Function configuration method and device based on gray release, equipment and storage medium

Abstract

The application discloses a function configuration method and device based on gray release, equipment and storage medium, the function configuration method based on gray release includes: obtaining the feature information of the gray version application; based on the feature information, determine the gray user, and gray release the gray version application to the gray user; collect the multi-dimensional index data of the gray user running the function program of the gray version application, and determine the target function configuration of the gray version application based on the index data. The application tests and releases the gray version application to the gray user in the form of gray release, finally collects the multi-dimensional index data of the gray user running the function program of the gray version application, and selects the optimal target function configuration according to the index data of the gray user, so as to realize sufficient test of all functions of the application, and improve the accuracy of the test of the new version application.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a function configuration method, apparatus, device and storage medium based on grayscale release. Background Technology

[0002] Currently, during the iterative development of app features, functional defects in new versions are typically discovered by testers through test cases. However, as the number and complexity of app client features increase, it becomes difficult for testers to fully test all functions within the application.

[0003] In related technologies, application developers typically invite a subset of users to participate in functional testing of a new version of the application. After these users have completed using the new version's features, user feedback is manually collected to ensure thorough testing of the application's functionalities. However, this method involves random user invitations, which can result in some features not being fully covered, leading to incomplete application configuration and low accuracy in testing the new version. Summary of the Invention

[0004] The main purpose of this application is to provide a function configuration method, apparatus, device and storage medium based on gray-scale release, which aims to solve the technical problem of low accuracy in testing new version applications in the prior art.

[0005] To achieve the above objectives, this application provides a feature configuration method based on canary release, the feature configuration method based on canary release including:

[0006] Obtain feature information of the grayscale version application;

[0007] Based on the aforementioned feature information, gray-scale users are identified, and the gray-scale version of the application is then distributed to these gray-scale users in a gray-scale manner.

[0008] Collect multi-dimensional indicator data after the grayscale users run various functions of the grayscale version application, and determine the target function configuration of the grayscale version application based on the indicator data.

[0009] Optionally, the step of determining the target feature configuration of the grayscale version application based on the indicator data includes:

[0010] Based on the aforementioned indicator data, the optimal function configuration is selected through a preset function configuration model to obtain the target function configuration of the grayscale application. The function configuration model is obtained by iteratively training a preset training model based on indicator data samples and the function configuration scheme labels of the indicator data samples.

[0011] Optionally, the step of selecting the optimal function configuration based on the indicator data using a preset function configuration model to obtain the target function configuration of the grayscale application includes:

[0012] Based on the aforementioned indicator data, the function evaluation value of the grayscale version is calculated using a preset function configuration model to obtain the first function evaluation value of the grayscale version.

[0013] Obtain a second functional evaluation value for the target version, wherein the target version includes at least one;

[0014] Based on the first and second function evaluation values, the optimal function configuration of the grayscale version and the target version is selected as the target function configuration of the grayscale version application.

[0015] Optionally, before the step of selecting the optimal function configuration based on the indicator data using a preset function configuration model to obtain the target function configuration of the gray-scale application, the method includes:

[0016] Acquire indicator data samples, the functional configuration scheme tags of the indicator data samples, and the functional adaptability evaluation values;

[0017] Based on the indicator data sample, the functional configuration scheme label of the indicator data sample, and the functional adaptability evaluation value, the preset training model is iteratively trained to obtain a functional configuration model that meets the accuracy requirements.

[0018] Optionally, the step of iteratively training a preset model to be trained based on the indicator data samples, the functional configuration scheme labels of the indicator data samples, and the functional adaptability evaluation values ​​to obtain a functional configuration model that meets the accuracy requirements includes:

[0019] Based on the indicator data samples and the function adaptability evaluation value, a prediction function configuration scheme is obtained by performing prediction processing through a preset training model.

[0020] The difference between the prediction function configuration scheme and the function configuration scheme label of the indicator data sample is calculated to obtain the error result;

[0021] Determine whether the error result meets the error standard indicated by the preset error threshold range;

[0022] If the error result does not meet the error standard indicated by the preset error threshold range, the process returns to the step of performing prediction processing based on the indicator data sample and the function fit evaluation value through the preset model to be trained to obtain the predicted function configuration scheme. Training stops when the error result meets the error standard indicated by the preset error threshold range, and a function configuration model that meets the accuracy conditions is obtained.

