model migration system based on data classification, the structure of which is as follows: Figure 2 As shown, including:
[0061] A data classification module, used to classify and mark the model data using the first verification information;
[0062] A data processing module is used to process model data of different categories respectively using processing templates;
[0063] A document updating module, used for updating the model parsing document according to the first verification information and the processing template;
[0064] The migration module is used to migrate the processed model data and the model combination with the updated model parsing document.
[0065] By using this system, the above-mentioned operation processing method can be executed and the corresponding technical effects can be achieved.
[0066] An embodiment of the present invention also provides a computer-readable storage medium capable of implementing all the steps in the method in the above embodiment, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, all the steps in the method in the above embodiment are implemented.
[0067] An embodiment of the present invention also provides an electronic device for executing the above method. As an implementation device of the method, the electronic device has at least a processor and a memory, and in particular, the memory stores data and related computer programs required for executing the method, such as model data and model parsing documents, etc., and the processor calls the data and programs in the memory to execute all steps of the implementation method and obtains the corresponding technical effect.
[0068] Preferably, the electronic device may include a bus architecture, the bus may include any number of interconnected buses and bridges, and the bus will include various circuits linked together by one or more processors and memories. The bus may also link together various other circuits such as peripherals, voltage regulators, and power management circuits, which are all well known in the art and are therefore not further described herein. The bus interface provides an interface between the bus and the receiver and transmitter. The receiver and transmitter may be the same component, i.e., a transceiver, providing a unit for communicating with various other systems over a transmission medium. The processor is responsible for managing the bus and general processing, while the memory may be used to store data used by the processor when performing operations.
[0069] In addition, the electronic device may further include components such as a communication module, an input unit, an audio processor, a display, and a power supply. The processor (or controller, operation control) used may include a microprocessor or other processor device and / or a logic device, which receives input and controls the operation of various components of the electronic device; the memory may be one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory, or other suitable devices, which may store the above-mentioned related data information, and may also store a program for executing related information, and the processor may execute the program stored in the memory to realize information storage or processing, etc.; the input unit is used to provide input to the processor, for example, it may be a key or a touch input device; the power supply is used to provide power to the electronic device; the display is used to display display objects such as images and text, for example, it may be an LCD display. The communication module is a transmitter / receiver that sends and receives signals via an antenna. The communication module (transmitter / receiver) is coupled to the processor to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal. Based on different communication technologies, multiple communication modules may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless local area network module, etc. The communication module (transmitter / receiver) is also coupled to a speaker and a microphone via an audio processor to provide an audio output via the speaker and receive an audio input from the microphone, thereby realizing a common telecommunication function. The audio processor may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor is also coupled to a central processor, so that recording can be performed on the machine through the microphone, and the sound stored on the machine can be played through the speaker.
[0070] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0071] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or multiple boxes.
[0072] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction system, which is implemented in the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0073] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the functions specified in a box or multiple boxes. Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the attached claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.
[0074] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A model migration method based on data classification, It is characterized in that include: S1. Setting data classification and first verification information for distinguishing different data classifications and processing templates corresponding to different data classifications; The data classification includes shared data, common data and personalized data; the shared data includes model data that can be directly obtained and used through public channels; the common data includes model data that cannot be directly obtained through public channels but can be directly used after being obtained; the personalized data includes model data that cannot be used directly and must be processed; The processing templates include: a shared data processing template, including shared data features and a shared data import path; a common data processing template, including common data features and a common data interface; a personalized data processing template, including personalized data features, a personalized data collection interface, a personalized data import interface, and a personalized data mapping conversion template; S2. Obtain model data and extract feature information from each item of model data; S3, matching the feature information with the first verification information, and marking the model data as a corresponding data classification according to the matching result; S4, grouping the model data into a set of data to be processed according to data classification; S5, using the corresponding processing template to update the data set to be processed to obtain the migratable data set; the step S5 includes the following steps: S51, recording the shared data features of the model data marked as shared data and the corresponding shared data import path according to the shared data processing template, and obtaining a shared migratable data set, wherein the shared migratable data set does not include the model data marked as shared data itself; S52, recording the common data features of the model data marked as common data and the corresponding common data interfaces according to the common data processing template, and obtaining a common transferable data set, wherein the common transferable data set does not include the model data marked as common data itself; S53, according to the personality data processing template, recording the personality data features of the model data marked as personality data and the corresponding personality data import interface, using the personality data collection interface to collect the model data and using the personality data mapping conversion template to convert the model data into pre-processed data suitable for the personality data import interface, to obtain a personality transferable data set, wherein the personality transferable data set includes the pre-processed data converted from the model data marked as personality data; S6, combining various transferable data sets to form transferable model data; S7. Export the model to be migrated and extract the model parsing document; S8. Update the model parsing document according to the first verification information and the processing template; S9. Migrate the migratable model data and the model combination with the updated model parsing document.
2. The method according to claim 1, It is characterized in that The first verification information includes data source, data level, data usage authority and data storage format.
3. The method according to claim 2, It is characterized in that The step S3 comprises the following sub-steps: S31, matching the feature information with the first verification information, verifying whether the model data belongs to the shared data, and marking the model data that passes the verification as the shared data; S32, when the verification model data does not belong to the shared data, matching the feature information with the first verification information, verifying whether the model data belongs to the common data, and marking the model data that passes the verification as the common data; S33. When the verification model data does not belong to common data, the model data is marked as personalized data.
4. The method according to claim 1, It is characterized in that The step S8 comprises: Based on data classification, information on shared data features, shared data import paths, common data features, common data interfaces, personalized data features, personalized data collection interfaces, personalized data import interfaces and personalized data mapping conversion templates is added to the model parsing document.
5. A model migration system based on data classification, It is characterized in that include: A data classification module, used to classify and mark the model data using the first verification information; Data classification includes shared data, common data and personalized data; the shared data includes model data that can be directly obtained and used through public channels; the common data includes model data that cannot be directly obtained through public channels but can be directly used after being obtained; the personalized data includes model data that cannot be used directly and must be processed; A data processing module, used to process model data of different categories respectively using processing templates; the processing templates include: a shared data processing template, including shared data features and shared data import paths; a common data processing template, including common data features and common data interfaces; a personalized data processing template, including personalized data features, personalized data collection interfaces, personalized data import interfaces and personalized data mapping conversion templates; The document updating module is used to update the model parsing document according to the first verification information and the processing template; update the to-be-processed data set using the corresponding processing template to obtain the migratable data set, including executing the following sub-steps: S51, recording the shared data features of the model data marked as shared data and the corresponding shared data import path according to the shared data processing template, and obtaining a shared migratable data set, wherein the shared migratable data set does not include the model data marked as shared data itself; S52, recording the common data features of the model data marked as common data and the corresponding common data interfaces according to the common data processing template, and obtaining a common transferable data set, wherein the common transferable data set does not include the model data marked as common data itself; S53, according to the personality data processing template, recording the personality data features of the model data marked as personality data and the corresponding personality data import interface, using the personality data collection interface to collect the model data and using the personality data mapping conversion template to convert the model data into pre-processed data suitable for the personality data import interface, to obtain a personality transferable data set, wherein the personality transferable data set includes the pre-processed data converted from the model data marked as personality data; The migration module is used to migrate the processed model data and the model combination with the updated model parsing document.
6. A computer-readable storage medium, It is characterized in that The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.
7. An electronic device, It is characterized in that including a processor and a memory; The memory is used to store model data and model parsing documents; The processor is used to execute the method according to any one of claims 1 to 4 by calling model data and model parsing documents.
