Performance parameter determination method and device, electronic equipment and readable storage medium

Abstract

The invention discloses a performance parameter determination method and device, electronic equipment and a readable storage medium. The method comprises the following steps: acquiring a performance parameter node group; performing global evaluation on the performance parameter node group to obtain a global evaluation result; performing local evaluation on each performance parameter node in the performance parameter node group to obtain a local evaluation result; combining a comprehensive evaluation result of each performance parameter node of the global evaluation result and the local evaluation result; newly adding performance parameter nodes into a performance parameter node group according to a comprehensive evaluation result; in response to the fact that the number of the performance parameter nodes in the performance parameter node group is larger than a preset value, reconstructing all the performance parameter nodes in the parameter space and corresponding second parameter information by using the first parameter information of all the performance parameter nodes in the performance parameter node group; and determining the performance parameter nodes meeting the requirements according to the second parameter information. Through the mode, the determination time of the performance parameters is shortened, and the working efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of performance parameter processing, and in particular to a performance parameter determination method, device, electronic device and readable storage medium. Background Art

[0002] With the widespread popularization and application of electronic devices, electronic devices can support more and more applications and their functions are becoming more and more powerful. Electronic devices are developing in the direction of diversification and personalization, becoming indispensable electronic products in users' lives.

[0003] Therefore, the performance of electronic equipment will also have different performance parameters. How to obtain high-quality performance parameters needs to be solved urgently. Summary of the invention

[0004] The present application provides a performance parameter determination method, device, electronic device and readable storage medium, which can reduce the performance parameter determination time and improve work efficiency.

[0005] In a first aspect, the present application provides a performance parameter determination method, the method comprising: obtaining a performance parameter node group; the performance parameter node group comprising at least two performance parameter nodes, each performance parameter node comprising a number of first parameter information; performing a global evaluation on the performance parameter node group to obtain a global evaluation result; and performing a local evaluation on each performance parameter node in the performance parameter node group to obtain a local evaluation result; combining the global evaluation result and the local evaluation result to obtain a comprehensive evaluation result of each performance parameter node; adding new performance parameter nodes to the performance parameter node group based on the comprehensive evaluation result; in response to the number of performance parameter nodes in the performance parameter node group being greater than a preset value, reconstructing all performance parameter nodes in the parameter space and corresponding second parameter information using the first parameter information of all performance parameter nodes in the performance parameter node group; and determining the performance parameter nodes that meet the requirements based on the second parameter information.

[0006] In a second aspect, the present application provides a performance parameter determination device, which includes: an acquisition module, used to acquire a performance parameter node group; the performance parameter node group includes at least two performance parameter nodes, and each performance parameter node includes a number of first parameter information; an evaluation module, used to perform a global evaluation on the performance parameter node group to obtain a global evaluation result; and perform a local evaluation on each performance parameter node in the performance parameter node group to obtain a local evaluation result; and a comprehensive evaluation result of each performance parameter node combining the global evaluation result and the local evaluation result; a first processing module, used to add a new performance parameter node to the performance parameter node group according to the comprehensive evaluation result; a second processing module, used to respond to the number of performance parameter nodes in the performance parameter node group being greater than a preset value, and reconstruct all performance parameter nodes in the parameter space and the corresponding second parameter information using the first parameter information of all performance parameter nodes in the performance parameter node group; a third processing module, used to determine the performance parameter nodes that meet the requirements according to the second parameter information.

[0007] In a third aspect, the present application provides an electronic device, comprising a processor and a memory coupled to the processor; wherein the memory is used to store a computer program, and the processor is used to execute the computer program to implement the method provided in the first aspect above.

[0008] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method provided in the first aspect above.

[0009] The beneficial effects of the present application are as follows: Different from the prior art, the performance parameter determination method, device, electronic device and readable storage medium provided by the present application, in the process of determining the performance parameters, use a combination of global evaluation and local evaluation to conduct a comprehensive evaluation on a small number of performance parameter nodes, improve the evaluation accuracy, and expand the evaluation scope by gradually increasing the performance parameter nodes, and then when the number of performance parameter nodes in the performance parameter node group is greater than the preset value, use the first parameter information of all performance parameter nodes in the performance parameter node group to reconstruct all performance parameter nodes in the parameter space and the corresponding second parameter information, and determine the performance parameter nodes that meet the requirements based on the second parameter information, thereby reducing the number of evaluation iterations, reducing the performance parameter determination time, and improving work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] in:

[0012] Figure 1 It is a flow chart of an embodiment of a method for determining performance parameters provided by the present application;

[0013] Figure 2 is a flow chart of another embodiment of the performance parameter determination method provided by the present application;

[0014] Figure 3 is a flow chart of another embodiment of the performance parameter determination method provided by the present application;

[0015] Figure 4 It is a flowchart of an embodiment of step 303 provided by this application;

[0016] Figure 5 It is a flowchart of an embodiment of step 305 provided by the present application;

[0017] Figure 6 is a flow chart of another embodiment of the performance parameter determination method provided by the present application;

[0018] Figure 7 It is a structural diagram of an embodiment of a performance parameter determination device provided by the present application;

[0019] Figure 8 It is a structural schematic diagram of an embodiment of an electronic device provided by the present application;

[0020] Fig. 9 is a structural schematic diagram of another embodiment of the electronic device provided by the present application;

[0021] Fig.10 It is a structural schematic diagram of an embodiment of a computer-readable storage medium provided by the present application. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It will be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some but not all structures related to the present application are shown in the drawings. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0023] 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 application. The appearance of the phrase in various locations 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.

