Determination Method, Device, Equipment and Storage Medium of Recommendation Model

Abstract

The present disclosure relates to a method, apparatus, device, and storage medium for determining a recommendation model. The method includes: in response to detecting that a service provider meets the model recommendation conditions, determining an initial model service for the service provider; determining a target model service group based on the initial model service, where the target model service group is a pre-determined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information; determining a target model service from each model service within the target model service group; and determining a model for making recommendations to the service provider based on the model corresponding to the target model service. The present disclosure can achieve model recommendation using existing model services without introducing additional service resources, can both achieve cross-domain model recommendation and reduce the software and hardware costs of the model recommendation service provider as well as the labor costs for model evaluation.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and particularly to a method, apparatus, device, and storage medium for determining a recommendation model. Background Art

[0002] With the development of deep learning technology and the significant improvement of GPU computing power, artificial intelligence technology has also achieved great development. Nowadays, many enterprises have launched machine learning-related services for other enterprises, including model training services and model inference services, etc., among which there is a service for model recommendation according to business requirements.

[0003] In related technologies, models are usually designed and recommended based on specific fields and requirements. For example, a product recommendation model designed for the e-commerce field and requirements, a power system data analysis platform designed for the power field and requirements, and a gene analysis system designed for the medical field and requirements, etc. However, the modeling process of these models designed based on specific fields and requirements is relatively complex, and it will consume huge software and hardware costs of the model recommendation service provider and the labor cost for model evaluation. Summary of the Invention

[0004] To overcome the problems existing in related technologies, embodiments of the present disclosure provide a method, apparatus, device, and storage medium for determining a recommendation model to solve the defects in related technologies.

[0005] According to a first aspect of an embodiment of the present disclosure, there is provided a method for determining a recommendation model, the method including:

[0006] In response to detecting that a business party meets the model recommendation condition, determining an initial model service of the business party;

[0007] Based on the initial model service, determining a target model service group, where the target model service group is a group to which the initial model service belongs determined in advance, and each model service in the target model service group has the same interface declaration set information;

[0008] Determining a target model service from each model service in the target model service group;

[0009] Based on the model corresponding to the target model service, determining a model for recommending to the business party.

[0010] In an embodiment, the determining the initial model service of the business party includes:

[0011] Obtaining a business party identifier of the business party;

[0012] Determining the initial model service corresponding to the business party identifier as the initial model service of the business party.

[0013] In one embodiment, the method further includes determining a target model service group to which the initial model service belongs based on the following:

[0014] Detecting information on a set of target interface declarations of the initial model service;

[0015] Searching for a target model service group that matches the information on the set of target interface declarations;

[0016] In response to finding the target model service group, adding the initial model service to the target model service group.

[0017] In one embodiment, the method further includes:

[0018] In response to not finding the target model service group, creating a target model service group to which the initial model service belongs, and registering the information on the set of target interface declarations under the target model service group;

[0019] Adding the initial model service to the created target model service group.

[0020] In one embodiment, determining a target model service from among various model services within the target model service group includes:

[0021] Obtaining evaluation information values of various model services within the target model service group;

[0022] Determining a target model service based on the model service with the highest evaluation information value among the various model services.

[0023] In one embodiment, the method further includes pre-determining the evaluation information value of the model service based on the following:

[0024] Collecting actual operation data of the model service based on a preset period;

[0025] In response to the actual operation data meeting an evaluation condition, determining the evaluation information value of the model service based on the actual operation data.

[0026] In one embodiment, determining the evaluation information value of the model service based on the actual operation data includes:

[0027] Performing preprocessing on the actual operation data to obtain preprocessed data;

[0028] Performing normalization processing on the preprocessed data to obtain normalized data;

[0029] Performing a weighted summation operation on the normalized data to obtain the evaluation information value of the model service.

[0030] According to a second aspect of the embodiments of the present disclosure, there is provided an apparatus for determining a recommendation model, the apparatus comprising:

[0031] An initial service determination module, configured to determine an initial model service of the service party in response to detecting that the service party meets the model recommendation condition;

[0032] A service group determination module, configured to determine a target model service group based on the initial model service, where the target model service group is a pre-determined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information;

[0033] A target service determination module, configured to determine a target model service from each model service in the target model service group;

[0034] A recommendation model determination module, configured to determine a model for recommending to the service party based on the model corresponding to the target model service.

