Model database management method and system based on parameter matching

Abstract

The present invention discloses a model database management method and system based on parameter matching, the method comprising: obtaining multiple algorithm model data uploaded by multiple users; grouping all the algorithm model data according to the model parameters and user parameters corresponding to each of the algorithm model data to obtain multiple model sets; in response to a website access operation of a target user, screening out a target model set from the multiple model sets according to the user parameters of the target user; and determining multiple recommended model data in the target model set according to the website access operation. It can be seen that the present invention can effectively improve the management efficiency and grouping rationality of model data, provide users with more reasonable model recommendation services, and improve user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a model database management method and system based on parameter matching. Background Art

[0002] As artificial intelligence technology continues to develop and gain attention, many technicians or technical institutions have begun to use algorithm technology to train and research algorithm models, and upload their trained algorithm models on data sharing platforms or computing platforms. How to effectively and efficiently manage these algorithm models has become an important technical issue. Most of the existing algorithm model data management technologies only implement the storage and search functions of model data, without further considering user parameters to group and recommend model data. Therefore, the management effect is general and the user experience is poor. It can be seen that the existing technology has defects that need to be solved urgently. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a model database management method and system based on parameter matching, which can effectively improve the management efficiency and grouping rationality of model data, provide users with more reasonable model recommendation services, and improve user experience.

[0004] In order to solve the above technical problems, the first aspect of the present invention discloses a model database management method based on parameter matching, the method comprising:

[0005] Obtain multiple algorithm model data uploaded by multiple users;

[0006] According to the model parameters and user parameters corresponding to each of the algorithm model data, all the algorithm model data are grouped to obtain multiple model sets;

[0007] In response to a website access operation of a target user, filtering a target model set from the multiple model sets according to a user parameter of the target user;

[0008] According to the website access operation, a plurality of recommended model data are determined in the target model set.

[0009] As an optional embodiment, in the first aspect of the present invention, the model parameters include model size, model purpose, model training data type, model architecture, model input and output data type and model historical usage record.

[0010] As an optional implementation, in the first aspect of the present invention, the user parameters include user type, user level, number of user uploaded models, user historical upload records and user historical download records.

[0011] As an optional implementation, in the first aspect of the present invention, all the algorithm model data are grouped according to the model parameters and user parameters corresponding to each of the algorithm model data to obtain multiple model sets, including:

[0012] Calculate the similarity of the model parameters and the user parameters between any two of the algorithm model data to obtain the parameter similarity between the two algorithm model data;

[0013] According to the parameter similarity, based on a dynamic programming algorithm, all the algorithm model data are grouped to obtain multiple model sets.

[0014] As an optional implementation, in the first aspect of the present invention, the calculating the similarity of the model parameters and the user parameters between any two of the algorithm model data to obtain the parameter similarity between the two algorithm model data includes:

[0015] For any two of the algorithm model data, calculating a first similarity between the model parameters corresponding to the two algorithm model data;

[0016] Calculating a second similarity between the users corresponding to the two algorithm model data;

[0017] The product of the first similarity and the second similarity is calculated to obtain the parameter similarity between the two algorithm models.

[0018] As an optional implementation, in the first aspect of the present invention, all the algorithm model data are grouped according to the parameter similarity based on a dynamic programming algorithm to obtain multiple model sets, including:

[0019] The objective function is set to maximize the number of algorithm model data in each model set;

[0020] The setting restriction condition includes that the parameter similarity between any two algorithm model data in the same model set is greater than a first similarity threshold, and the parameter similarity between any two algorithm model data belonging to different model sets is less than a second similarity threshold; the second similarity threshold is less than the first similarity threshold;

[0021] Based on the spatial search algorithm, according to the objective function and the restriction conditions, all the algorithm model data are iteratively grouped and calculated until the optimal result is obtained, thereby obtaining multiple model sets.

