Digital model product calling method and device and readable storage medium

Abstract

The invention discloses a digital model product calling method and device and a readable storage medium, and the method comprises the steps: firstly receiving a model calling request of an external application, carrying out authority and product version verification through middleware, determining a digital model product of a target version, and executing a model execution request of the external application after confirmation. The method can ensure that the access to the model is subjected to proper authority verification, and can also ensure that a correct product version is being used at the same time.

Description

Technical Field

[0001] The present invention relates to the technical field of digital model management, and more particularly, to a method, apparatus, and readable storage medium for calling digital model products. Background Art

[0002] In the context of cloud computing and big data, the demand for digital model products is increasing day by day. These models are widely used in various fields such as decision support systems and predictive analysis systems. However, when using these models, some key issues need to be addressed, such as permission control and version management. Traditional methods often directly publish the model as an API interface for external applications to call, but this approach has drawbacks. For example, without an appropriate permission control mechanism, anyone can access and manipulate the model, which may lead to information leakage or incorrect operations. In addition, if there are multiple versions of the model, how to ensure that external applications always call the latest or specific version is also a challenge. Therefore, this method for calling digital model products has emerged to solve these problems. Summary of the Invention

[0003] The purpose of the present invention is to provide a method, apparatus, and readable storage medium for calling digital model products.

[0004] In a first aspect, an embodiment of the present invention provides a method for calling a digital model product, including:

[0005] Obtaining a model call request from an external application;

[0006] Using a middleware to perform permission verification and product version verification on the model call request;

[0007] Determining a digital model product of a target version according to the permission verification result and the product version verification result;

[0008] In response to the model execution request of the external application, performing an execution task corresponding to the model execution request through the digital model product of the target version.

[0009] In a possible implementation manner, the using a middleware to perform permission verification and product version verification on the model call request includes:

[0010] Using the middleware to obtain the digital model product code, digital model product requirement parameters, requester identifier, and call request signature included in the model call request;

[0011] Performing a normalization verification on the digital model product code, the digital model product requirement parameters, the requester identifier, and the call request signature;

[0012] In the case where the normative verification passes, perform the permission verification according to the requester identification and the call request signature;

[0013] Obtain the current digital model product version;

[0014] Perform the product version verification according to the digital model product code, the digital model product requirement parameters, and the current digital model product version.

[0015] In a possible implementation manner, the obtaining the current digital model product version includes:

[0016] Obtain the current digital model product version from a digital product status memory, where the digital product status memory and the middleware are integrated in the same system;

[0017] The obtaining the current digital model product version further includes:

[0018] Obtain the current digital model product version from an external database, where the external database is communicatively connected to the middleware.

[0019] In a possible implementation manner, the determining the digital model product of the target version according to the permission verification result and the product version verification result includes:

[0020] In the case where the permission verification passes, obtain the current digital model product code and the current digital model product requirement parameters of the current digital model product version;

[0021] If the digital model product code is the same as the current digital model product code, and the digital model product requirement parameters are the same as the current digital model product requirement parameters, then use the digital model product corresponding to the current digital model product version as the digital model product of the target version;

[0022] If the digital model product code is different from the current digital model product code and / or the digital model product requirement parameters are different from the current digital model product requirement parameters, then use the new digital model product determined based on the digital model product code and the digital model product requirement parameters as the digital model product of the target version.

[0023] In a possible implementation manner, after using the new digital model product determined based on the digital model product code and the digital model product requirement parameters as the digital model product of the target version, the method further includes:

[0024] Update the target version to the digital product status memory.

[0025] In a possible implementation, in response to a model execution request from an external application, executing an execution task corresponding to the model execution request through the digital model product of the target version includes:

[0026] Determine the definitions of parameter configuration, parameter rule verification, data query, data processing, and model file management corresponding to the digital model product of the target version;

[0027] In response to the model execution request from the external application, execute the execution task corresponding to the model execution request through the definitions of the parameter configuration, the parameter rule verification, the data query, the data processing, and the model file management.

[0028] In a possible implementation, the method further includes:

[0029] Obtain multiple model invocation requests from the external application;

[0030] Use a reverse proxy to process the multiple model invocation requests to obtain the digital model product of the required version corresponding to each model invocation request.

[0031] In a second aspect, an embodiment of the present invention provides a digital model product invocation device, including:

[0032] An acquisition module, configured to acquire a model invocation request from an external application;

[0033] A verification module, configured to perform permission verification and product version verification on the model invocation request by using middleware; and determine a digital model product of a target version according to the permission verification result and the product version verification result;

[0034] An execution module, configured to, in response to the model execution request from the external application, execute an execution task corresponding to the model execution request through the digital model product of the target version.