[0035] In an embodiment, the initial service determination module includes:

[0036] A service identity acquisition unit, configured to acquire the service party identity of the service party;

[0037] An initial service determination unit, configured to determine the initial model service corresponding to the service party identity as the initial model service of the service party.

[0038] In an embodiment, the apparatus further includes an affiliated group determination module;

[0039] The affiliated group determination module includes:

[0040] A set information detection unit, configured to detect the target interface declaration set information of the initial model service;

[0041] An affiliated group determination unit, configured to search for a target model service group that matches the target interface declaration set information;

[0042] A target group addition unit, configured to add the initial model service to the target model service group in response to finding the target model service group.

[0043] In an embodiment, the affiliated group determination module further includes:

[0044] A target group creation unit, configured to create a target model service group to which the initial model service belongs and register the target interface declaration set information under the target model service group in response to not finding the target model service group.

[0045] The target group adding unit is further configured to add the initial model service to the created target model service group.

[0046] In one embodiment, the target service determining module includes:

[0047] An evaluation information value obtaining unit, configured to obtain the evaluation information values of each model service in the target model service group;

[0048] A target service determining unit, configured to determine a target model service based on the model service with the highest evaluation information value among the model services.

[0049] In one embodiment, the apparatus further includes an evaluation information value determining module;

[0050] The evaluation information value determining module includes:

[0051] An operation data collecting unit, configured to collect the actual operation data of the model service based on a preset period;

[0052] An evaluation information value determining unit, configured to determine the evaluation information value of the model service based on the actual operation data in response to the actual operation data satisfying the evaluation condition.

[0053] In one embodiment, the evaluation information value determining unit is further configured to:

[0054] Preprocess the actual operation data to obtain preprocessed data;

[0055] Normalize the preprocessed data to obtain normalized data;

[0056] Perform a weighted summation operation on the normalized data to obtain the evaluation information value of the model service.

[0057] According to a third aspect of the embodiments of the present disclosure, there is provided an electronic device, including:

[0058] A processor and a memory for storing processor-executable instructions;

[0059] Wherein, the processor is configured to:

[0060] In response to detecting that a service provider meets the model recommendation condition, determine the initial model service of the service provider;

[0061] Determine a target model service group based on the initial model service, where the target model service group is a pre-determined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information;

[0062] Determine a target model service from each model service within the target model service group;

[0063] Based on the model corresponding to the target model service, determine a model for recommendation to the business party.

[0064] According to a fourth aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, it implements:

[0065] In response to detecting that a business party meets the model recommendation condition, determine the initial model service of the business party;

[0066] Based on the initial model service, determine a target model service group, where the target model service group is a pre-determined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information;

[0067] Determine a target model service from each model service within the target model service group;

[0068] Based on the model corresponding to the target model service, determine a model for recommendation to the business party.

[0069] The technical solutions provided by the embodiments of the present disclosure may include the following beneficial effects:

[0070] In the present disclosure, in response to detecting that a business party meets the model recommendation condition, the initial model service of the business party is determined, and based on the initial model service, a target model service group is determined. Then, a target model service is determined from each model service within the target model service group, and further, based on the model corresponding to the target model service, a model for recommendation to the business party is determined. Since the target model service group is a pre-determined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information, it is possible to determine the target model service based on the model service in the group to which the initial model service belongs and having the same interface declaration set information as the initial model service when detecting that the business party meets the model recommendation condition, and further determine the model for recommendation to the business party. It is possible to implement model recommendation using existing model services without introducing additional service resources, which can not only achieve cross-domain model recommendation but also reduce the software and hardware costs of the model recommendation service provider and the labor costs for model evaluation.

[0071] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] The accompanying drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure.