[0022] As an optional implementation, in the first aspect of the present invention, the step of selecting a target model set from the multiple model sets according to the user parameter of the target user includes:

[0023] For each of the model sets, vectorize the union of the user parameters corresponding to all the algorithm model data in the model set to determine the user parameter vector corresponding to the model set;

[0024] Calculate the vectorization result corresponding to the user parameters of the target user;

[0025] Calculating the vector distance between the user parameter vector and the vectorization result to obtain the user parameter similarity corresponding to the model set;

[0026] All the model sets whose user parameter similarity is higher than a third similarity threshold are determined as target model sets.

[0027] As an optional implementation, in the first aspect of the present invention, the determining of a plurality of recommendation model data in the target model set according to the website access operation includes:

[0028] Determine a cursor movement path and a cursor click object corresponding to the website access operation;

[0029] According to the cursor movement path, a plurality of path passing areas are determined on the current browser page;

[0030] According to the cursor click object, determining a plurality of clicked objects from a plurality of model objects of the current browser page;

[0031] Determine the model parameters corresponding to all model objects in the areas where all the paths pass as a first model parameter set;

[0032] Determine the model parameters corresponding to all the clicked objects as a second model parameter set;

[0033] For each model data in the target model set, calculating a third similarity between a model parameter of the model data and the first model parameter set;

[0034] Calculating a fourth similarity between the model parameters of the model data and the second model parameter set;

[0035] Calculating the product of the third similarity and the fourth similarity to obtain a priority parameter corresponding to the model data;

[0036] All the model data whose priority parameters are greater than a preset parameter threshold are determined as recommended model data.

[0037] A second aspect of an embodiment of the present invention discloses a model database management system based on parameter matching, the system comprising:

[0038] An acquisition module is used to acquire multiple algorithm model data uploaded by multiple users;

[0039] A grouping module, used to group all the algorithm model data according to the model parameters and user parameters corresponding to each of the algorithm model data to obtain multiple model sets;

[0040] A screening module, configured to screen a target model set from the plurality of model sets in response to a website access operation of a target user and according to a user parameter of the target user;

[0041] A determination module is used to determine a plurality of recommended model data in the target model set according to the website access operation.

[0042] As an optional embodiment, in the second aspect of the present invention, the model parameters include model size, model purpose, model training data type, model architecture, model input and output data type and model historical usage records.

[0043] As an optional implementation, in the second aspect of the present invention, the user parameters include user type, user level, number of user uploaded models, user historical upload records and user historical download records.

[0044] As an optional implementation, in the second aspect of the present invention, the grouping module groups all the algorithm model data according to the model parameters and user parameters corresponding to each of the algorithm model data to obtain a specific manner of multiple model sets, including:

[0045] Calculate the similarity of the model parameters and the user parameters between any two of the algorithm model data to obtain the parameter similarity between the two algorithm model data;

[0046] According to the parameter similarity, based on a dynamic programming algorithm, all the algorithm model data are grouped to obtain multiple model sets.

[0047] As an optional implementation, in the second aspect of the present invention, the grouping module calculates the similarity of the model parameters and the user parameters between any two of the algorithm model data, and obtains the specific manner of the parameter similarity between the two algorithm model data, including:

[0048] For any two of the algorithm model data, calculating a first similarity between the model parameters corresponding to the two algorithm model data;

[0049] Calculating a second similarity between the users corresponding to the two algorithm model data;

[0050] The product of the first similarity and the second similarity is calculated to obtain the parameter similarity between the two algorithm models.

[0051] As an optional implementation, in the second aspect of the present invention, the grouping module groups all the algorithm model data according to the parameter similarity based on a dynamic programming algorithm to obtain a specific method of multiple model sets, including:

[0052] The objective function is set to maximize the number of algorithm model data in each model set;

[0053] The setting restriction condition includes that the parameter similarity between any two algorithm model data in the same model set is greater than a first similarity threshold, and the parameter similarity between any two algorithm model data belonging to different model sets is less than a second similarity threshold; the second similarity threshold is less than the first similarity threshold;

[0054] Based on the spatial search algorithm, according to the objective function and the restriction conditions, all the algorithm model data are iteratively grouped and calculated until the optimal result is obtained, thereby obtaining multiple model sets.