[0024] With the widespread popularization and application of electronic devices, electronic devices can support more and more applications and their functions are becoming more and more powerful. Electronic devices are developing in the direction of diversification and personalization, becoming indispensable electronic products in users' lives.

[0025] Therefore, the performance of electronic equipment will also have different performance parameters. How to obtain high-quality performance parameters needs to be solved urgently.

[0026] The current performance parameter optimization method is very time-consuming and laborious to test the performance of each of the too many performance parameters.

[0027] Based on this, the present application proposes any of the following embodiments to solve at least one of the above technical problems.

[0028] See also Figure 1 , Figure 1 1 is a flow chart of an embodiment of a method for determining performance parameters provided by the present application. The method comprises:

[0029] Step 11: Get the performance parameter node group.

[0030] The performance parameter node group includes at least two performance parameter nodes, and each performance parameter node includes some first parameter information.

[0031] Among them, the first parameter information may include parameters of the central processing unit and corresponding parameter values, parameters of the graphics processing unit and corresponding parameter values, frame rate of the display screen and other parameters that affect the performance of the electronic device. That is, a performance parameter node includes multiple different parameters and parameter values ​​corresponding to these parameters. For example, performance parameter node A includes parameter a, parameter b, parameter c; wherein parameter a=a1, parameter b=b1, parameter c=c1. Performance parameter node B includes parameter a, parameter b, parameter c; wherein parameter a=a2, parameter b=b2, parameter c=c2. That is, at least one parameter in each performance parameter node is different or the parameter value corresponding to the parameter is different.

[0032] In some embodiments, the performance parameter node group may be manually input in advance. It is understood that the number of performance parameter nodes in the performance parameter node group here is much smaller than the number of performance parameter nodes composed of all performance parameters. For example, the number of performance parameter nodes composed of all performance parameters may be counted in thousands or tens of thousands, while the number of performance parameter nodes in the performance parameter node group may be counted in units or tens.

[0033] It can be understood that with subsequent calculations, the number of performance parameter nodes in the performance parameter node group will gradually increase until it meets the requirements and no longer increases.

[0034] For example, some performance parameters are as follows: the minimum value of the maximum frequency of the CPU large core (cl: 1-19), the maximum value of the maximum frequency of the CPU large core (ch: 1-19), the maximum frequency of the CPU small core (sm: 1-19), and the maximum frequency of the super core (gm: 1-19); the above are only 4 performance parameters. If only these 4 performance parameters are adjusted, the example is as follows (cl: 4, ch: 9, sm: 9, gm: 17), then (cl: 4, ch: 9, sm: 9, gm: 17) can be used as a performance parameter node.

[0035] Step 12: Perform a global evaluation on the performance parameter node group to obtain a global evaluation result.

[0036] In some embodiments, a global evaluation function is used to perform a global evaluation on all performance parameter nodes in the performance parameter node group to obtain a global evaluation result.

[0037] Step 13: Perform local evaluation on each performance parameter node in the performance parameter node group to obtain a local evaluation result.

[0038] In some embodiments, a local evaluation function is used to perform a global evaluation on all performance parameter nodes in the performance parameter node group to obtain a global evaluation result. For example, the local evaluation function may be a local linear approximation function.

[0039] Step 14: Combine the global evaluation results and the local evaluation results to obtain the comprehensive evaluation results of each performance parameter node.

[0040] In some embodiments, the global evaluation result and the local evaluation result are both numerically represented, and a weighted sum of the global evaluation result and the local evaluation result can be performed to obtain a comprehensive evaluation result of each performance parameter node.

[0041] Step 15: Add new performance parameter nodes to the performance parameter node group based on the comprehensive evaluation results.

[0042] Since each performance parameter node has a corresponding comprehensive evaluation result, a target performance parameter node can be determined from the comprehensive evaluation result. A node is determined near the target performance parameter node and added to the performance parameter node group as a newly added performance parameter node.

[0043] The larger the value corresponding to the comprehensive evaluation result is, the less information the corresponding performance parameter node has. Therefore, a node is determined from the neighborhood of the performance parameter node and added to the performance parameter node group as a new performance parameter node, so that the new performance parameter node can be supplemented with information in the subsequent process.

[0044] Step 16: In response to the number of performance parameter nodes in the performance parameter node group being greater than a preset value, all performance parameter nodes and corresponding second parameter information in the parameter space are reconstructed using the first parameter information of all performance parameter nodes in the performance parameter node group.

[0045] In some embodiments, in response to the number of performance parameter nodes in the performance parameter node group being less than or equal to a preset value, the process of steps 11 to 15 is executed again using the performance parameter nodes in the performance parameter node group to increase the performance parameter nodes in the performance parameter node group.

[0046] In some embodiments, the number of performance parameter nodes in the performance parameter node group is much smaller than the number of performance parameter nodes in the parameter space composed of all performance parameters. Therefore, when the number of performance parameter nodes in the performance parameter node group is greater than a preset value, all performance parameter nodes and corresponding second parameter information in the parameter space are reconstructed using the first parameter information of all performance parameter nodes in the performance parameter node group. In this way, performance parameter nodes that can be formed by all performance parameters can be obtained.