[0073] Figure 1 is a flowchart of a method for determining a recommendation model shown according to an exemplary embodiment;

[0074] Figure 2 is a flowchart of how to determine the initial model service of the business party shown according to an exemplary embodiment;

[0075] Figure 3 is a flowchart of how to determine the target model service group to which the initial model service belongs shown according to an exemplary embodiment;

[0076] Figure 4 is a flowchart of how to determine a target model service from each model service within the target model service group shown according to an exemplary embodiment;

[0077] Figure 5 is a flowchart of how to determine the evaluation information value of the model service shown according to an exemplary embodiment;

[0078] Figure 6 is a flowchart of how to determine the evaluation information value of the model service based on the actual operation data shown according to an exemplary embodiment;

[0079] Figure 7 is a block diagram of a device for determining a recommendation model shown according to an exemplary embodiment;

[0080] Figure 8 is a block diagram of a device for determining a recommendation model shown according to another exemplary embodiment;

[0081] Figure 9 is a block diagram of an electronic device shown according to an exemplary embodiment. Detailed Description of the Invention

[0082] Exemplary embodiments will be described in detail herein, and examples thereof are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0083] Figure 1It is a flowchart of a method for determining a recommendation model according to an exemplary embodiment; the method of this embodiment can be applied to a cluster including multiple node devices, wherein the node devices include physical machines, virtual machines or containers, etc.

[0084] like Figure 1 As shown, the method includes the following steps S101-S103:

[0085] In step S101, in response to detecting that a business party meets a model recommendation condition, an initial model service of the business party is determined.

[0086] In this embodiment, when it is detected that a business party meets the model recommendation condition, the initial model service of the business party may be determined.

[0087] The initial model service may be a model service created for the business party in the current cluster. In one embodiment, the initial model service may be a model service provided by the business party for creation and operation in the current cluster. The model service may be a service for reasoning using a model in related technologies, and the type of the external interface of the model service may be set based on actual needs, such as http, grpc or other rpc types, etc. This embodiment does not limit this.

[0088] For example, the specific way in which the above-mentioned business party meets the model recommendation conditions can be set based on actual needs, such as setting it to the poor performance of the current model service of the business party, or the performance of the current model service is not optimal within the model service group to which it belongs, etc.

[0089] In step S102, a target model service group is determined based on the initial model service.

[0090] In this embodiment, in response to detecting that a business party meets a model recommendation condition, after determining an initial model service of the business party, a target model service group may be determined based on the initial model service.

[0091] The target model service group is a predetermined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information. That is to say, the interface declaration set information of other model services in the target model service group except the initial model service is the same as the interface declaration set information of the initial model service.

[0092] In another embodiment, the above method of determining the target model service group based on the initial model service can also refer to the following Figure 3 The illustrated embodiment will not be described in detail here.

[0093] In step S103, a target model service is determined from each model service within the target model service group.

[0094] In this embodiment, after determining the target model service group based on the initial model service, a target model service can be determined from each model service within the target model service group.

[0095] For example, after determining the target model service group based on the initial model service, each model service within the target model service group can be obtained, and then a model service with better performance than the initial model service can be determined from these model services as the target model service.

[0096] In another embodiment, the method of determining the target model service from each model service within the target model service group can also refer to the following Figure 4 illustrated embodiment, which will not be elaborated here first.

[0097] In step S104, a model for recommending to the business party is determined based on the model corresponding to the target model service.

[0098] In this embodiment, after determining the target model service from each model service within the target model service group, a model for recommending to the business party can be determined based on the model corresponding to the target model service. For example, the model corresponding to the target model service can be obtained, and then this model can be recommended to the business party.

[0099] As can be seen from the above description, the method of this embodiment responds to detecting that the business party meets the model recommendation condition, determines the initial model service of the business party, determines the target model service group based on the initial model service, then determines the target model service from each model service within the target model service group, and further determines the model for recommending to the business party based on the model corresponding to the target model service. Since the target model service group is the group to which the initial model service belongs that is determined in advance, and each model service in the target model service group has the same interface declaration set information, it is possible to determine the target model service based on the model service in the group to which the initial model service belongs and having the same interface declaration set information as the initial model service when detecting that the business party meets the model recommendation condition, and then determine the model for recommending to the business party. It is possible to perform model recommendation using existing model services without introducing additional service resources, which can not only achieve cross - domain model recommendation but also reduce the software and hardware costs of the model recommendation service provider and the labor costs for model evaluation.

[0100] Figure 2It is a flowchart showing how to determine the initial model service of the business party according to an exemplary embodiment; this embodiment takes how to determine the initial model service of the business party as an example for exemplary illustration on the basis of the above embodiment. As Figure 2 shown, in the above step S101, determining the initial model service of the business party may include the following steps S201 - S202:

[0101] In step S201, obtain the business party identifier of the business party.