[0055] As an optional implementation, in the second aspect of the present invention, the specific manner in which the screening module screens out the target model set from the multiple model sets according to the user parameters of the target user includes:

[0056] For each of the model sets, vectorize the union of the user parameters corresponding to all the algorithm model data in the model set to determine the user parameter vector corresponding to the model set;

[0057] Calculate the vectorization result corresponding to the user parameters of the target user;

[0058] Calculating the vector distance between the user parameter vector and the vectorization result to obtain the user parameter similarity corresponding to the model set;

[0059] All the model sets whose user parameter similarity is higher than a third similarity threshold are determined as target model sets.

[0060] As an optional implementation, in the second aspect of the present invention, the specific manner in which the determination module determines a plurality of recommendation model data in the target model set according to the website access operation includes:

[0061] Determine a cursor movement path and a cursor click object corresponding to the website access operation;

[0062] According to the cursor movement path, a plurality of path passing areas are determined on the current browser page;

[0063] According to the cursor click object, determining a plurality of clicked objects from a plurality of model objects of the current browser page;

[0064] Determine the model parameters corresponding to all model objects in the areas where all the paths pass as a first model parameter set;

[0065] Determine the model parameters corresponding to all the clicked objects as a second model parameter set;

[0066] For each model data in the target model set, calculating a third similarity between a model parameter of the model data and the first model parameter set;

[0067] Calculating a fourth similarity between the model parameters of the model data and the second model parameter set;

[0068] Calculating the product of the third similarity and the fourth similarity to obtain a priority parameter corresponding to the model data;

[0069] All the model data whose priority parameters are greater than a preset parameter threshold are determined as recommended model data.

[0070] The third aspect of the present invention discloses another model database management system based on parameter matching, the system comprising:

[0071] A memory storing executable program code;

[0072] a processor coupled to the memory;

[0073] The processor calls the executable program code stored in the memory to execute part or all of the steps in the model database management method based on parameter matching disclosed in the first aspect of the present invention.

[0074] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute part or all of the steps in the model database management method based on parameter matching disclosed in the first aspect of the present invention.

[0075] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0076] The present invention can accurately and reasonably group the model data uploaded by users based on model parameters and user parameters, and then perform set screening and recommendation based on the parameters and operations of the target users to determine multiple more matching recommended model data, thereby effectively improving the management efficiency and grouping rationality of model data, providing users with more reasonable model recommendation services, and improving user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0078] Figure 1 It is a flow chart of a model database management method based on parameter matching disclosed in an embodiment of the present invention.

[0079] Figure 2 It is a structural diagram of a model database management system based on parameter matching disclosed in an embodiment of the present invention.

[0080] Figure 3 It is a structural diagram of another model database management system based on parameter matching disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0081] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0082] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.

[0083] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0084] The present invention discloses a model database management method and system based on parameter matching, which can accurately and reasonably group the model data uploaded by users based on model parameters and user parameters, and then perform set screening and recommendation based on the parameters and operations of the target users to determine a plurality of more matching recommended model data, thereby effectively improving the management efficiency and grouping rationality of the model data, providing users with more reasonable model recommendation services, and improving user experience. The following are detailed descriptions.

[0085] Embodiment 1

[0086] See also Figure 1 , Figure 1 : is a flow chart of a model database management method based on parameter matching disclosed in an embodiment of the present invention. Figure 1 The model database management method based on parameter matching described above can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 1 As shown, the model database management method based on parameter matching may include the following operations:

[0087] 101. Obtain multiple algorithm model data uploaded by multiple users.