[0047] In some embodiments, the first parameter information includes performance parameters and corresponding performance parameter values, and also includes experience quality and current values ​​when running the corresponding performance parameter nodes.

[0048] Based on this, during reconstruction, the second parameter information also includes the performance parameter and the corresponding performance parameter value, and also includes the experience quality and current value when running the corresponding performance parameter node.

[0049] Step 17: Determine a performance parameter node that meets the requirements according to the second parameter information.

[0050] In some embodiments, the required judgment can be made based on the experience quality and the current value in the second parameter information, and then the performance parameter node that meets the requirements can be determined.

[0051] In some embodiments, the ratio between the experience quality and the current value can be used as a judgment reference. If the ratio is greater than a first threshold, it means that the experience quality is high and the performance parameter node meets the requirements.

[0052] In some embodiments, the ratio between the current value and the experience quality can be used as a judgment reference. If the ratio is less than the second threshold, it means that the experience quality is high and the performance parameter node meets the requirements.

[0053] In the process of determining the performance parameters of this embodiment, a comprehensive evaluation is performed on a small number of performance parameter nodes by combining global evaluation and local evaluation to improve the evaluation accuracy, and the evaluation scope is expanded by gradually increasing the performance parameter nodes. Then, when the number of performance parameter nodes in the performance parameter node group is greater than the preset value, all performance parameter nodes in the parameter space and the corresponding second parameter information are reconstructed using the first parameter information of all performance parameter nodes in the performance parameter node group, and the performance parameter nodes that meet the requirements are determined based on the second parameter information, thereby reducing the number of evaluation iterations, reducing the determination time of the performance parameter test nodes, and improving work efficiency.

[0054] See also Figure 2 , Figure 2 : is a flow chart of another embodiment of the performance parameter determination method provided by the present application. The method comprises:

[0055] Step 21: Get the performance parameter node group.

[0056] The performance parameter node group includes at least two performance parameter nodes, and each performance parameter node includes some first parameter information.

[0057] Step 21 has the same or similar technical solution as other embodiments of the present application and will not be described in detail here.

[0058] Step 22: Generate several test points.

[0059] In some embodiments, several test points may be randomly generated, such as based on a Monte Carlo method.

[0060] Step 23: Determine the distance from each test point to each performance parameter node, and count the target test points whose distances meet preset requirements.

[0061] Since the test points and the performance parameter nodes have corresponding coordinates in the parameter space, the distance from each test point to each performance parameter node can be calculated through their respective coordinates.

[0062] According to these distances, the target test points that meet the preset requirements are determined. Since the test points are randomly generated, the corresponding distances are close or far. Therefore, the test points with distances less than the preset distance are selected as the target test points.

[0063] Step 24: Use the target test point to determine the Voronoi unit corresponding to each performance parameter node, and use the Voronoi unit as the global evaluation result.

[0064] Then, the Voronoi unit corresponding to each performance parameter node is determined using these target test points. It can be understood that the Voronoi unit is an area, and the performance parameter node corresponds to the Voronoi unit. Therefore, the two-dimensional area or three-dimensional volume of the Voronoi unit can be used as the global evaluation result.

[0065] Step 25: Perform local evaluation on each performance parameter node in the performance parameter node group to obtain a local evaluation result.

[0066] Step 26: Combine the global evaluation result and the local evaluation result to obtain the comprehensive evaluation result of each performance parameter node.

[0067] Step 27: Add new performance parameter nodes to the performance parameter node group based on the comprehensive evaluation results.

[0068] Step 28: In response to the number of performance parameter nodes in the performance parameter node group being greater than a preset value, all performance parameter nodes and corresponding second parameter information in the parameter space are reconstructed using the first parameter information of all performance parameter nodes in the performance parameter node group.

[0069] Step 29: Determine a performance parameter node that meets the requirements according to the second parameter information.

[0070] Steps 25 to 29 have the same or similar technical solutions as the other embodiments of the present application and will not be described in detail here.

[0071] In the process of determining the performance parameters of this embodiment, a comprehensive evaluation is performed on a small number of performance parameter nodes by combining global evaluation and local evaluation to improve the evaluation accuracy, and the evaluation scope is expanded by gradually increasing the performance parameter nodes. Then, when the number of performance parameter nodes in the performance parameter node group is greater than the preset value, all performance parameter nodes in the parameter space and the corresponding second parameter information are reconstructed using the first parameter information of all performance parameter nodes in the performance parameter node group, and the performance parameter nodes that meet the requirements are determined based on the second parameter information, thereby reducing the number of evaluation iterations, reducing the determination time of the performance parameter test nodes, and improving work efficiency.

[0072] See also Figure 3 , Figure 3 : is a flow chart of another embodiment of the performance parameter determination method provided by the present application. The method comprises:

[0073] Step 301: Obtain a performance parameter node group.

[0074] The performance parameter node group includes at least two performance parameter nodes, and each performance parameter node includes some first parameter information.

[0075] Step 302: Perform a global evaluation on the performance parameter node group to obtain a global evaluation result.

[0076] Steps 301 to 302 have the same or similar technical solutions as the other embodiments of the present application, and will not be described in detail here.