[0035] In a third aspect, an embodiment of the present invention provides a computer device, where the computer device includes a processor and a non-volatile memory storing computer instructions, and when the computer instructions are executed by the processor, the computer device executes the digital model product invocation method in at least one possible implementation of the first aspect.

[0036] In a fourth aspect, an embodiment of the present invention provides a readable storage medium, where the readable storage medium includes a computer program, and when the computer program runs, it controls the computer device where the readable storage medium is located to execute the digital model product invocation method in at least one possible implementation of the first aspect.

[0037] Compared with the prior art, the beneficial effects provided by the present invention include: By adopting a method, device and readable storage medium for calling digital model products disclosed in the present invention, upon receiving a model call request from an external application, a middleware is used to perform permission and product version verification, determine the digital model product of the target version, and after confirmation, execute the model execution request of the external application. This method can ensure that all accesses to the model have undergone appropriate permission verification and also ensure that the correct product version is being used. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0039] Figure 1 Schematic diagram of the steps of the method for calling digital model products provided by the embodiments of the present invention;

[0040] Figure 2 Schematic block diagram of the structure of the device for calling digital model products provided by the embodiments of the present invention;

[0041] Figure 3 Schematic block diagram of the structure of the computer device provided by the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0043] The following will, with reference to the drawings, provide a detailed description of the specific embodiments of the present invention.

[0044] To solve the technical problems in the foregoing background art, Figure 2 The following is a schematic flowchart of the method for calling digital model products provided by the embodiments of the present disclosure. The method for calling digital model products will be introduced in detail below.

[0045] Step S201, obtain a model call request from an external application;

[0046] Step S202, use middleware to perform permission verification and product version verification on the model call request;

[0047] Step S203: Determine the digital model product of the target version based on the permission verification result and the product version verification result.

[0048] Step S204: In response to the model execution request of the external application, execute the execution task corresponding to the model execution request through the digital model product of the target version.

[0049] Exemplarily, assume that we have an online image recognition application where users can upload pictures to obtain corresponding tag information. The external application will send the pictures uploaded by the users and related parameters to the digital model product for processing. The middleware, as an interceptor, will receive the model call request sent by the external application. The middleware will first perform permission verification, such as checking whether the external application has sufficient permissions to call the digital model product. Next, the middleware will perform product version verification to determine whether the digital model product requested by the external application is the currently available version. Assume that the digital model product has two versions: A and B. If both the permission verification and the product version verification pass, the middleware will determine the target version as the latest digital model product (such as version B). If the permission verification passes but the product version verification fails, the middleware will determine the target version as the stable version of the digital model product (such as version A). According to the decision of the middleware, the external application will receive a response from the selected target version of the digital model product. The external application can continue to send model execution requests and execute corresponding tasks through the selected target version of the digital model product, such as identifying the uploaded pictures and returning relevant tag information. Designed in this way, by performing permission verification and product version verification through the middleware, it is ensured that only external applications with appropriate permissions can call the appropriate version of the digital model product. This approach can ensure that each call is independent, and even if a new version is released, it will not affect the currently invoked process.

[0050] In the embodiment of the present invention, the foregoing step S202 can be implemented by the following method.

[0051] (1) Use the middleware to obtain the digital model product code, digital model product requirement parameters, requester identifier, and call request signature included in the model call request.

[0052] (2) Perform normalization verification on the digital model product code, the digital model product requirement parameters, the requester identifier, and the call request signature.

[0053] (3) In the case where the normalization verification passes, perform the permission verification according to the requester identifier and the call request signature.

[0054] (4) Obtain the current digital model product version.

[0055] (5)Perform the product version verification according to the digital model product code, the digital model product requirement parameters, and the current digital model product version.