[0088] 102. According to the model parameters and user parameters corresponding to each algorithm model data, all the algorithm model data are grouped to obtain multiple model sets.

[0089] 103. In response to a website access operation of a target user, a target model set is screened out from a plurality of model sets according to a user parameter of the target user.

[0090] 104. According to the website access operation, a plurality of recommended model data are determined in the target model set.

[0091] It can be seen that the above-mentioned embodiments of the invention can accurately and reasonably group the model data uploaded by users based on model parameters and user parameters, and then perform collection screening and recommendation based on the parameters and operations of the target users to determine multiple more matching recommended model data, thereby effectively improving the management efficiency and grouping rationality of model data, providing users with more reasonable model recommendation services, and improving user experience.

[0092] As an optional embodiment, in the above steps, the model parameters include model size, model purpose, model training data type, model architecture, model input and output data type and model historical usage records.

[0093] It can be seen that through the above optional embodiments, the content of the model parameters is limited to more comprehensively characterize the characteristics of the model data, assist in improving the management efficiency and grouping rationality of the model data, provide users with more reasonable model recommendation services, and improve user experience.

[0094] As an optional embodiment, in the above steps, the user parameters include user type, user level, number of models uploaded by the user, user historical upload records, and user historical download records.

[0095] It can be seen that through the above optional embodiments, the content of user parameters is limited to more comprehensively characterize the characteristics of the uploading user, assist in improving the management efficiency and grouping rationality of model data, provide users with more reasonable model recommendation services, and improve user experience.

[0096] As an optional embodiment, in the above steps, all algorithm model data are grouped according to the model parameters and user parameters corresponding to each algorithm model data to obtain multiple model sets, including:

[0097] Calculate the similarity of model parameters and user parameters between any two algorithm model data to obtain the parameter similarity between the two algorithm model data;

[0098] According to parameter similarity, based on the dynamic programming algorithm, all algorithm model data are grouped to obtain multiple model sets.

[0099] It can be seen that through the above optional embodiments, it is possible to accurately and reasonably group the model data uploaded by users based on similarity calculation and dynamic programming algorithm, assist in improving the management efficiency and grouping rationality of model data, provide users with more reasonable model recommendation services, and improve user experience.

[0100] As an optional embodiment, in the above steps, calculating the similarity of the model parameters and the user parameters between any two algorithm model data to obtain the parameter similarity between the two algorithm model data includes:

[0101] For any two algorithm model data, calculating a first similarity between model parameters corresponding to the two algorithm model data;

[0102] Calculating a second similarity between users corresponding to the two algorithm model data;

[0103] The product of the first similarity and the second similarity is calculated to obtain the parameter similarity between the two algorithm models.

[0104] It can be seen that through the above-mentioned optional embodiments, the parameter similarity between the model data can be calculated by similarity calculation and product calculation of the model parameters of the model data and the user parameters, so as to facilitate the subsequent accurate and reasonable grouping of the model data uploaded by the user, assist in improving the management efficiency and grouping rationality of the model data, provide users with more reasonable model recommendation services, and improve user experience.

[0105] As an optional embodiment, in the above steps, all algorithm model data are grouped according to parameter similarity based on a dynamic programming algorithm to obtain multiple model sets, including:

[0106] The objective function is set to maximize the number of algorithm model data in each model set;

[0107] The setting restriction condition includes that the parameter similarity between any two algorithm model data in the same model set is greater than a first similarity threshold, and the parameter similarity between any two algorithm model data belonging to different model sets is less than a second similarity threshold; optionally, the second similarity threshold is less than the first similarity threshold;

[0108] Based on the spatial search algorithm, according to the objective function and constraints, all algorithm model data are iteratively grouped and calculated until the optimal result is obtained, and multiple model sets are obtained.