[0023] Optionally, before the step of obtaining a predicted function configuration scheme by performing prediction processing through a preset model to be trained based on the indicator data sample and the function fit evaluation value, the method includes:

[0024] The indicator data sample is subjected to random information erasure processing to obtain the erased indicator data sample, wherein the erased indicator data sample includes a portion of the indicator data sample, or the erased indicator data sample includes all the indicator data samples.

[0025] Optionally, after the step of determining the target feature configuration of the grayscale version application based on the indicator data, the method includes:

[0026] Determine whether the target function configuration is consistent with the gray-scale function configuration of the gray-scale version application;

[0027] If the target function configuration is consistent with the grayscale function configuration, then the grayscale function configuration is released in a single release manner;

[0028] If the target function configuration is inconsistent with the grayscale function configuration, the grayscale function configuration will be rolled back using a one-click rollback.

[0029] This application also provides a function configuration device based on grayscale release, the function configuration device based on grayscale release includes:

[0030] The acquisition module is used to obtain feature information of the grayscale version application.

[0031] The publishing module is used to identify gray-scale users based on the feature information and to publish the gray-scale version of the application to the gray-scale users in a gray-scale manner.

[0032] The configuration module is used to collect multi-dimensional indicator data after the grayscale users run various functions of the grayscale version application, and to determine the target function configuration of the grayscale version application based on the indicator data.

[0033] This application also provides a function configuration device based on grayscale release, the function configuration device based on grayscale release comprising: a memory, a processor, and a program stored in the memory for implementing the function configuration method based on grayscale release.

[0034] The memory is used to store programs that implement the function configuration method based on grayscale release;

[0035] The processor is used to execute a program that implements the gray-scale release-based feature configuration method, so as to implement the steps of the gray-scale release-based feature configuration method.

[0036] This application also provides a storage medium storing a program that implements a grayscale release-based function configuration method, wherein the program is executed by a processor to implement the steps of the grayscale release-based function configuration method.

[0037] This application screens gray-scale users and releases the gray-scale version of the application to them for testing through a gray-scale release method. Finally, it collects multi-dimensional indicator data after gray-scale users run various functions of the gray-scale version of the application, and selects the optimal target function configuration based on the gray-scale user indicator data, thereby achieving full testing of all functions of the application and improving the accuracy of testing the new version of the application. Attached Figure Description

[0038] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0039] Figure 1 This is a flowchart illustrating the first embodiment of the feature configuration method based on grayscale release in this application;

[0040] Figure 2 This is a schematic diagram of the functional configuration device based on grayscale release in this application;

[0041] Figure 3 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of this application.

[0042] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0043] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0044] Reference Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the grayscale release-based feature configuration method of this application.

[0045] In the first embodiment, the function configuration method based on canary release includes the following steps:

[0046] Step S100: Obtain the feature information of the grayscale version application;

[0047] It should be noted that the execution subject of the gray-scale release-based function configuration method is the gray-scale release-based function configuration device. Preferably, the gray-scale release-based function configuration device is a software (application) client, but it can also be other terminals with data transmission and data processing functions. No specific restrictions are made here.

[0048] It is understood that a gray-scale version application refers to a new version of the application that is to be released in a gray-scale manner. The characteristic information of the gray-scale version application refers to the characteristics (or characteristic range) of the requirements testing of the gray-scale version application, including but not limited to terminal model, test region, etc. The characteristic information of the gray-scale version application is usually determined by testers based on the testing requirements of the new version of the application.

[0049] In specific implementations, the device can obtain the feature information of the grayscale version application by receiving feature information uploaded / input by testers / or administrators, or by extracting relevant feature information from the database; no specific limitation is made here.

[0050] Step S200: Based on the feature information, determine the grayscale users and release the grayscale version of the application to the grayscale users in a grayscale manner;

[0051] It should be noted that the term "grayscale users" refers to terminal users who meet the conditions for grayscale release (i.e., testing conditions) of the application. The device filters and / or selects eligible terminal users based on the aforementioned characteristic information. After the device identifies the grayscale users, it releases the grayscale version of the application to them. After the grayscale release, the grayscale users can experience (or use) the grayscale version of the application. Specific scenarios include: the application developer further develops an application version X2 based on application version X1. Application version X2 is still in the testing phase. The device filters grayscale users from the user base of application version X1 and sends test invitations to the grayscale users' terminals, thus releasing the grayscale version of the application to them so that they can experience the various untested functions of the grayscale version of the application.