[0077] Step 303: Determine the target neighboring node of each performance parameter node in the parameter space.

[0078] In some embodiments, the neighboring node may be located above, below, to the left, or to the right of the performance parameter node.

[0079] In some embodiments, if only distance is used as a metric and only the nearest neighboring nodes to the performance parameter node are selected, other areas will inevitably be ignored; but if the neighboring nodes are selected as dispersedly as possible to ensure full coverage of the area, although every direction is covered, the distance to the performance parameter node is too far and the meaning of the "neighborhood" is lost. Therefore, in order to make the best selection of the neighboring performance parameter nodes, refer to Figure 4 , step 303 may be the following process:

[0080] Step 41: Determine a first average distance between all to-be-determined neighboring nodes and each performance parameter node.

[0081] Specifically, it can be defined by the following formula:

[0082]

[0083] Among them, D(P mi ) represents the first average distance between all neighboring nodes to be determined and the mth performance parameter node, P m represents the mth performance parameter node, P mi represents the i-th neighboring node to be determined corresponding to the m-th performance parameter node, and 2d represents the total number of neighboring nodes to be determined.

[0084] Step 42: Determine the minimum average distance between all neighboring nodes to be determined.

[0085] Specifically, it can be defined by the following formula:

[0086]

[0087] D'(P mi ) represents the minimum average distance between all neighboring nodes to be determined, P mj Indicates the jth neighboring node to be determined corresponding to the mth performance parameter node.

[0088] Step 43: Screening is performed according to the minimum average distance and the first average distance to obtain the target neighborhood node of each performance parameter node in the parameter space.

[0089] Specifically, it can be defined by the following formula:

[0090]

[0091] R(P mi ) represents the screening criteria, with R(P mi ) is maximized and the target neighborhood nodes are selected. Suitable neighborhood nodes need to be as close to the performance parameter nodes as possible and as far away from each other as possible, that is, try to ensure that there are neighborhood nodes on both sides of the performance parameter nodes in any dimension.

[0092] Step 304: Fit each performance parameter node and the corresponding target neighborhood node to form a hyperplane.

[0093] After the target neighborhood node is determined, each performance parameter node and the corresponding target neighborhood node can be fitted to form a hyperplane. In order to ensure that the hyperplane passes through the performance parameter node, the least square method can be used to solve it.

[0094] Step 305: Obtain a local linear approximation based on the hyperplane.

[0095] In the least squares method, we need to know the corresponding gradient information, see Figure 5 , step 305 may be the following process:

[0096] Step 51: Obtain the gradient information of each performance parameter node in different dimensions based on the hyperplane.

[0097] Among them, the least squares method can be used to solve:

[0098]

[0099] in, It is the gradient matrix composed of the gradients to be solved in each dimension.

[0100] Because f(P mi ) represents the actual response value corresponding to the i-th neighborhood node, which can be calculated by the corresponding model. So the gradient matrix can be obtained.

[0101] Step 52: Obtain a local linear approximation according to the gradient information.

[0102] The first data may be obtained by using the gradient information, the performance parameter node and the target neighborhood node, and then combined with the actual response value corresponding to the performance parameter node to obtain the local linear approximation value.

[0103] That is, for a given performance parameter node, the local linear approximation at a nearby target neighborhood node can be expressed by the following formula:

[0104] F(P mi )=f(P m )+g·(P mi -P m ).

[0105] F(P mi ) represents the local linear approximation of the i-th target neighborhood node of the m-th performance parameter node, f(P m ) indicates the actual response value corresponding to the mth performance parameter node.

[0106] Step 306: Obtain a local evaluation result based on the actual response value corresponding to each target neighborhood node and the local linear approximation value.

[0107] Therefore, the local evaluation result obtained by combining the actual response value and the local linear approximation corresponding to each target neighborhood node can be used to characterize the local evaluation result of the corresponding performance parameter node.

[0108] Step 307: Combine the global evaluation result and the local evaluation result to obtain a comprehensive evaluation result of each performance parameter node.

[0109] Step 308: Add new performance parameter nodes to the performance parameter node group based on the comprehensive evaluation results.

[0110] Step 309: In response to the number of performance parameter nodes in the performance parameter node group being greater than a preset value, all performance parameter nodes in the performance parameter node group are reconstructed using the first parameter information of all performance parameter nodes in the performance parameter node group and corresponding second parameter information in the parameter space.

[0111] Step 310: Determine a performance parameter node that meets the requirements according to the second parameter information.

[0112] Steps 307 to 310 have the same or similar technical solutions as the other embodiments of the present application and are not described in detail here.

[0113] In the process of determining the performance parameters of this embodiment, a comprehensive evaluation is performed on a small number of performance parameter nodes by combining global evaluation and local evaluation to improve the evaluation accuracy, and the evaluation scope is expanded by gradually increasing the performance parameter nodes. Then, when the number of performance parameter nodes in the performance parameter node group is greater than the preset value, all performance parameter nodes in the parameter space and the corresponding second parameter information are reconstructed using the first parameter information of all performance parameter nodes in the performance parameter node group, and the performance parameter nodes that meet the requirements are determined based on the second parameter information, thereby reducing the number of evaluation iterations, reducing the determination time of the performance parameter test nodes, and improving work efficiency.