[0109] It can be seen that through the above optional embodiments, it is possible to accurately and reasonably group the model data uploaded by users based on the preset objective function and constraints and based on the spatial search algorithm, thereby helping to improve the management efficiency and grouping rationality of the model data, providing users with more reasonable model recommendation services, and improving the user experience.

[0110] As an optional embodiment, in the above step, selecting a target model set from multiple model sets according to the user parameters of the target user includes:

[0111] For each model set, vectorize the union of user parameters corresponding to all algorithm model data in the model set to determine the user parameter vector corresponding to the model set;

[0112] Calculate the vectorized results corresponding to the user parameters of the target user;

[0113] Calculate the vector distance between the user parameter vector and the vectorized result to obtain the user parameter similarity corresponding to the model set;

[0114] All model sets whose user parameter similarity is higher than a third similarity threshold are determined as target model sets.

[0115] It can be seen that through the above optional embodiments, the user parameter similarity corresponding to the model set can be calculated based on the vector distance between the user parameters, and then the target model set can be screened out based on this, so as to subsequently determine more matching recommended model data, thereby improving the management efficiency and grouping rationality of the model data, providing users with more reasonable model recommendation services, and improving user experience.

[0116] As an optional embodiment, in the above step, according to the website access operation, a plurality of recommended model data are determined in the target model set, including:

[0117] Determine the cursor movement path and cursor click object corresponding to the website access operation;

[0118] According to the cursor movement path, multiple path passing areas are determined on the current browser page;

[0119] According to the cursor click object, multiple clicked objects are determined from multiple model objects on the current browser page;

[0120] Determine the model parameters corresponding to all model objects in the areas where all paths pass as a first model parameter set;

[0121] Determine the model parameters corresponding to all clicked objects as a second model parameter set;

[0122] For each model data in the target model set, calculating a third similarity between a model parameter of the model data and the first model parameter set;

[0123] Calculate a fourth similarity between the model parameters of the model data and a second model parameter set;

[0124] Calculate the product of the third similarity and the fourth similarity to obtain the priority parameter corresponding to the model data;

[0125] All model data with priority parameters greater than a preset parameter threshold are determined as recommended model data.

[0126] It can be seen that through the above optional embodiments, multiple model objects that the user may be interested in can be determined based on the user's website access operations, and based on the similarity calculation and screening between the model parameters, multiple more matching recommended model data can be accurately determined, thereby improving the management efficiency and grouping rationality of the model data, providing users with more reasonable model recommendation services, and improving user experience.

[0127] Embodiment 2

[0128] See also Figure 2 , Figure 2 : is a schematic diagram of a model database management system based on parameter matching disclosed in an embodiment of the present invention. Figure 2 The model database management system based on parameter matching described above can be applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 2 As shown, the model database management system based on parameter matching may include:

[0129] The acquisition module 201 is used to acquire multiple algorithm model data uploaded by multiple users.

[0130] The grouping module 202 is used to group all the algorithm model data according to the model parameters and user parameters corresponding to each algorithm model data to obtain multiple model sets.

[0131] The screening module 203 is used to screen out a target model set from multiple model sets in response to a website access operation of the target user and according to user parameters of the target user.

[0132] The determination module 204 is used to determine a plurality of recommendation model data in the target model set according to the website access operation.

[0133] It can be seen that the above-mentioned embodiments of the invention can accurately and reasonably group the model data uploaded by users based on model parameters and user parameters, and then perform collection screening and recommendation based on the parameters and operations of the target users to determine multiple more matching recommended model data, thereby effectively improving the management efficiency and grouping rationality of model data, providing users with more reasonable model recommendation services, and improving user experience.

[0134] As an optional embodiment, the model parameters include model size, model purpose, model training data type, model architecture, model input and output data type, and model historical usage records.

[0135] It can be seen that through the above optional embodiments, the content of the model parameters is limited to more comprehensively characterize the characteristics of the model data, assist in improving the management efficiency and grouping rationality of the model data, provide users with more reasonable model recommendation services, and improve user experience.