[0052] In a specific implementation, the device determines gray-scale users based on the aforementioned feature information by filtering out eligible terminal users from the user database. The process of establishing the user database involves the device collecting runtime logs generated by users using the original version of the application, cleaning the runtime logs to obtain cleaned runtime logs, and finally establishing a user database involving each user based on the cleaned runtime logs. In other words, this application involves gray-scale release of the application's test version to users, and the user database is established based on the user data of the original version of the application prior to the test version.

[0053] Furthermore, in the specific implementation, the device collects the operation logs generated by the App (the application before the version update), and then the device cleans the operation logs and puts them into the database. The data cleaning includes data desensitization, filtering and format conversion to retain usable, correctly formatted and valid user data. Finally, the device performs data analysis on the cleaned operation logs to establish a user database involving each user.

[0054] Furthermore, in a specific implementation, the device uses the aforementioned feature information to filter gray-scale users from a user database. This feature information typically consists of multi-dimensional features of various functions within a gray-scale application. The feature information is usually filtered based on a preset filtering model to obtain gray-scale users who meet the criteria. It is understood that the device uses the aforementioned feature information to filter target gray-scale users from the user profile library who match the program's feature information. Specifically, the device can filter based on the number of times each user's user profile matches the program's feature information. For example, if the feature information includes region A, terminal brand B, terminal model B1, and user credit score C, then users who meet these criteria are selected from each user's user profile as target gray-scale users.

[0055] Furthermore, in a specific implementation, the device selects target grayscale users from the user profile database based on the program feature information using the following methods:

[0056] The device acquires feature samples and user screening result tags of the feature samples;

[0057] It should be noted that the feature samples and their user selection result labels are used for model training. The user selection result labels are manually labeled labels of the optimal grayscale users corresponding to the feature samples. The manual labeling of the user selection result labels can be done through A / B testing to determine the user selection results with the most successful invitations among users meeting different feature information. These user selection result labels are then used as the user selection result labels for the feature samples. For example, the feature information includes region A, mobile phone brand B, mobile phone model B1, and user credit score C. The user must meet the three conditions of region A, mobile phone brand B, and user credit score C. The success rate of user invitation testing is higher than the success rate of user invitation testing that simultaneously meets all four conditions: region A, mobile phone brand B, mobile phone model B1, and user credit score C. Therefore, the user screening result label is users who meet three of the conditions: region A, mobile phone brand B, and user credit score C. In this process, the weight of different program features can also be determined to further accurately screen users. For example, the weight of region A is 0.8, the weight of mobile phone brand B is 0.7, the weight of mobile phone model B1 is 0.2, and the weight of user credit score C is 0.9. Since the weight of mobile phone model B1 is low, users who meet all three conditions of region A, mobile phone brand B, and user credit score C can be selected as the user screening result label.

[0058] Understandably, the determination of the above feature weights and the success rate of invitation testing can also be obtained through parameter training during the screening model training process.

[0059] In a specific implementation, the device may acquire feature samples and user screening result tags of the feature samples by receiving feature samples and user screening result tags uploaded / input by testers / or users, or by reading relevant feature samples and user screening result tags of the feature samples from a database. No specific limitation is made here.

[0060] Based on the feature samples and the user screening result labels of the feature samples, the preset training model is iteratively trained to obtain a screening model that meets the accuracy requirements.

[0061] Understandably, the device iteratively trains a preset training model based on the feature samples and the user screening result labels of the feature samples to obtain a screening model that meets the accuracy requirements. The training model is a preset model that has the basic ability to process the feature samples and predict the grayscale users to be screened. The only difference between the training model and the screening model is the accuracy.

[0062] Based on the program feature information and the user profile database, users are matched using the filtering model to obtain target grayscale users.

[0063] It should be noted that during the user matching process using the filtering model, the device can identify a user as a target grayscale user as long as the user profile meets the filtering conditions obtained by the filtering model based on the program feature information.