[0102] In this embodiment, when it is detected that the business party meets the model recommendation condition, the business party identifier of the business party can be obtained.

[0103] For example, the above business party identifier may include the company name, organization code or other preset identifiers of the business party, and this embodiment does not limit this.

[0104] In step S202, determine the initial model service corresponding to the business party identifier as the initial model service of the business party.

[0105] In this embodiment, after obtaining the business party identifier of the business party, the initial model service corresponding to the business party identifier can be determined as the initial model service of the business party.

[0106] For example, when creating an initial model service for the above business party in the current cluster, a corresponding relationship between the business party identifier of the business party and this initial model service can be constructed. Then, after obtaining the business party identifier of this business party, the initial model service corresponding to the above business party identifier can be determined based on this corresponding relationship, that is, the initial model service of the business party.

[0107] From the above description, it can be seen that in this embodiment, by obtaining the business party identifier of the business party and determining the initial model service corresponding to the business party identifier as the initial model service of the business party, it is possible to accurately determine the initial model service of the business party based on the business party identifier. Furthermore, it can lay a foundation for subsequently determining the target model service group based on the initial model service and determining the target model service from each model service within the target model service group, enabling cross - domain model recommendation, and reducing the software and hardware costs of the model recommendation service provider as well as the labor costs for model evaluation.

[0108] Figure 3 It is a flowchart showing how to determine the target model service group to which the initial model service belongs according to an exemplary embodiment; this embodiment may further include determining the target model service group to which the initial model service belongs based on the following steps S301 - S305 on the basis of the above embodiment:

[0109] In step S301, detect the target interface declaration set information of the initial model service.

[0110] In this embodiment, the model file of the initial model service can be detected based on a preset model detection method, so as to determine the target interface declaration set information of the initial model service.

[0111] Among them, the above preset model detection method can be selected from related technologies according to actual business needs. For example, a manual detection method or an automatic detection method can be adopted, and this embodiment does not limit this.

[0112] For example, the target interface declaration set information of the above initial model service may include the interface type of the initial model service and the corresponding data dimension, etc.

[0113] In step S302, search for a target model service group that matches the target interface declaration set information.

[0114] In this embodiment, after detecting the target interface declaration set information of the initial model service, a target model service group that matches the target interface declaration set information can be searched for.

[0115] Among them, each model service included in the target model service group has the same interface declaration set information as the initial model service.

[0116] For example, based on the correspondence relationship pre-established between the interface declaration set information and the model service group, a target model service group that matches the target interface declaration set information can be searched for.

[0117] In step S303, whether a target model service group is found: if yes, execute step S304; if no, execute step S305.

[0118] In this embodiment, after performing the operation of searching for a target model service group that matches the target interface declaration set information, it can be determined whether a target model service group is found. If found, the following step S304 can be continued; on the contrary, if not found, the following step S305 can be executed first and then step S304.

[0119] In step S304, add the initial model service to the target model service group.

[0120] In this embodiment, when the target model service group is found, the initial model service can be added to the target model service group, so that the association relationship between the initial model service and the target model service group can be constructed, and then the model for recommendation to the business party can be determined based on other model services in the target model service group.

[0121] In step S305, in response to the failure to find the target model service group, create the target model service group to which the initial model service belongs, and register the target interface declaration set information under the target model service group.

[0122] In this embodiment, when the target model service group is not found, the target model service group to which the initial model service belongs can be created, and the target interface declaration set information can be registered under the target model service group. On this basis, step S304 can be executed, that is, the initial model service is added to the newly created target model service group.

[0123] That is to say, when there is currently no target model service group to which the initial model service belongs, a target model service group can be newly created for the initial model service, and the target interface declaration set information of the initial model service can be registered under the target model service group, so as to facilitate the subsequent addition of more model services to the target model service group based on the target interface declaration set information, so that the model for recommendation to the business party can be determined based on other model services in the target model service group.

[0124] As can be seen from the above description, in this embodiment, by detecting the target interface declaration set information of the initial model service and finding the target model service group that matches the target interface declaration set information, and then in response to finding the target model service group, adding the initial model service to the target model service group, or in response to not finding the target model service group, creating the target model service group to which the initial model service belongs, and registering the target interface declaration set information under the target model service group, so as to add the initial model service to the created target model service group, the accurate division of the target model service group to which the initial model service belongs can be realized, and then the target model service can be determined from each model service in the target model service group, and the model recommendation can be realized by using the existing model services without introducing additional service resources, thus reducing the software and hardware costs of the model recommendation service provider and the labor costs for model evaluation.