[0056] Exemplarily, assume there is an online machine translation platform, and users call the translation service of the platform through the API. The external application sends a model call request, including the digital model product code (the unique identifier of the translation model), the digital model product requirement parameters (the text to be translated), the requester identifier (the identity information of the external application), and the call request signature (used to verify the validity of the request). The middleware will verify the received digital model product code, digital model product requirement parameters, requester identifier, and call request signature to ensure that their formats and contents meet the expectations. For example, check whether the digital model product code exists, whether the requirement parameters meet the specified data type and length limits, whether the requester identifier is valid, and whether the call request signature is correct. If the normative verification passes, the middleware will perform the permission verification using the requester identifier and the call request signature. For example, the middleware can check the identity information and signature key of the external application to ensure that it has sufficient permissions to call a specific digital model product. The middleware will query the information stored in the built-in data product status storage module to obtain the current available digital model product version. For example, if the current stable version is recorded as 2.0 in the storage module, the middleware will obtain this information. Perform the product version verification according to the digital model product code, the digital model product requirement parameters, and the current digital model product version: The middleware will use the digital model product code, the digital model product requirement parameters, and the current digital model product version to verify whether the target version of the request is valid. For example, the middleware can compare the requested digital model product code and requirement parameters with the requirements of the current stable version to determine whether the current version can be selected as the target version. Designed in this way, the middleware can perform permission verification and product version verification on the model call request. It will ensure the correct format of the request through the normative verification, and then perform the permission verification according to the requester identifier and the call request signature. Subsequently, the middleware will obtain the current digital model product version and verify it with the requested digital model product code and requirement parameters to determine the digital model product of the target version. This can ensure that only requesters with appropriate permissions can call the appropriate version of the digital model product.

[0057] In the embodiment of the present invention, the foregoing step of obtaining the current digital model product version can be executed through the following example.

[0058] (1)Obtain the current digital model product version from the digital product status memory, and the digital product status memory is integrated with the middleware in the same system;

[0059] The foregoing steps for obtaining the current digital model product version may also be implemented through the following examples.

[0060] (1) Obtain the current digital model product version from an external database, where the external database is communicatively connected to the middleware.

[0061] Exemplarily, assume we have an online recommendation system, and the middleware is used to process users' recommendation requests. The middleware needs to obtain the version of the currently available recommendation algorithm model. The digital product status memory is a module in this system, which records the version information of each recommendation algorithm model. The middleware can obtain the current digital model product version from the digital product status memory to determine which version of the recommendation algorithm model to use to process the request. Assume we have an online payment system, and the middleware is used to process payment requests. The middleware needs to obtain the version of the currently available payment service. The external database is the place where the version information is stored in this system and is communicatively connected to the middleware. The middleware can send a request to the external database to obtain the information of the current digital model product version. Through the above steps, the middleware can obtain the information of the current digital model product version. This can be achieved in two ways: one is to obtain the version information from the digital product status memory, and the other is to communicate with the external database and obtain the version information. Whichever way is used, the middleware can obtain the currently available digital model product version so as to select an appropriate version to process the request according to the requirements.

[0062] In the embodiments of the present invention, the foregoing step S203 may also be implemented through the following examples.

[0063] (1) In the case where the permission verification passes, obtain the current digital model product code and the current digital model product requirement parameters of the current digital model product version;

[0064] (2) If the digital model product code is the same as the current digital model product code, and the digital model product requirement parameters are the same as the current digital model product requirement parameters, then use the digital model product corresponding to the current digital model product version as the digital model product of the target version;

[0065] (3) If the digital model product code is different from the current digital model product code and / or the digital model product requirement parameters are different from the current digital model product requirement parameters, then use the new digital model product determined based on the digital model product code and the digital model product requirement parameters as the digital model product of the target version.

[0066] Exemplarily, assume there is an online music recommendation system, and the middleware is used to process users' music recommendation requests. After the permission verification passes, the middleware needs to obtain the version of the currently available music recommendation model, and obtain the digital model product code and requirement parameters of this version. The middleware can obtain the relevant information of the current digital model product version from the digital product status memory or an external database. If the requested digital model product code and requirement parameters exactly match the current digital model product code and requirement parameters, it indicates that the request has specified the target version to be used. In this case, the middleware directly selects the digital model product corresponding to the current digital model product version as the target version and uses it to execute the request task. If the requested digital model product code and requirement parameters do not exactly match the current digital model product code and requirement parameters, it indicates that the request needs to select a different target version. In this case, the middleware will determine the new target version based on the requested digital model product code and requirement parameters. The middleware can query the digital product status memory or an external database to find the digital model product version that best matches the request and use it as the target version. Designed in this way, the middleware can determine the digital model product of the target version based on the permission verification result and the product version verification result. If the request exactly matches the current version, the middleware directly selects the current version. If the request does not exactly match the current version, the middleware will select a new suitable version as the target version according to the code and requirement parameters of the request. This can ensure that the correct digital model product version is selected for processing according to specific requirements.