[0064] In a specific implementation, before the step of matching users based on the program feature information and the user profile database using the filtering model to obtain the target grayscale users, the pre-training of the filtering model includes the following methods:

[0065] The device inputs the feature samples into a preset training model to obtain user screening prediction results; calculates the difference between the user screening prediction results and the user screening result labels of the feature samples to obtain an error result; determines whether the error result meets the error standard indicated by a preset error threshold range; if the error result does not meet the error standard indicated by the preset error threshold range, it returns to the step of inputting the feature samples into the preset training model to obtain user screening prediction results, until the error result meets the error standard indicated by the preset error threshold range, and then stops training to obtain a screening model that meets the accuracy requirements.

[0066] Understandably, the device inputs the feature samples into a preset model to be trained to obtain user selection prediction results, which are obtained by predictive analysis of the model during training.

[0067] In a specific implementation, the device calculates the difference between the user screening prediction result and the user screening result label of the feature sample to obtain the error result. That is, it verifies whether the result obtained by the model in training is consistent with the known result and calculates the difference between the results to obtain the error result.

[0068] It should be noted that the device further determines whether the error result meets the error standard indicated by the preset error threshold range. Specifically, since there is an error between the result after model training and the actual result, the error result is allowed to be within the preset error threshold range, thereby further determining whether the error result meets the error standard indicated by the preset error threshold range.

[0069] In the specific implementation, if the error result does not meet the error standard indicated by the preset error threshold range, it indicates that the model has too large an error in this training. The device then returns to the step of inputting the feature sample into the preset model to be trained to obtain the user screening prediction result, realizing iterative training of the model until the error result meets the error standard indicated by the preset error threshold range, and then stops training to obtain a screening model that meets the accuracy conditions, thereby improving the accuracy of the model in screening target grayscale users.

[0070] Step S300: Collect multi-dimensional indicator data after the grayscale user runs various functions of the grayscale version application, and determine the target function configuration of the grayscale version application based on the indicator data.

[0071] In a specific implementation, the method for determining the target feature configuration of the grayscale version application based on the aforementioned indicator data further includes the following steps:

[0072] Based on the aforementioned indicator data, the optimal function configuration is selected through a preset function configuration model to obtain the target function configuration of the grayscale application. The function configuration model is obtained by iteratively training a preset training model based on indicator data samples and the function configuration scheme labels of the indicator data samples.

[0073] In a specific implementation, the method for obtaining the target functional configuration of the grayscale application by selecting the optimal functional configuration based on the indicator data and a preset functional configuration model further includes the following steps:

[0074] Based on the aforementioned indicator data, the function evaluation value of the grayscale version is calculated using a preset function configuration model to obtain the first function evaluation value of the grayscale version; the second function evaluation value of the target version is obtained, wherein the target version includes at least one; based on the first function evaluation value and the second function evaluation value, the function configuration of the optimal version among the grayscale version and the target version is selected as the target function configuration of the grayscale version application.

[0075] In a specific implementation, after step S300, which involves collecting multi-dimensional indicator data after the grayscale user runs various functional programs of the grayscale version application, and determining the target functional configuration of the grayscale version application based on the indicator data, the method includes:

[0076] Determine whether the target function configuration is consistent with the gray-scale function configuration of the gray-scale version application; if the target function configuration is consistent with the gray-scale function configuration, then publish the gray-scale function configuration through a single release; if the target function configuration is inconsistent with the gray-scale function configuration, then roll back the gray-scale function configuration through a one-click rollback.

[0077] This application screens gray-scale users and releases the gray-scale version of the application to them for testing through a gray-scale release method. Finally, it collects multi-dimensional indicator data after gray-scale users run various functions of the gray-scale version of the application, and selects the optimal target function configuration based on the gray-scale user indicator data, thereby achieving full testing of all functions of the application and improving the accuracy of testing the new version of the application.