[0125] Figure 4is a flowchart showing how to determine a target model service from each model service within the target model service group according to an exemplary embodiment; on the basis of the above embodiment, this embodiment takes how to determine a target model service from each model service within the target model service group as an example for exemplary illustration. As Figure 4 shown, the step of determining a target model service from each model service within the target model service group in the above step S103 may include the following steps S401 - S402:

[0126] In step S401, obtain the evaluation information values of each model service within the target model service group.

[0127] In this embodiment, after determining the target model service group based on the initial model service, the evaluation information values of each model service within the target model service group can be obtained.

[0128] Among them, the above evaluation information value may include an evaluation score determined in advance during the operation of each model service, and this evaluation score can be used to reflect information such as the performance of each model service.

[0129] In step S402, determine the target model service based on the model service with the highest evaluation information value among each model service.

[0130] In this embodiment, after obtaining the evaluation information values of each model service within the target model service group, the target model service can be determined based on the model service with the highest evaluation information value among each model service.

[0131] For example, after obtaining the evaluation information values of each model service within the target model service group, each model service can be sorted based on the evaluation information values of each model service within this target model service group, and then the model service with the highest evaluation information value can be selected as the target model service based on the sorting result.

[0132] From the above description, it can be seen that in this embodiment, by obtaining the evaluation information values of each model service within the target model service group and determining the target model service based on the model service with the highest evaluation information value among each model service, it is possible to determine the target model service from within the target model service group based on the evaluation information values of each model service, which can improve the accuracy and rationality of determining the target model service, and further can realize the subsequent determination of the model for recommendation to the business party based on the model corresponding to the target model service.

[0133] Figure 5is a flowchart showing how to determine the evaluation information value of the model service according to an exemplary embodiment; this embodiment takes how to determine the evaluation information value of the model service as an example for exemplary illustration on the basis of the above embodiment. As Figure 5 shown, this embodiment may further include determining the evaluation information value of the model service based on the following steps S501 - S502 on the basis of the above embodiment:

[0134] In step S501, the actual operation data of the model service is collected based on a preset period.

[0135] In this embodiment, after the model service is deployed to the current cluster, corresponding services can be provided to the upstream service callers of the business based on the external interface of the model service, and the actual operation data of the model service can be collected based on a preset period.

[0136] Among them, the above preset period can be set according to actual needs, and this embodiment does not limit this.

[0137] In one embodiment, the above operation data includes but is not limited to at least one of the single - call time consumption, the number of calls per second, the CPU occupancy rate consumed by the model, the memory size consumed by the model, etc., and this embodiment does not limit this.

[0138] In step S502, in response to the actual operation data meeting the evaluation conditions, the evaluation information value of the model service is determined based on the actual operation data.

[0139] In this embodiment, after the actual operation data of the model service is collected based on a preset period, in response to the actual operation data meeting the evaluation conditions, the evaluation information value of the model service is determined based on the actual operation data.

[0140] In one embodiment, the situation where the above actual operation data meets the evaluation conditions can be set according to actual business requirements, such as setting the data collection duration to be greater than or equal to a set duration threshold, and / or the number of calls to the initial model service to be greater than or equal to a set number threshold, etc., and this embodiment does not limit this.

[0141] For example, Figure 6 is a flowchart showing how to determine the evaluation information value of the model service based on the actual operation data according to an exemplary embodiment. As Figure 6 shown, the determination of the evaluation information value of the model service based on the actual operation data in the above step S502 may further include:

[0142] In step S601, the actual operation data is pre - processed to obtain pre - processed data.

[0143] In this embodiment, after collecting the actual operation data of the model service based on a preset period, the actual operation data can be preprocessed. For example, the units of the actual operation data of the above model service can be unified to obtain the data with unified units, that is, the preprocessed data.

[0144] In step S602, the preprocessed data is normalized to obtain normalized data.

[0145] In this embodiment, after preprocessing the actual operation data to obtain preprocessed data, the preprocessed data can be normalized. For example, various actual operation data of the same model service are normalized to obtain normalized data.