[0067] In an embodiment of the present invention, after using the new digital model product determined based on the digital model product code and the digital model product requirement parameters as the digital model product of the target version as described above, the following examples are further provided.

[0068] (1) Update the target version to the digital product status memory.

[0069] Exemplarily, assume there is an online recommendation system, and the middleware is used to process users' recommendation requests. After selecting a new suitable version according to the digital model product code and demand parameters, the middleware determines the digital model product of the target version. The digital product status memory is the place to record the version information of the digital model product. After determining the target version, the middleware updates and stores the information of this target version in the digital product status memory so that subsequent requests can use the latest target version. Assume that in the recommendation system, the current version of the recommendation algorithm model is version 1.0, but due to system improvement or changes in user requirements, it is necessary to switch to version 2.0. When the middleware receives a new recommendation request, after passing the permission verification and product version verification, it determines that the target version is 2.0 and updates this information to the digital product status memory. The next time a similar recommendation request comes, the middleware will select and use the recommendation algorithm model of version 2.0 according to the target version information recorded in the digital product status memory. Designed in this way, the middleware can use the new digital model product based on the determined digital model product code and demand parameters as the target version and update this target version to the digital product status memory. This can ensure that subsequent requests can use the latest target version, enabling the system to flexibly adapt to different requirements and changes.

[0070] In the embodiment of the present invention, the foregoing step S204 can be implemented through the following examples.

[0071] (1) Determine the definitions of parameter configuration, parameter rule verification, data query, data processing, and model file management corresponding to the digital model product of the target version;

[0072] (2) In response to the model execution request of the external application, execute the execution task corresponding to the model execution request through the definitions of the parameter configuration, the parameter rule verification, the data query, the data processing, and the model file management.

[0073] Exemplarily, assume there is an online risk control system, and the middleware is used to process risk assessment requests sent by external applications. After determining the digital model product of the target version, the middleware needs to determine the parameter configuration, parameter rule verification, data query, data processing, and model file management definitions required for this target version. These definitions include setting the input and output parameters of the digital model, defining parameter rules and constraints, specifying database query statements, designing data processing flows, and managing model files related to the target version. When an external application sends a risk assessment request, the middleware receives the request and starts to execute the corresponding tasks. The middleware obtains the required input parameters according to the previously defined parameter configuration and performs parameter rule verification to ensure that the input meets the expectations. The middleware obtains the required data from the database according to the defined data query rules. Next, the middleware performs data processing, appropriately processing and transforming the obtained data according to the defined data processing flow. Finally, the middleware uses the digital model product of the target version to perform model execution, passing the processed data to the digital model and obtaining an evaluation result. Designed in this way, the middleware can respond to the model execution requests of external applications and use the digital model product of the target version to execute the corresponding tasks. This involves determining the parameter configuration, parameter rule verification, data query, data processing, and model file management definitions required for the target version, and performing corresponding operations according to these definitions during the execution process, and finally completing the tasks required by the model execution request. This method can ensure that the system can correctly execute model tasks according to the requirements of the target version and generate accurate results.

[0074] In an embodiment of the present invention, the following examples are also provided.

[0075] (1) Obtain multiple model call requests of the external application;

[0076] (2) Use a reverse proxy to process the multiple model call requests to obtain the digital model product of the required version corresponding to each model call request.

[0077] Exemplarily, assume there is an online advertising system, and the middleware is responsible for processing advertising placement requests sent by external applications. Before processing these requests, the middleware needs to obtain multiple model call requests from the external applications. These model call requests may include classification of advertising materials, analysis of target audiences, optimization of advertising placement timing, etc. The middleware uses the function of the reverse proxy to route the received multiple model call requests to the corresponding digital model services. The reverse proxy selects the appropriate target version of the digital model product for processing according to the parameters and rules required by each model call request. Through the processing of the reverse proxy, the middleware can obtain the digital model products of the required versions corresponding to each model call request. In the online advertising system, the external application sends multiple advertising placement requests, including classification and recommendation of different advertising materials. As the reverse proxy, the middleware will route these requests to the corresponding digital model services after receiving them. Assume there are two digital model services, namely the advertising material classification model and the advertising recommendation model. Among them, the latest version of the advertising material classification model is 2.0, and the latest version of the advertising recommendation model is 1.5. The middleware routes the advertising material classification requests to the advertising material classification model of version 2.0 and the advertising recommendation requests to the advertising recommendation model of version 1.5 according to the parameters and rules required by each request. In this way, the middleware can obtain the digital model products of the required versions corresponding to each model call request based on the processing of the reverse proxy. Designed in this way, the middleware can obtain multiple model call requests sent by the external application and use the function of the reverse proxy to route these requests to the corresponding digital model services. By processing with the digital model products of the target versions applicable to each request, it is ensured that each model call request can obtain a result that meets the requirements. This method enables the middleware to select appropriate digital model products for processing according to different requests, improving the flexibility and performance of the system.