[0078] Based on the first embodiment described above, this application also provides another embodiment, wherein the function configuration method based on grayscale release includes:

[0079] Before the step of the device selecting the optimal function configuration based on the indicator data and a preset function configuration model to obtain the target function configuration of the grayscale version application, the method further includes the following steps:

[0080] Step A100: Obtain the indicator data sample, the function configuration scheme label of the indicator data sample, and the function adaptability evaluation value;

[0081] Step A200: Based on the indicator data sample, the functional configuration scheme label of the indicator data sample, and the functional adaptability evaluation value, the preset training model is iteratively trained to obtain a functional configuration model that meets the accuracy conditions.

[0082] In a specific implementation, the method for iteratively training a preset training model based on the indicator data samples, the functional configuration scheme labels of the indicator data samples, and the functional adaptability evaluation value to obtain a functional configuration model that meets the accuracy requirements further includes the following steps:

[0083] The device performs prediction processing based on the indicator data samples and the function fit evaluation value, using a preset training model to obtain a predicted function configuration scheme; it calculates the difference between the predicted function configuration scheme and the function configuration scheme label of the indicator data samples to obtain an error result; it determines whether the error result meets the error standard indicated by a preset error threshold range; if the error result does not meet the error standard indicated by the preset error threshold range, it returns to the step of performing prediction processing based on the indicator data samples and the function fit evaluation value, using a preset training model to obtain a predicted function configuration scheme, until the error result meets the error standard indicated by the preset error threshold range, at which point training stops, resulting in a function configuration model that meets the accuracy requirements.

[0084] In its specific implementation, before the step of obtaining the predicted function configuration scheme by performing prediction processing through a preset training model based on the indicator data sample and the function adaptability evaluation value, the method includes the following steps:

[0085] The indicator data sample is subjected to random information erasure processing to obtain the erased indicator data sample, wherein the erased indicator data sample includes a portion of the indicator data sample, or the erased indicator data sample includes all the indicator data samples.

[0086] It should be noted that the random information erasure process proposed in this application may erase some information or not erase any information, but it will not erase all information. Since some information is inevitably missing when collecting samples of the actual infant soothing process (such as missing time information), this application designs a random information erasure module to enable the algorithm model to perform intelligent control even when some information is missing, enhancing the robustness of the algorithm. This ensures that the algorithm model can stably and reliably perform the corresponding information processing work even when some information is unavailable in product applications.

[0087] This application also provides a function configuration device based on grayscale release, as shown in the following reference. Figure 2 The grayscale-based deployment configuration device includes:

[0088] Module 10 is used to acquire feature information of the grayscale version application;

[0089] The publishing module 20 is used to determine grayscale users based on the feature information and to grayscale publish the grayscale version application to the grayscale users.

[0090] The configuration module 30 is used to collect multi-dimensional indicator data after the grayscale user runs various functional programs of the grayscale version application, and to determine the target function configuration of the grayscale version application based on the indicator data.

[0091] Optionally, the configuration module 30 includes:

[0092] The selection module is used to select the optimal function configuration based on the indicator data and through a preset function configuration model to obtain the target function configuration of the grayscale version application. The function configuration model is obtained by iteratively training a preset model to be trained based on the indicator data samples and the function configuration scheme labels of the indicator data samples.

[0093] Optionally, the selection module includes:

[0094] The evaluation module is used to calculate the function evaluation value of the gray version based on the indicator data and through a preset function configuration model, so as to obtain the first function evaluation value of the gray version.

[0095] The functional evaluation value acquisition module is used to acquire a second functional evaluation value of the target version, wherein the target version includes at least one;

[0096] The optimal version selection module is used to select the optimal version of the gray version and the target version based on the first function evaluation value and the second function evaluation value, and use it as the target function configuration of the gray version application.

[0097] Optionally, the grayscale-based deployment configuration device further includes:

[0098] The sample acquisition module is used to acquire indicator data samples, functional configuration scheme tags of the indicator data samples, and functional adaptability evaluation values;

[0099] The training module is used to iteratively train a preset model to be trained based on the indicator data sample, the functional configuration scheme label of the indicator data sample, and the functional adaptability evaluation value, so as to obtain a functional configuration model that meets the accuracy conditions.

[0100] Optionally, the training module includes:

[0101] The prediction module is used to perform prediction processing based on the indicator data samples and the function adaptability evaluation value, through a preset training model, to obtain a prediction function configuration scheme.