[0136] As an optional embodiment, the user parameters include user type, user level, number of user uploaded models, user historical upload records, and user historical download records.

[0137] It can be seen that through the above optional embodiments, the content of user parameters is limited to more comprehensively characterize the characteristics of the uploading user, assist in improving the management efficiency and grouping rationality of model data, provide users with more reasonable model recommendation services, and improve user experience.

[0138] As an optional embodiment, the grouping module groups all algorithm model data according to the model parameters and user parameters corresponding to each algorithm model data to obtain a specific method of multiple model sets, including:

[0139] Calculate the similarity of model parameters and user parameters between any two algorithm model data to obtain the parameter similarity between the two algorithm model data;

[0140] According to parameter similarity, based on the dynamic programming algorithm, all algorithm model data are grouped to obtain multiple model sets.

[0141] It can be seen that through the above optional embodiments, it is possible to accurately and reasonably group the model data uploaded by users based on similarity calculation and dynamic programming algorithm, assist in improving the management efficiency and grouping rationality of model data, provide users with more reasonable model recommendation services, and improve user experience.

[0142] As an optional embodiment, the grouping module calculates the similarity of the model parameters and the user parameters between any two algorithm model data, and obtains the specific manner of the parameter similarity between the two algorithm model data, including:

[0143] For any two algorithm model data, calculating a first similarity between model parameters corresponding to the two algorithm model data;

[0144] Calculating a second similarity between users corresponding to the two algorithm model data;

[0145] The product of the first similarity and the second similarity is calculated to obtain the parameter similarity between the two algorithm models.

[0146] It can be seen that through the above-mentioned optional embodiments, the parameter similarity between the model data can be calculated by similarity calculation and product calculation of the model parameters of the model data and the user parameters, so as to facilitate the subsequent accurate and reasonable grouping of the model data uploaded by the user, assist in improving the management efficiency and grouping rationality of the model data, provide users with more reasonable model recommendation services, and improve user experience.

[0147] As an optional embodiment, the grouping module groups all algorithm model data according to parameter similarity based on a dynamic programming algorithm to obtain a specific method of multiple model sets, including:

[0148] The objective function is set to maximize the number of algorithm model data in each model set;

[0149] The setting restriction condition includes that the parameter similarity between any two algorithm model data in the same model set is greater than a first similarity threshold, and the parameter similarity between any two algorithm model data belonging to different model sets is less than a second similarity threshold; optionally, the second similarity threshold is less than the first similarity threshold;

[0150] Based on the spatial search algorithm, according to the objective function and constraints, all algorithm model data are iteratively grouped and calculated until the optimal result is obtained, and multiple model sets are obtained.

[0151] It can be seen that through the above optional embodiments, it is possible to accurately and reasonably group the model data uploaded by users based on the preset objective function and constraints and based on the spatial search algorithm, thereby helping to improve the management efficiency and grouping rationality of the model data, providing users with more reasonable model recommendation services, and improving the user experience.

[0152] As an optional embodiment, the specific manner in which the screening module screens out the target model set from the multiple model sets according to the user parameters of the target user includes:

[0153] For each model set, vectorize the union of user parameters corresponding to all algorithm model data in the model set to determine the user parameter vector corresponding to the model set;

[0154] Calculate the vectorized results corresponding to the user parameters of the target user;

[0155] Calculate the vector distance between the user parameter vector and the vectorized result to obtain the user parameter similarity corresponding to the model set;

[0156] All model sets whose user parameter similarity is higher than a third similarity threshold are determined as target model sets.

[0157] It can be seen that through the above optional embodiments, the user parameter similarity corresponding to the model set can be calculated based on the vector distance between the user parameters, and then the target model set can be screened out based on this, so as to subsequently determine more matching recommended model data, thereby improving the management efficiency and grouping rationality of the model data, providing users with more reasonable model recommendation services, and improving user experience.