[0146] In step S603, a weighted sum operation is performed on the normalized data to obtain the evaluation information value of the model service.

[0147] In this embodiment, after normalizing the preprocessed data to obtain normalized data, a weighted sum operation can be performed on the normalized data to obtain the evaluation information value of the model service.

[0148] For example, the weights of various actual operation data can be determined based on the preset importance levels of various actual operation data in the normalized data, and then a weighted sum operation is performed based on the normalized data and the corresponding weights, so that the evaluation information value of the model service can be obtained.

[0149] As can be seen from the above description, in this embodiment, by collecting the actual operation data of the model service based on a preset period and in response to the actual operation data satisfying the evaluation conditions, determining the evaluation information value of the model service based on the actual operation data, it is possible to accurately determine the evaluation information value of the model service based on the collected actual operation data of the model service. Thus, it is possible to determine the target model service based on the model service with the highest evaluation information value among each model service subsequently, which can improve the accuracy of determining the target model service, and thus can improve the accuracy of determining the model for recommendation to the business party based on the model corresponding to the target model service.

[0150] Figure 7 is a block diagram of a device for determining a recommendation model shown according to an exemplary embodiment; the device of this embodiment can be applied to a cluster including multiple node devices, and the node devices include physical machines, virtual machines, containers, etc. As Figure 7 shown, the device includes: an initial service determination module 110, a service group determination module 120, a target service determination module 130, and a recommendation model determination module 140, where:

[0151] An initial service determination module 110, configured to determine an initial model service for the service provider in response to detecting that the service provider meets the model recommendation condition;

[0152] A service group determination module 120, configured to determine a target model service group based on the initial model service, where the target model service group is a pre-determined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information;

[0153] A target service determination module 130, configured to determine a target model service from each model service in the target model service group;

[0154] A recommended model determination module 140, configured to determine a model for recommending to the service provider based on the model corresponding to the target model service.

[0155] As can be seen from the above description, the device in this embodiment determines the initial model service for the service provider by responding to detecting that the service provider meets the model recommendation condition, determines the target model service group based on the initial model service, then determines the target model service from each model service in the target model service group, and further determines the model for recommending to the service provider based on the model corresponding to the target model service. Since the target model service group is a pre-determined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information, it is possible to determine the target model service based on the model service in the group to which the initial model service belongs and having the same interface declaration set information as the initial model service when detecting that the service provider meets the model recommendation condition, and further determine the model for recommending to the service provider, which can realize model recommendation using existing model services without introducing additional service resources, and can not only realize cross-domain model recommendation, but also reduce the software and hardware costs of the model recommendation service provider and the labor costs for model evaluation.

[0156] Figure 8 It is a block diagram of a device for determining a recommended model shown according to another exemplary embodiment; the device in this embodiment can be applied to a cluster including multiple node devices, and the node devices include physical machines, virtual machines, containers, etc. Among them, the initial service determination module 210, the service group determination module 220, the target service determination module 230, and the recommended model determination module 240 have the same functions as the initial service determination module 110, the service group determination module 120, the target service determination module 130, and the recommended model determination module 140 in the foregoing Figure 7 shown embodiment, and will not be elaborated here. As Figure 8 shown, the initial service determination module 210 may include:

[0157] A service identifier acquisition unit 211, configured to acquire a service identifier of the service provider.

[0158] An initial service determination unit 212, configured to determine the initial model service corresponding to the service identifier as the initial model service of the service provider.

[0159] In one embodiment, the above device may further include an affiliated group determination module 250;

[0160] The affiliated group determination module 250 may include:

[0161] A set information detection unit 251, configured to detect target interface declaration set information of the initial model service;

[0162] An affiliated group determination unit 252, configured to find a target model service group that matches the target interface declaration set information;

[0163] A target group joining unit 253, configured to, in response to finding the target model service group, add the initial model service to the target model service group.

[0164] In one embodiment, the above affiliated group determination module may further include:

[0165] A target group creation unit 254, configured to, in response to not finding the target model service group, create a target model service group to which the initial model service belongs, and register the target interface declaration set information under the target model service group;

[0166] The target group joining unit 253 may also be configured to add the initial model service to the created target model service group.