[0102] The difference calculation module is used to calculate the difference between the prediction function configuration scheme and the function configuration scheme label of the indicator data sample to obtain the error result.

[0103] The judgment module is used to determine whether the error result meets the error standard indicated by the preset error threshold range;

[0104] The iterative training module is used to return a step of predicting a function configuration scheme based on the indicator data sample and the function fit evaluation value by performing prediction processing through a preset model to be trained if the error result does not meet the error standard indicated by the preset error threshold range. Training stops when the error result meets the error standard indicated by the preset error threshold range, and a function configuration model that meets the accuracy conditions is obtained.

[0105] Optionally, the grayscale-based deployment configuration device further includes:

[0106] The random erasure module is used to perform random information erasure processing on the indicator data sample to obtain the erased indicator data sample, wherein the erased indicator data sample includes a portion of the indicator data sample, or the erased indicator data sample includes all the indicator data samples.

[0107] Optionally, the grayscale-based deployment configuration device further includes:

[0108] A configuration judgment module is used to determine whether the target function configuration is consistent with the gray-scale function configuration of the gray-scale version application;

[0109] The configuration release module is used to release the grayscale function configuration in a single release manner if the target function configuration is consistent with the grayscale function configuration.

[0110] The rollback module is used to roll back the grayscale function configuration by one click if the target function configuration is inconsistent with the grayscale function configuration.

[0111] The specific implementation of the function configuration device based on grayscale release in this application is basically the same as the embodiments of the function configuration method based on grayscale release described above, and will not be repeated here.

[0112] Reference Figure 3 , Figure 3 This is a schematic diagram of the terminal structure of the hardware operating environment involved in the embodiments of this application.

[0113] like Figure 3As shown, the terminal may include: a processor 1001, such as a CPU; a network interface 1004; a user interface 1003; a memory 1005; and a communication bus 1002. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen and an input unit such as a keyboard. Optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0114] Optionally, the grayscale-release-based functional configuration device may also include a rectangular user interface, a network interface, a camera, RF (Radio Frequency) circuitry, sensors, audio circuitry, a WiFi module, etc. The rectangular user interface may include a display screen and an input submodule such as a keyboard. Optionally, the rectangular user interface may also include a standard wired interface or a wireless interface. The network interface may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0115] Those skilled in the art will understand that Figure 2 The grayscale-based functional configuration device structure shown does not constitute a limitation on grayscale-based functional configuration devices, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0116] like Figure 3 As shown, the memory 1005, as a storage medium, may include an operating system, a network communication module, and a gray-scale release-based function configuration program. The operating system is a program that manages and controls the hardware and software resources of the gray-scale release-based function configuration device, supporting the operation of the gray-scale release-based function configuration program and other software and / or programs. The network communication module is used to enable communication between the various components within the memory 1005, as well as communication with other hardware and software in the gray-scale release-based function configuration system.

[0117] exist Figure 3 Figure 3 In the grayscale release-based function configuration device shown, the processor 1001 is used to execute the grayscale release-based function configuration program stored in the memory 1005 to implement the steps of the grayscale release-based function configuration method described above.

[0118] The specific implementation method of the function configuration device based on grayscale release in this application is basically the same as the various embodiments of the function configuration method based on grayscale release described above, and will not be repeated here.

[0119] This application also provides a storage medium storing a program implementing a gray-scale release-based function configuration method, wherein the program implementing the gray-scale release-based function configuration method is executed by a processor to implement the gray-scale release-based function configuration method as follows:

[0120] Obtain feature information of the grayscale version application;

[0121] Based on the aforementioned feature information, gray-scale users are identified, and the gray-scale version of the application is then distributed to these gray-scale users in a gray-scale manner.

[0122] Collect multi-dimensional indicator data after the grayscale users run various functions of the grayscale version application, and determine the target function configuration of the grayscale version application based on the indicator data.

[0123] Optionally, the step of determining the target feature configuration of the grayscale version application based on the indicator data includes:

[0124] Based on the aforementioned indicator data, the optimal function configuration is selected through a preset function configuration model to obtain the target function configuration of the grayscale application. The function configuration model is obtained by iteratively training a preset training model based on indicator data samples and the function configuration scheme labels of the indicator data samples.