[0078] In addition, the embodiments of the present invention also provide the following implementation processes.

[0079] In the embodiments of the present invention, the product version is defined in terms of parameter configuration, parameter rule verification, data query, data processing, and model file management.

[0080] Parameter configuration: mainly used to configure the parameters and parameter types required by the interface when the product version is called through httpapi, as well as the default values that need to be automatically filled by the system when empty or not passed in the request, and whether the parameters need to enter the model file for calculation.

[0081] Parameter verification rules: mainly verify the legality of the parameter content in the request through parameter name rule verification, type rule verification, and expression. Among them, the verification type includes regular and calculation. When the verification type is selected as calculation, the content filled in the rule expression is a lua language expression.

[0082] Data query: For the operation model, it may not be enough just through the parameter content sent by the requester. Data supplementation from the data warehouse or big data platform may also be required. Therefore, in data query, it is possible to select which query conditions to use to query which data from which system, which database, and which table to supplement the operation model with data.

[0083] Data processing:

[0084] Pre - SQL: If data combination or aggregation operations need to be performed on the data query results and parameters before the data enters the operation model, then SQL is used to complete the data processing of the parameters and data query results. Post - SQL: Post - SQL is a function for finally secondary - processing the results of the operation model operation combined with data query and parameter content. The result of post - SQL is the final interface return result.

[0085] Model file management: For the model files required for the current data product version, and the request traffic allocation ratio of each model file can be configured. The sum of the total traffic ratios of multiple model files is 1.

[0086] In summary, in the embodiment of the present invention, version control is performed through the built - in data product status storage module, and the currently launched data product is determined and called in real - time through the middleware module, thereby ensuring the efficiency of each request, and at the same time achieving true isolation between processes. Even when a new version is released, it will not affect the running processes.

[0087] Please refer to Figure 2 , Figure 2 which is a digital model product calling device 110 provided by an embodiment of the present invention, including:

[0088] An obtaining module 1101, configured to obtain a model call request of an external application;

[0089] A verification module 1102, configured to perform permission verification and product version verification on the model call request by using middleware; and determine a digital model product of a target version according to the permission verification result and the product version verification result;

[0090] An execution module 1103, configured to, in response to a model execution request of the external application, execute an execution task corresponding to the model execution request through the digital model product of the target version.

[0091] It should be noted that the implementation principle of the aforementioned digital model product calling device 110 can refer to the implementation principle of the digital model product calling device 110 described above, which will not be elaborated here. It should be understood that the division of each module of the above device is only a logical function division. In actual implementation, it can be fully or partially integrated into a physical entity, or physically separated. And these modules can all be implemented in the form of software called by processing elements; they can also all be implemented in the form of hardware; or some modules can be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. For example, the digital model product calling device 110 can be a separately established processing element, or can be integrated in a certain chip of the above device. In addition, it can also be stored in the memory of the above device in the form of program code, and the function of the above digital model product calling device 110 can be called and executed by a certain processing element of the above device. The implementation of other modules is similar. In addition, all or part of these modules can be integrated together or can be independently implemented. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed by the integrated logic circuit in the processor element or the instruction in the form of software.

[0092] For example, the above modules can be one or more integrated circuits configured to implement the above method, such as: one or more application specific integrated circuits (ASICs), or, one or more digital signal processors (DSPs), or, one or more field programmable gate arrays (FPGAs), etc. Again, when a certain module above is implemented in the form of a processing element scheduling program code, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processors that can call program code. Again, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0093] An embodiment of the present invention provides a computer device 100. The computer device 100 includes a processor and a non-volatile memory storing computer instructions. When the computer instructions are executed by the processor, the computer device 100 executes the aforementioned digital model product calling device 110. As Figure 3 shown, Figure 3The computer device 100 provided in the embodiment of the present invention is a structural block diagram. The computer device 100 includes a digital model product calling device 110, a memory 111, a processor 112 and a communication unit 113.