[0167] In one embodiment, the above target service determination module 230 may include:

[0168] An evaluation information value acquisition unit 231, configured to acquire evaluation information values of each model service in the target model service group;

[0169] A target service determination unit 232, configured to determine a target model service based on the model service with the highest evaluation information value among the respective model services.

[0170] In one embodiment, the above device may further include an evaluation information value determination module 260;

[0171] The evaluation information value determination module 260 may include:

[0172] An operation data acquisition unit 261, configured to acquire actual operation data of the model service based on a preset period;

[0173] An evaluation information value determination unit 262, configured to determine an evaluation information value of the model service based on the actual operation data in response to the actual operation data satisfying an evaluation condition.

[0174] In one embodiment, the above-mentioned evaluation information value determination unit 262 may further be configured to:

[0175] Preprocess the actual operation data to obtain preprocessed data;

[0176] Normalize the preprocessed data to obtain normalized data;

[0177] Perform a weighted summation operation on the normalized data to obtain the evaluation information value of the model service.

[0178] Regarding the device in the above embodiment, the specific manners in which each module performs operations have been described in detail in the embodiment related to the method, and will not be elaborated herein.

[0179] Figure 9 is a block diagram of an electronic device shown according to an exemplary embodiment. For example, the device 900 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0180] Refer to Figure 9 , the device 900 may include one or more of the following components: a processing component 902, a memory 904, a power component 906, a multimedia component 908, an audio component 910, an input / output (I / O) interface 912, a sensor component 914, and a communication component 916.

[0181] The processing component 902 generally controls the overall operation of the device 900, such as operations associated with display, telephone calls, data communication, camera operations, and recording operations. The processing component 902 may include one or more processors 920 to execute instructions to complete all or part of the steps of the above method. In addition, the processing component 902 may include one or more modules to facilitate the interaction between the processing component 902 and other components. For example, the processing component 902 may include a multimedia module to facilitate the interaction between the multimedia component 908 and the processing component 902.

[0182] The memory 904 is configured to store various types of data to support the operation of the device 900. Examples of such data include instructions for any application or method operating on the device 900, contact data, phone book data, messages, pictures, videos, and the like. The memory 904 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disks, or optical disks.

[0183] The power supply component 906 provides power to various components of the device 900. The power supply component 906 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 900.

[0184] The multimedia component 908 includes a screen that provides an output interface between the device 900 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can not only sense the boundaries of touch or swipe actions but also detect the duration and pressure associated with the touch or swipe operation. In some embodiments, the multimedia component 908 includes a front camera and / or a rear camera. When the device 900 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each of the front camera and the rear camera can be a fixed optical lens system or have a focal length and optical zoom capabilities.

[0185] The audio component 910 is configured to output and / or input audio signals. For example, the audio component 910 includes a microphone (MIC) that is configured to receive external audio signals when the device 900 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals can be further stored in the memory 904 or transmitted via the communication component 916. In some embodiments, the audio component 910 further includes a speaker for outputting audio signals.

[0186] The I / O interface 912 provides an interface between the processing component 902 and a peripheral interface module, which can be a keyboard, a click wheel, buttons, etc. These buttons can include, but are not limited to: a home button, a volume button, a power-on button, and a lock button.

[0187] The sensor assembly 914 includes one or more sensors for providing an assessment of the status of the device 900 in various aspects. For example, the sensor assembly 914 can detect the on / off state of the device 900, the relative positioning of components, such as the display and keypad of the device 900. The sensor assembly 914 can also detect a change in the position of the device 900 or a component of the device 900, the presence or absence of user contact with the device 900, the orientation or acceleration / deceleration of the device 900, and a change in the temperature of the device 900. The sensor assembly 914 can also include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 914 can also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 914 can also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0188] The communication component 916 is configured to facilitate communication between the device 900 and other devices in a wired or wireless manner. The device 900 can access a wireless network based on communication standards, such as WiFi, 2G or 3G, 4G or 5G, or a combination thereof. In an exemplary embodiment, the communication component 916 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 916 further includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0189] In an exemplary embodiment, the device 900 can be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing the above-described methods.

[0190] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions, such as a memory 904 including instructions, is also provided. The above instructions can be executed by the processor 920 of the device 900 to complete the above-described methods. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0191] Other embodiments of the present disclosure will be readily apparent to those skilled in the art upon consideration of the specification and practice of the disclosure herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include known common general knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and examples are only to be considered as exemplary, and the true scope and spirit of the present disclosure are pointed out by the following claims.