[0125] Optionally, the step of selecting the optimal function configuration based on the indicator data using a preset function configuration model to obtain the target function configuration of the grayscale application includes:

[0126] Based on the aforementioned indicator data, the function evaluation value of the grayscale version is calculated using a preset function configuration model to obtain the first function evaluation value of the grayscale version.

[0127] Obtain a second functional evaluation value for the target version, wherein the target version includes at least one;

[0128] Based on the first and second function evaluation values, the optimal function configuration of the grayscale version and the target version is selected as the target function configuration of the grayscale version application.

[0129] Optionally, before the step of selecting the optimal function configuration based on the indicator data using a preset function configuration model to obtain the target function configuration of the gray-scale application, the method includes:

[0130] Acquire indicator data samples, the functional configuration scheme tags of the indicator data samples, and the functional adaptability evaluation values;

[0131] Based on the indicator data sample, the functional configuration scheme label of the indicator data sample, and the functional adaptability evaluation value, the preset training model is iteratively trained to obtain a functional configuration model that meets the accuracy requirements.

[0132] Optionally, the step of iteratively training a preset model to be trained based on the indicator data samples, the functional configuration scheme labels of the indicator data samples, and the functional adaptability evaluation values ​​to obtain a functional configuration model that meets the accuracy requirements includes:

[0133] Based on the indicator data samples and the function adaptability evaluation value, a prediction function configuration scheme is obtained by performing prediction processing through a preset training model.

[0134] The difference between the prediction function configuration scheme and the function configuration scheme label of the indicator data sample is calculated to obtain the error result;

[0135] Determine whether the error result meets the error standard indicated by the preset error threshold range;

[0136] If the error result does not meet the error standard indicated by the preset error threshold range, the process returns to the step of performing prediction processing based on the indicator data sample and the function fit evaluation value through the preset model to be trained to obtain the predicted function configuration scheme. Training stops when the error result meets the error standard indicated by the preset error threshold range, and a function configuration model that meets the accuracy conditions is obtained.

[0137] Optionally, before the step of obtaining a predicted function configuration scheme by performing prediction processing through a preset model to be trained based on the indicator data sample and the function fit evaluation value, the method includes:

[0138] The indicator data sample is subjected to random information erasure processing to obtain the erased indicator data sample, wherein the erased indicator data sample includes a portion of the indicator data sample, or the erased indicator data sample includes all the indicator data samples.

[0139] Optionally, after the step of determining the target feature configuration of the grayscale version application based on the indicator data, the method includes:

[0140] Determine whether the target function configuration is consistent with the gray-scale function configuration of the gray-scale version application;

[0141] If the target function configuration is consistent with the grayscale function configuration, then the grayscale function configuration is released in a single release manner;

[0142] If the target function configuration is inconsistent with the grayscale function configuration, the grayscale function configuration will be rolled back using a one-click rollback.

[0143] The specific implementation of the storage medium in this application is basically the same as the embodiments of the above-described grayscale release-based functional configuration method, and will not be repeated here.

[0144] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described grayscale release-based functional configuration method.

[0145] The specific implementation methods of the computer program product in this application are basically the same as those of the above-described embodiments of the function configuration method based on grayscale release, and will not be repeated here.

[0146] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0147] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0148] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0149] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A feature configuration method based on canary release, characterized in that, The feature configuration method based on canary release includes: Obtain feature information of the grayscale version application; Based on the aforementioned feature information, gray-scale users are identified, and the gray-scale version of the application is then distributed to these gray-scale users in a gray-scale manner. Collect multi-dimensional indicator data after the grayscale users run the various functions of the grayscale version application. Based on the indicator data, calculate the function evaluation value of the grayscale version through a preset function configuration model to obtain the first function evaluation value of the grayscale version. Obtain the second feature evaluation value for the target version; Based on the first function evaluation value and the second function evaluation value, the function configuration of the optimal version between the gray version and the target version is selected as the target function configuration of the gray version application. Determine whether the target function configuration is consistent with the gray-scale function configuration of the gray-scale version application; if consistent, release it through a single release; if inconsistent, roll back it through a one-click rollback. The functional configuration model is obtained by iteratively training a preset training model based on indicator data samples, functional configuration scheme labels of the indicator data samples, and functional adaptability evaluation values.