[0114] See also Figure 6 , Figure 6 : is a flow chart of another embodiment of the performance parameter determination method provided by the present application. The method comprises:

[0115] Step 61: Obtain a performance parameter node group.

[0116] The performance parameter node group includes at least two performance parameter nodes, and each performance parameter node includes some first parameter information.

[0117] Step 62: Perform a global evaluation on the performance parameter node group to obtain a global evaluation result.

[0118] Step 63: Perform local evaluation on each performance parameter node in the performance parameter node group to obtain a local evaluation result.

[0119] Step 64: Combine the global evaluation result and the local evaluation result to obtain a comprehensive evaluation result of each performance parameter node.

[0120] Steps 61 to 64 have the same or similar technical solutions as the other embodiments of the present application and will not be described in detail here.

[0121] Step 65: Determine the target performance parameter node corresponding to the maximum value of the comprehensive evaluation result.

[0122] It can be understood that in this embodiment, the global evaluation result and the local evaluation result are both quantified in numerical values, so the comprehensive evaluation result is also a numerical value. Therefore, the current performance parameter nodes in the performance parameter node group can be sorted according to the comprehensive evaluation result.

[0123] The larger the value corresponding to the comprehensive evaluation result is, the less information the corresponding performance parameter node has. Therefore, a node is determined from the neighborhood of the performance parameter node and added to the performance parameter node group as a new performance parameter node, so that the new performance parameter node can be supplemented with information in the subsequent process.

[0124] Therefore, in this embodiment, the target performance parameter node corresponding to the maximum value of the comprehensive evaluation result is selected.

[0125] Step 66: Determine a node in the neighborhood of the target performance parameter node and add it to the performance parameter node group as a new performance parameter node.

[0126] The node may be selected from a plurality of target neighboring nodes of the target performance parameter node.

[0127] Step 67: In response to the number of performance parameter nodes in the performance parameter node group being greater than a preset value, first parameter information of all performance parameter nodes in the performance parameter node group is input into the reconstruction model so that the reconstruction model outputs all performance parameter nodes in the parameter space and corresponding second parameter information.

[0128] Among them, the reconstruction model is based on Kriging interpolation, compressed sensing reconstruction or neural network construction.

[0129] Step 68: Determine a performance parameter node that meets the requirements according to the second parameter information.

[0130] Step 68 has the same or similar technical solution as the other embodiments of the present application and will not be described in detail here.

[0131] In an application scenario, the performance parameter nodes in the performance parameter node group may be evaluated using LOLA (Local Linear Approximation) to obtain local evaluation results, and may be evaluated using Voronoi global evaluation to obtain global evaluation results.

[0132] Among them, Voronoi evaluation calculates the size of Thiessen polygons based on Monte Carlo simulation, which is used to evaluate the distribution of nodes in the entire parameter space. Its purpose is to fill the space as evenly as possible.

[0133] Among them, LOLA local linear approximation evaluation is used to evaluate the intensity of nonlinearity near existing performance parameter nodes. The purpose is to find the key areas in the parameter space and densely sample in the key areas. If the model output response is linear in the area, then only fewer nodes are needed to reconstruct and restore the data in the area. For areas with higher nonlinearity, more nodes are required.

[0134] In the actual local linear evaluation, the gradient of the reference point is estimated by screening the existing data points in the neighborhood of the performance parameter node (considering adhesion and cohesion) and the model response.

[0135] Then the two evaluation methods are considered comprehensively to offset their respective shortcomings and obtain a comprehensive evaluation result.

[0136] Furthermore, when the number of performance parameter nodes in the performance parameter node group exceeds the preset number, the Kriging interpolation method is used to reconstruct the performance parameter nodes in the parameter space. The Kriging interpolation method is an interpolation method based on statistics, also known as the spatial local interpolation method. The principle is to use regionalized variables as the basis and the variogram as the basic tool to perform linear unbiased and optimal estimation of unknown sample points. It mainly includes four parts: calculating the sample variogram, modeling the estimated data according to the variogram, and using the constructed model to perform Kriging interpolation estimation and variance estimation. Specifically, the performance parameter nodes screened by adaptive sampling and the corresponding AGTS test QoE (quality of experience) / current values ​​are used as input, and the Kriging interpolation method is used for interpolation to reconstruct all performance parameter nodes and the corresponding QoE / current values ​​in the parameter space. The optimal performance parameters can be selected from the reconstructed data later. Among them, the larger the ratio of QoE / current value, the better the performance of the corresponding performance parameter node.

[0137] The specific steps for the above application scenario are as follows:

[0138] (1) Initialize the GPA (General Performance Adjustor) parameters that need to be debugged (such as the performance parameters of this application), set the adjustment range, and the number of sampling nodes (performance parameter nodes).

[0139] (2) Select the initial node. The initial points in two-dimensional space are generally two points on the diagonal of the space, and similarly in higher dimensions.

[0140] (3) Based on the existing nodes, the current value and quality of experience (QoE) when the node is running under the parameters of the node are obtained through AGTS (Auto Game Tool Serials, game automated testing tool), and the ratio is calculated. For example, the parameters of the node are run in the game scenario to obtain the corresponding current value and quality of experience.

[0141] (4) Through the LOLA-Voronoi hybrid strategy, global and local evaluations are performed on each node to calculate the comprehensive evaluation result S.