[0158] As an optional embodiment, the specific manner in which the determination module determines multiple recommendation model data in the target model set according to the website access operation includes:

[0159] Determine the cursor movement path and cursor click object corresponding to the website access operation;

[0160] According to the cursor movement path, multiple path passing areas are determined on the current browser page;

[0161] According to the cursor click object, multiple clicked objects are determined from multiple model objects on the current browser page;

[0162] Determine the model parameters corresponding to all model objects in the area where all paths pass as a first model parameter set;

[0163] Determine the model parameters corresponding to all clicked objects as a second model parameter set;

[0164] For each model data in the target model set, calculating a third similarity between a model parameter of the model data and the first model parameter set;

[0165] Calculate a fourth similarity between the model parameters of the model data and a second model parameter set;

[0166] Calculate the product of the third similarity and the fourth similarity to obtain the priority parameter corresponding to the model data;

[0167] All model data with priority parameters greater than a preset parameter threshold are determined as recommended model data.

[0168] It can be seen that through the above optional embodiments, multiple model objects that the user may be interested in can be determined based on the user's website access operations, and based on the similarity calculation and screening between the model parameters, multiple more matching recommended model data can be accurately determined, thereby improving the management efficiency and grouping rationality of the model data, providing users with more reasonable model recommendation services, and improving user experience.

[0169] Embodiment 3

[0170] See also Figure 3 , Figure 3 It is another model database management system based on parameter matching disclosed in an embodiment of the present invention. Figure 3 The model database management system based on parameter matching is applied to a data processing system / data processing device / data processing server (wherein the server includes a local processing server or a cloud processing server). Figure 3 As shown, the model database management system based on parameter matching may include:

[0171] A memory 301 storing executable program codes;

[0172] a processor 302 coupled to the memory 301;

[0173] The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the model database management method based on parameter matching described in the first embodiment.

[0174] Embodiment 4

[0175] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the model database management method based on parameter matching described in the first embodiment.

[0176] Embodiment 5

[0177] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps of the model database management method based on parameter matching described in the first embodiment.

[0178] The above describes specific embodiments of the present specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily have to be performed in the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0179] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0180] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0181] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may be in the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of this specification may be in the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0182] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0183] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0184] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0185] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0186] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0187] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0188] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0189] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0190] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0191] Finally, it should be noted that the model database management method and system based on parameter matching disclosed in the embodiment of the present invention only discloses the preferred embodiment of the present invention, which is only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A model database management method based on parameter matching, It is characterized in that The method comprises: Obtain multiple algorithm model data uploaded by multiple users; According to the model parameters and user parameters corresponding to each of the algorithm model data, all the algorithm model data are grouped to obtain multiple model sets, including: For any two of the algorithm model data, calculating a first similarity between the model parameters corresponding to the two algorithm model data; Calculating a second similarity between the users corresponding to the two algorithm model data; Calculating the product of the first similarity and the second similarity to obtain the parameter similarity between the two algorithm models; The objective function is set to maximize the number of algorithm model data in each model set; The setting restriction condition includes that the parameter similarity between any two algorithm model data in the same model set is greater than a first similarity threshold, and the parameter similarity between any two algorithm model data belonging to different model sets is less than a second similarity threshold; the second similarity threshold is less than the first similarity threshold; Based on the spatial search algorithm, according to the objective function and the restriction conditions, all the algorithm model data are iteratively grouped and calculated until the optimal result is obtained, thereby obtaining multiple model sets; In response to a website access operation of a target user, a target model set is screened out from the plurality of model sets according to a user parameter of the target user, including: For each of the model sets, vectorize the union of the user parameters corresponding to all the algorithm model data in the model set to determine the user parameter vector corresponding to the model set; Calculate the vectorization result corresponding to the user parameters of the target user; Calculating the vector distance between the user parameter vector and the vectorization result to obtain the user parameter similarity corresponding to the model set; Determine all the model sets whose user parameter similarities are higher than a third similarity threshold as target model sets; According to the website access operation, a plurality of recommended model data are determined in the target model set, including: Determine a cursor movement path and a cursor click object corresponding to the website access operation; According to the cursor movement path, a plurality of path passing areas are determined on the current browser page; According to the cursor click object, determining a plurality of clicked objects from a plurality of model objects of the current browser page; Determine the model parameters corresponding to all model objects in the areas where all the paths pass as a first model parameter set; Determine the model parameters corresponding to all the clicked objects as a second model parameter set; For each model data in the target model set, calculating a third similarity between a model parameter of the model data and the first model parameter set; Calculating a fourth similarity between the model parameters of the model data and the second model parameter set; Calculating the product of the third similarity and the fourth similarity to obtain a priority parameter corresponding to the model data; All the model data whose priority parameters are greater than a preset parameter threshold are determined as recommended model data.