[0192] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A method for determining a recommendation model, characterized in that, the method includes: In response to detecting that a business party meets the model recommendation condition, determining an initial model service for the business party, where the initial model service is a service for the business party to perform inference using a model; Based on the initial model service, determining a target model service group, where the target model service group is a pre-determined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information; Determining a model service with better performance than the initial model service from each model service in the target model service group as the target model service; Based on the model corresponding to the target model service, determining a model for recommending to the business party; The determining a model service with better performance than the initial model service from each model service in the target model service group as the target model service includes: Obtaining the evaluation information values of each model service in the target model service group; Based on the model service with the highest evaluation information value among each model service, determining the target model service.

2. The method according to claim 1, characterized in that, the determining the initial model service for the business party includes: Obtaining the business party identifier of the business party; Determining the initial model service corresponding to the business party identifier as the initial model service for the business party.

3. The method according to claim 1, characterized in that, the method further includes determining the target model service group to which the initial model service belongs based on the following method: Detecting the target interface declaration set information of the initial model service; Searching for a target model service group that matches the target interface declaration set information; In response to finding the target model service group, adding the initial model service to the target model service group.

4. The method according to claim 3, characterized in that, the method further includes: In response to not finding the target model service group, creating a target model service group to which the initial model service belongs, and registering the target interface declaration set information under the target model service group; Adding the initial model service to the created target model service group.

5. The method according to claim 1, characterized in that, the method further includes pre-determining the evaluation information value of the model service based on the following method: Collecting the actual operation data of the model service based on a preset period; In response to the actual operation data meeting the evaluation condition, determining the evaluation information value of the model service based on the actual operation data.

6. The method according to claim 5, characterized in that, the determining the evaluation information value of the model service based on the actual operation data includes: Performing preprocessing on the actual operation data to obtain preprocessed data; Performing normalization processing on the preprocessed data to obtain normalized data; Performing a weighted sum operation on the normalized data to obtain the evaluation information value of the model service.

7. A device for determining a recommendation model, characterized in that, the device includes: An initial service determination module, configured to determine an initial model service for the service provider in response to detecting that the service provider meets the model recommendation condition, where the initial model service is a service for the service provider to perform inference using a model; A service group determination module, configured to determine a target model service group based on the initial model service, where the target model service group is a pre-determined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information; A target service determination module, configured to determine, from each model service in the target model service group, a model service with better performance than the initial model service as the target model service; A recommended model determination module, configured to determine a model for recommending to the service provider based on the model corresponding to the target model service; The target service determination module includes: An evaluation information value acquisition unit, configured to acquire the evaluation information values of each model service in the target model service group; A target service determination unit, configured to determine a target model service based on the model service with the highest evaluation information value among each model service; 8. An electronic device characterized in that the device includes: a processor and a memory for storing processor-executable instructions; wherein the processor is configured to: determine an initial model service for the service provider in response to detecting that the service provider meets the model recommendation condition, where the initial model service is a service for the service provider to perform inference using a model; determine a target model service group based on the initial model service, where the target model service group is a pre-determined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information; determine, from each model service in the target model service group, a model service with better performance than the initial model service as the target model service; determine a model for recommending to the service provider based on the model corresponding to the target model service; The determining, from each model service in the target model service group, a model service with better performance than the initial model service as the target model service includes: acquiring the evaluation information values of each model service in the target model service group; determining a target model service based on the model service with the highest evaluation information value among each model service; 9. A computer-readable storage medium, on which a computer program is stored characterized in that when the program is executed by a processor, it realizes: determining an initial model service for the service provider in response to detecting that the service provider meets the model recommendation condition, where the initial model service is a service for the service provider to perform inference using a model; determining a target model service group based on the initial model service, where the target model service group is a pre-determined group to which the initial model service belongs, and each model service in the target model service group has the same interface declaration set information; Determine, from each model service within the target model service group, a model service with better performance than the initial model service as the target model service; Based on the model corresponding to the target model service, determine a model for recommendation to the business party; The determining, from each model service within the target model service group, a model service with better performance than the initial model service as the target model service includes: Obtain the evaluation information values of each model service within the target model service group; Based on the model service with the highest evaluation information value among the model services, determine the target model service.

Images