[0142] (5) Sort the nodes according to the mixed score S corresponding to each node, add a new node near the node with the largest S value, add the node to the node set (such as the performance parameter node group mentioned above), so as to achieve the update and repeat step (3).

[0143] (6) Repeat steps 3-5 until the number of nodes in the node set meets the requirements.

[0144] (7) Using the nodes and corresponding QoE / current stored in the node set at this time as basic data, the Kriging method is used to reconstruct the model corresponding values ​​of all nodes in the parameter space (such as the parameters and values ​​of each node, and the corresponding QoE / current).

[0145] (8) Select the optimal parameters from all nodes.

[0146] In the process of determining the performance parameters of this embodiment, a comprehensive evaluation is performed on a small number of performance parameter nodes by combining global evaluation and local evaluation to improve the evaluation accuracy, and the evaluation scope is expanded by gradually increasing the performance parameter nodes. Then, when the number of performance parameter nodes in the performance parameter node group is greater than the preset value, all performance parameter nodes in the parameter space and the corresponding second parameter information are reconstructed using the first parameter information of all performance parameter nodes in the performance parameter node group, and the performance parameter nodes that meet the requirements are determined based on the second parameter information, thereby reducing the number of evaluation iterations, reducing the determination time of the performance parameter test nodes, and improving work efficiency.

[0147] See also Figure 7 , Figure 7 1 is a schematic diagram of the structure of an embodiment of a performance parameter determination device provided in the present application. The performance parameter determination device 70 includes an acquisition module 71 , an evaluation module 72 , a first processing module 73 , a second processing module 74 and a third processing module 75 .

[0148] The acquisition module 71 is used to acquire a performance parameter node group; the performance parameter node group includes at least two performance parameter nodes, and each performance parameter node includes a number of first parameter information.

[0149] The evaluation module 72 is used to perform a global evaluation on the performance parameter node group to obtain a global evaluation result; and to perform a local evaluation on each performance parameter node in the performance parameter node group to obtain a local evaluation result; and a comprehensive evaluation result of each performance parameter node combining the global evaluation result and the local evaluation result.

[0150] The first processing module 73 is used to add a new performance parameter node to the performance parameter node group according to the comprehensive evaluation result.

[0151] The second processing module 74 is used for reconstructing all performance parameter nodes and corresponding second parameter information in the parameter space using the first parameter information of all performance parameter nodes in the performance parameter node group in response to the number of performance parameter nodes in the performance parameter node group being greater than a preset value.

[0152] The third processing module 75 is used to determine a performance parameter node that meets the requirements according to the second parameter information.

[0153] In some embodiments, the evaluation module 72 includes a first evaluation module and a second evaluation module. The first evaluation module is used to generate a number of test points; determine the distance from each test point to each performance parameter node, and count the target test points whose distances meet the preset requirements; and use the target test points to determine the Voronoi unit corresponding to each performance parameter node, and use the Voronoi unit as the global evaluation result.

[0154] The second evaluation module is used to determine the target neighborhood node of each performance parameter node in the parameter space; and to fit each performance parameter node and the corresponding target neighborhood node to form a hyperplane; and to obtain a local linear approximation based on the hyperplane; and to obtain a local evaluation result based on the actual response value corresponding to each target neighborhood node and the local linear approximation.

[0155] The second evaluation module is also used to obtain the gradient information of each performance parameter node in different dimensions based on the hyperplane; and obtain the local linear approximation value according to the gradient information.

[0156] The second evaluation module is also used to determine the first average distance between all the neighboring nodes to be determined and each performance parameter node; determine the minimum average distance between all the neighboring nodes to be determined; and screen according to the minimum average distance and the first average distance to obtain the target neighboring node of each performance parameter node in the parameter space.

[0157] The first processing module 73 is further used to determine a target performance parameter node corresponding to the maximum value of the comprehensive evaluation result; and determine a node in the neighborhood of the target performance parameter node as a newly added performance parameter node to be added to the performance parameter node group.

[0158] The second processing module 74 is also used to input the first parameter information of all performance parameter nodes in the performance parameter node group into the reconstruction model, so that the reconstruction model outputs all performance parameter nodes in the parameter space and the corresponding second parameter information; wherein the reconstruction model is based on Kriging interpolation, compressed sensing reconstruction or neural network construction.

[0159] The acquisition module 71 , the evaluation module 72 , the first processing module 73 , the second processing module 74 and the third processing module 75 cooperate with each other to implement the method of any of the above embodiments.

[0160] See also Figure 8 , Figure 8 1 is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. The electronic device 80 includes a processor 81 and a memory 82 coupled to the processor 81;

[0161] The memory 82 is used to store computer programs, and the processor 81 is used to execute computer programs to implement the following method:

[0162] Acquire a performance parameter node group; the performance parameter node group includes at least two performance parameter nodes, each performance parameter node includes a number of first parameter information; perform a global evaluation on the performance parameter node group to obtain a global evaluation result; and perform a local evaluation on each performance parameter node in the performance parameter node group to obtain a local evaluation result; combine the global evaluation result and the local evaluation result to obtain a comprehensive evaluation result of each performance parameter node; add new performance parameter nodes to the performance parameter node group based on the comprehensive evaluation result; in response to the number of performance parameter nodes in the performance parameter node group being greater than a preset value, reconstruct all performance parameter nodes in the parameter space and the corresponding second parameter information using the first parameter information of all performance parameter nodes in the performance parameter node group; determine the performance parameter nodes that meet the requirements based on the second parameter information.