2. The model database management method based on parameter matching according to claim 1, It is characterized in that The model parameters include model size, model purpose, model training data type, model architecture, model input and output data type, and model historical usage records.

3. The model database management method based on parameter matching according to claim 1, It is characterized in that The user parameters include user type, user level, number of models uploaded by the user, user historical upload records and user historical download records.

4. A model database management system based on parameter matching, It is characterized in that The system comprises: An acquisition module is used to acquire multiple algorithm model data uploaded by multiple users; A grouping module is used to group all the algorithm model data according to the model parameters and user parameters corresponding to each of the algorithm model data to obtain multiple model sets, including: For any two of the algorithm model data, calculating a first similarity between the model parameters corresponding to the two algorithm model data; Calculating a second similarity between the users corresponding to the two algorithm model data; Calculating the product of the first similarity and the second similarity to obtain the parameter similarity between the two algorithm models; The objective function is set to maximize the number of algorithm model data in each model set; The setting restriction condition includes that the parameter similarity between any two algorithm model data in the same model set is greater than a first similarity threshold, and the parameter similarity between any two algorithm model data belonging to different model sets is less than a second similarity threshold; the second similarity threshold is less than the first similarity threshold; Based on the spatial search algorithm, according to the objective function and the restriction conditions, all the algorithm model data are iteratively grouped and calculated until the optimal result is obtained, thereby obtaining multiple model sets; A screening module, for responding to a website access operation of a target user and screening a target model set from the plurality of model sets according to a user parameter of the target user, comprises: For each of the model sets, vectorize the union of the user parameters corresponding to all the algorithm model data in the model set to determine the user parameter vector corresponding to the model set; Calculate the vectorization result corresponding to the user parameters of the target user; Calculating the vector distance between the user parameter vector and the vectorization result to obtain the user parameter similarity corresponding to the model set; Determine all the model sets whose user parameter similarities are higher than a third similarity threshold as target model sets; A determination module, used to determine a plurality of recommended model data in the target model set according to the website access operation, including: Determine a cursor movement path and a cursor click object corresponding to the website access operation; According to the cursor movement path, a plurality of path passing areas are determined on the current browser page; According to the cursor click object, determining a plurality of clicked objects from a plurality of model objects of the current browser page; Determine the model parameters corresponding to all model objects in the areas where all the paths pass as a first model parameter set; Determine the model parameters corresponding to all the clicked objects as a second model parameter set; For each model data in the target model set, calculating a third similarity between a model parameter of the model data and the first model parameter set; Calculating a fourth similarity between the model parameters of the model data and the second model parameter set; Calculating the product of the third similarity and the fourth similarity to obtain a priority parameter corresponding to the model data; All the model data whose priority parameters are greater than a preset parameter threshold are determined as recommended model data.

5. A model database management system based on parameter matching, It is characterized in that The system comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the model database management method based on parameter matching as described in any one of claims 1-3.

Images