2. The function configuration method based on canary release as described in claim 1, characterized in that, The step of determining the target feature configuration of the grayscale version application based on the indicator data includes: Based on the aforementioned indicator data, the optimal function configuration is selected through a preset function configuration model to obtain the target function configuration of the grayscale application. The function configuration model is obtained by iteratively training a preset training model based on indicator data samples and the function configuration scheme labels of the indicator data samples.

3. The function configuration method based on canary release as described in claim 2, characterized in that, The step of selecting the optimal function configuration based on the indicator data and a preset function configuration model to obtain the target function configuration of the gray-scale application includes: The target version includes at least one.

4. The function configuration method based on canary release as described in claim 2, characterized in that, Before the step of selecting the optimal function configuration based on the indicator data and a preset function configuration model to obtain the target function configuration of the gray-scale application, the method includes: Acquire indicator data samples, the functional configuration scheme tags of the indicator data samples, and the functional adaptability evaluation values; Based on the indicator data sample, the functional configuration scheme label of the indicator data sample, and the functional adaptability evaluation value, the preset training model is iteratively trained to obtain a functional configuration model that meets the accuracy requirements.

5. The function configuration method based on canary release as described in claim 4, characterized in that, The step of iteratively training a preset model to be trained based on the indicator data samples, the functional configuration scheme labels of the indicator data samples, and the functional adaptability evaluation values ​​to obtain a functional configuration model that meets the accuracy requirements includes: Based on the indicator data samples and the function adaptability evaluation value, a prediction function configuration scheme is obtained by performing prediction processing through a preset training model. The difference between the prediction function configuration scheme and the function configuration scheme label of the indicator data sample is calculated to obtain the error result; Determine whether the error result meets the error standard indicated by the preset error threshold range; If the error result does not meet the error standard indicated by the preset error threshold range, the process returns to the step of performing prediction processing based on the indicator data sample and the function fit evaluation value through the preset model to be trained to obtain the predicted function configuration scheme. Training stops when the error result meets the error standard indicated by the preset error threshold range, and a function configuration model that meets the accuracy conditions is obtained.

6. The function configuration method based on canary release as described in claim 5, characterized in that, Before the step of obtaining a predicted function configuration scheme by performing prediction processing through a preset training model based on the indicator data samples and the function fit evaluation value, the method includes: The indicator data sample is subjected to random information erasure processing to obtain the erased indicator data sample, wherein the erased indicator data sample includes a portion of the indicator data sample, or the erased indicator data sample includes all the indicator data samples.

7. A function configuration device based on grayscale release, characterized in that, The grayscale release-based function configuration device includes: The acquisition module is used to obtain feature information of the grayscale version application. The publishing module is used to identify gray-scale users based on the feature information and to publish the gray-scale version of the application to the gray-scale users in a gray-scale manner. The configuration module is used to collect multi-dimensional indicator data after the grayscale user runs the various functions of the grayscale version application. Based on the indicator data, the function evaluation value of the grayscale version is calculated through a preset function configuration model to obtain the first function evaluation value of the grayscale version. Obtain the second feature evaluation value for the target version; Based on the first function evaluation value and the second function evaluation value, the function configuration of the optimal version between the gray version and the target version is selected as the target function configuration of the gray version application. Determine whether the target function configuration is consistent with the gray-scale function configuration of the gray-scale version application; if consistent, release it through a single release; if inconsistent, roll back it through a one-click rollback. The functional configuration model is obtained by iteratively training a preset training model based on indicator data samples, functional configuration scheme labels of the indicator data samples, and functional adaptability evaluation values.

8. A function configuration device based on grayscale release, characterized in that, The grayscale-based function configuration device includes: a memory, a processor, and a program stored in the memory for implementing the grayscale-based function configuration method. The memory is used to store programs that implement the function configuration method based on grayscale release; The processor is configured to execute a program that implements the grayscale release-based feature configuration method, in order to implement the steps of the grayscale release-based feature configuration method as described in any one of claims 1 to 6.

9. A storage medium, characterized in that, The storage medium stores a program that implements a grayscale release-based function configuration method, which is executed by a processor to implement the steps of the grayscale release-based function configuration method as described in any one of claims 1 to 6.

Images