[0163] It can be understood that the processor 81 is used to execute a computer program to implement the method of any of the above embodiments.

[0164] See also Fig. 9 , Fig. 91 is a schematic diagram of the structure of another embodiment of the electronic device provided by the present application. The electronic device 110 may be, for example, a terminal device. The electronic device 110 may include: a memory 112, a processor (Central Processing Unit, CPU) 111, a circuit board (not shown in the figure), a power circuit and a microphone 123. The circuit board is placed inside the space surrounded by the shell; the processor 111 and the memory 112 are arranged on the circuit board; the power circuit is used to supply power to various circuits or devices of the electronic device; the memory 112 is used to store executable program code; the processor 111 runs a computer program corresponding to the executable program code by reading the executable program code stored in the memory 112, so as to implement the method of any of the above embodiments.

[0165] The electronic device may also include: a peripheral interface 114, an RF (Radio Frequency) circuit 116, an audio circuit 117, a speaker 122, a power management chip 119, an input / output (I / O) subsystem 120, other input / control devices 121, a display 113 and an external port 115. These components communicate through one or more communication buses or signal lines 118.

[0166] The memory 112 can be accessed by the processor 111, the peripheral interface 114, etc. The memory 112 can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more disk storage devices, flash memory devices, or other volatile solid-state storage devices. The peripheral interface 114 can connect the input and output peripherals of the device to the processor 111 and the memory 112.

[0167] The I / O subsystem 120 can connect the input and output peripherals on the device, such as the display 113 and other input / control devices 121, to the peripheral interface 114. The I / O subsystem 120 may include a display controller 1201 and one or more input controllers 1202 for controlling other input / control devices 121. Among them, the one or more input controllers 1202 receive electrical signals from other input / control devices 121 or send electrical signals to other input / control devices 121. Other input / control devices 121 may include physical buttons (press buttons, rocker buttons, etc.), dials, slide switches, joysticks, and click wheels. It is worth noting that the input controller 1202 can be connected to any of the following: a keyboard, an infrared port, a USB interface, and a pointing device such as a mouse.

[0168] The display 113 is an input interface and an output interface between the user electronic device and the user, and displays visual output to the user. The visual output may include graphics, text, icons, videos, etc.

[0169] The display controller 1201 in the I / O subsystem 120 receives an electrical signal from the display 113 or sends an electrical signal to the display 113. The display 113 detects contact on the touch screen, and the display controller 1201 converts the detected contact into interaction with a user interface object displayed on the display 113, that is, realizes human-computer interaction. The user interface object displayed on the display 113 may be an icon for running a game, an icon for connecting to a corresponding network, etc.

[0170] The RF circuit 116 is mainly used to establish communication between the mobile phone and the wireless network (i.e., the network side) to realize data reception and transmission between the mobile phone and the wireless network. For example, sending and receiving short messages, emails, etc. Specifically, the RF circuit 116 receives and sends RF signals, which are also called electromagnetic signals. The RF circuit 116 converts electrical signals into electromagnetic signals or converts electromagnetic signals into electrical signals, and communicates with the communication network and other devices through the electromagnetic signals. The RF circuit 116 may include known circuits for performing these functions, including but not limited to antenna systems, RF transceivers, one or more amplifiers, tuners, one or more oscillators, digital signal processors, CODEC (COder-DECoder, codec) chipsets, Subscriber Identity Module (SIM), etc.

[0171] The audio circuit 117 is mainly used to receive audio data from the peripheral interface 114, convert the audio data into an electrical signal, and send the electrical signal to the speaker 122. The speaker 122 is used to restore the voice signal received by the mobile phone from the wireless network through the RF circuit 116 to sound and play the sound to the user. The power management chip 119 is used to provide power supply and power management for the processor 111, the I / O subsystem 120 and the hardware connected to the peripheral interface 114.

[0172] The electronic device 110 mentioned above may be a mobile terminal, such as a smart phone, a tablet or other mobile device, or a terminal device, such as a personal computer, or a corresponding server.

[0173] See also Fig.10 , Fig.10 1 is a schematic diagram of a computer readable storage medium according to an embodiment of the present application. The computer readable storage medium 150 stores a computer program 151. When the computer program 151 is executed by a processor, the following method is implemented:

[0174] Acquire a performance parameter node group; the performance parameter node group includes at least two performance parameter nodes, each performance parameter node includes a number of first parameter information; perform a global evaluation on the performance parameter node group to obtain a global evaluation result; and perform a local evaluation on each performance parameter node in the performance parameter node group to obtain a local evaluation result; combine the global evaluation result and the local evaluation result to obtain a comprehensive evaluation result of each performance parameter node; add new performance parameter nodes to the performance parameter node group based on the comprehensive evaluation result; in response to the number of performance parameter nodes in the performance parameter node group being greater than a preset value, reconstruct all performance parameter nodes in the parameter space and the corresponding second parameter information using the first parameter information of all performance parameter nodes in the performance parameter node group; determine the performance parameter nodes that meet the requirements based on the second parameter information.