[0094] In order to realize data transmission or interaction, the memory 111, the processor 112 and the communication unit 113 are electrically connected to each other directly or indirectly. For example, the electrical connection between these elements can be realized through one or more communication buses or signal lines. The digital model product calling device 110 includes at least one software function module that can be stored in the memory 111 in the form of software or firmware or solidified in the operating system (OS) of the computer device 100. The processor 112 is used to execute the digital model product calling device 110 stored in the memory 111, such as the software function modules and computer programs included in the digital model product calling device 110.

[0095] An embodiment of the present invention provides a readable storage medium, which includes a computer program. When the computer program is running, the computer device where the readable storage medium is located is controlled to execute the aforementioned digital model product calling method.

[0096] For illustrative purposes, the foregoing description is made with reference to specific embodiments. However, the above illustrative discussion is not intended to be exhaustive or to limit the present disclosure to the precise form disclosed. Numerous modifications and variations are possible in accordance with the above teachings. These embodiments are selected and described in order to best illustrate the principles of the present disclosure and its practical application, so that those skilled in the art can best utilize the present disclosure and utilize various embodiments with different modifications to suit the intended specific application.

Claims

1. A method for calling a digital model product, characterized in that Including: Obtain a model call request of an external application; Use middleware to perform permission verification and product version verification on the model call request; Determine a digital model product of a target version according to the permission verification result and the product version verification result; In response to the model execution request of the external application, execute the execution task corresponding to the model execution request through the digital model product of the target version.

2. The method according to claim 1, characterized in that, The using middleware to perform permission verification and product version verification on the model call request includes: Use the middleware to obtain the digital model product code, digital model product requirement parameters, requester identifier, and call request signature included in the model call request; Perform normalization verification on the digital model product code, the digital model product requirement parameters, the requester identifier, and the call request signature; In the case where the normalization verification passes, perform the permission verification according to the requester identifier and the call request signature; Obtain the current digital model product version; Perform the product version verification according to the digital model product code, the digital model product requirement parameters, and the current digital model product version.

3. The method according to claim 2, characterized in that The obtaining the current digital model product version includes: Obtain the current digital model product version from a digital product status memory, and the digital product status memory is integrated with the middleware in the same system; The obtaining the current digital model product version further includes: Obtain the current digital model product version from an external database, and the external database is communicatively connected to the middleware.

4. The method according to claim 2, characterized in that The determining a digital model product of a target version according to the permission verification result and the product version verification result includes: In the case where the permission verification passes, obtain the current digital model product code and the current digital model product requirement parameters of the current digital model product version; If the digital model product code is the same as the current digital model product code, and the digital model product requirement parameters are the same as the current digital model product requirement parameters, then use the digital model product corresponding to the current digital model product version as the digital model product of the target version; If the digital model product code is different from the current digital model product code and / or the digital model product requirement parameters are different from the current digital model product requirement parameters, then use the new digital model product determined based on the digital model product code and the digital model product requirement parameters as the digital model product of the target version.

5. The method according to claim 4, wherein After using the new digital model product determined based on the digital model product code and the digital model product requirement parameters as the digital model product of the target version, the method further includes: Update the target version to the digital product status memory.

6. The method according to claim 1, characterized in that The responding to the model execution request of the external application and executing the execution task corresponding to the model execution request through the digital model product of the target version includes: Determine the definitions of parameter configuration, parameter rule verification, data query, data processing, and model file management corresponding to the digital model product of the target version; In response to a model execution request from the external application, an execution task corresponding to the model execution request is executed through the definitions of the parameter configuration, the parameter rule verification, the data query, the data processing, and the model file management.

7. The method according to claim 6, wherein The method further includes: Obtaining a plurality of model call requests of the external application; Processing the plurality of model call requests by using a reverse proxy to obtain a digital model product of a required version corresponding to each model call request.

8. A digital model product calling device, characterized in that, It includes: An obtaining module, configured to obtain a model call request of an external application; A verification module, configured to perform permission verification and product version verification on the model call request by using middleware; and determine a digital model product of a target version according to the permission verification result and the product version verification result; An execution module, configured to, in response to a model execution request from the external application, execute an execution task corresponding to the model execution request through the digital model product of the target version.

9. A computer device, characterized in that, The computer device includes a processor and a non-volatile memory storing computer instructions. When the computer instructions are executed by the processor, the computer device executes the digital model product calling method according to any one of claims 1-7.

10. A readable storage medium, characterized in that, The readable storage medium includes a computer program. When the computer program runs, it controls the computer device where the readable storage medium is located to execute the digital model product calling method according to any one of claims 1-7.

Images