[0175] It can be understood that when the computer program 151 is executed by a processor, it can also implement the method of any of the above embodiments.

[0176] In summary, the performance parameter determination method, device, electronic device and readable storage medium provided in the present application, in the process of determining the performance parameters, use a combination of global evaluation and local evaluation to conduct a comprehensive evaluation on a small number of performance parameter nodes, improve the evaluation accuracy, and expand the evaluation scope by gradually increasing the performance parameter nodes. Then, when the number of performance parameter nodes in the performance parameter node group is greater than the preset value, all performance parameter nodes in the parameter space and the corresponding second parameter information are reconstructed using the first parameter information of all performance parameter nodes in the performance parameter node group, and the performance parameter nodes that meet the requirements are determined based on the second parameter information, thereby reducing the number of evaluation iterations, reducing the determination time of performance parameter test nodes, and improving work efficiency.

[0177] In the several embodiments provided in this application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only illustrative, for example, the division of the modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0178] If the integrated units in the above other embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), disk or optical disk and other media that can store program codes.

[0179] The above description is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for determining performance parameters, characterized in that: The method comprises: Acquire a performance parameter node group; the performance parameter node group includes at least two performance parameter nodes, each of which includes a plurality of first parameter information; Performing a global evaluation on the performance parameter node group to obtain a global evaluation result; and performing a local evaluation on each of the performance parameter nodes in the performance parameter node group to obtain a local evaluation result; A comprehensive evaluation result of each of the performance parameter nodes combining the global evaluation result and the local evaluation result; Adding new performance parameter nodes to the performance parameter node group according to the comprehensive evaluation results; In response to the number of performance parameter nodes in the performance parameter node group being greater than a preset value, reconstructing all performance parameter nodes and corresponding second parameter information in a parameter space using the first parameter information of all the performance parameter nodes in the performance parameter node group; A performance parameter node that meets the requirements is determined according to the second parameter information.

2. The method according to claim 1, characterized in that The globally evaluating the performance parameter node group to obtain a global evaluation result includes: Generate several test points; Determine the distance from each test point to each of the performance parameter nodes, and count the target test points whose distances meet preset requirements; The target test point is used to determine the Voronoi unit corresponding to each of the performance parameter nodes, and the Voronoi unit is used as the global evaluation result.

3. The method according to claim 1, characterized in that The locally evaluating each of the performance parameter nodes in the performance parameter node group to obtain a local evaluation result includes: Determining a target neighboring node of each performance parameter node in the parameter space; Fitting each of the performance parameter nodes and the corresponding target neighborhood nodes to form a hyperplane; Obtaining a local linear approximation based on the hyperplane; The local evaluation result is obtained based on the actual response value corresponding to each target neighborhood node and the local linear approximation.

4. The method according to claim 3, characterized in that The obtaining of a local linear approximation based on the hyperplane comprises: Obtaining gradient information of each performance parameter node in different dimensions based on the hyperplane; The local linear approximation is obtained according to the gradient information.

5. The method according to claim 3, characterized in that: The determining of a target neighboring node of each performance parameter node in the parameter space includes: Determine a first average distance between all the neighboring nodes to be determined and each of the performance parameter nodes; Determine the minimum average distance between all neighboring nodes to be determined; Screening is performed according to the minimum average distance and the first average distance to obtain a target neighboring node of each performance parameter node in the parameter space.

6. The method according to claim 1, characterized in that The adding a new performance parameter node to the performance parameter node group according to the comprehensive evaluation result includes: Determine the target performance parameter node corresponding to the maximum value of the comprehensive evaluation result; A node is determined in a neighborhood of the target performance parameter node as a newly added performance parameter node and added to the performance parameter node group.

7. The method according to claim 1, characterized in that The reconstructing all the performance parameter nodes and the corresponding second parameter information in the parameter space by using the first parameter information of all the performance parameter nodes in the performance parameter node group includes: The first parameter information of all the performance parameter nodes in the performance parameter node group is input into a reconstruction model so that the reconstruction model outputs all the performance parameter nodes in the parameter space and the corresponding second parameter information; wherein the reconstruction model is constructed based on Kriging interpolation, compressed sensing reconstruction or neural network.

8. A performance parameter determination device, characterized in that: The performance parameter determination device comprises: An acquisition module, used for acquiring a performance parameter node group; the performance parameter node group includes at least two performance parameter nodes, each of which includes a plurality of first parameter information; An evaluation module, configured to perform a global evaluation on the performance parameter node group to obtain a global evaluation result; and perform a local evaluation on each of the performance parameter nodes in the performance parameter node group to obtain a local evaluation result; and a comprehensive evaluation result of each of the performance parameter nodes combining the global evaluation result and the local evaluation result; A first processing module, configured to add a new performance parameter node to the performance parameter node group according to the comprehensive evaluation result; A second processing module, configured to reconstruct all performance parameter nodes and corresponding second parameter information in a parameter space using the first parameter information of all the performance parameter nodes in the performance parameter node group in response to the number of the performance parameter nodes in the performance parameter node group being greater than a preset value; The third processing module is used to determine a performance parameter node that meets the requirements according to the second parameter information.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory coupled to the processor; The memory is used to store a computer program, and the processor is used to execute the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 7.

Images