CN111176758A - Configuration parameter recommendation method and device, terminal and storage medium - Google Patents
Configuration parameter recommendation method and device, terminal and storage medium Download PDFInfo
- Publication number
- CN111176758A CN111176758A CN201911415084.4A CN201911415084A CN111176758A CN 111176758 A CN111176758 A CN 111176758A CN 201911415084 A CN201911415084 A CN 201911415084A CN 111176758 A CN111176758 A CN 111176758A
- Authority
- CN
- China
- Prior art keywords
- game
- configuration parameters
- terminal
- configuration
- recommended
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/445—Program loading or initiating
- G06F9/44505—Configuring for program initiating, e.g. using registry, configuration files
Abstract
The application discloses a configuration parameter recommendation method, a configuration parameter recommendation device, a terminal and a storage medium. The method comprises the following steps: in the process of running a game application program by a terminal, obtaining game configuration parameters, terminal configuration parameters and game running data in real time; determining a recommended game configuration parameter and a recommended terminal configuration parameter according to the game configuration parameter, the terminal configuration parameter and the game operation data through a configuration recommendation model; and sending the recommended game configuration parameters to the game application program, and sending the recommended terminal configuration parameters to the operating system. According to the method and the device, the game configuration parameters and the terminal configuration parameters are dynamically recommended in real time in the game running process, and compared with the method and the device for statically determining the configuration parameters in the related technology, the technical scheme provided by the embodiment of the application can dynamically recommend the configuration parameters in real time, and the flexibility and the accuracy of recommending the configuration parameters are improved.
Description
Technical Field
The embodiment of the application relates to the technical field of artificial intelligence, in particular to a configuration parameter recommendation method, a configuration parameter recommendation device, a terminal and a storage medium.
Background
Currently, various game applications can be installed in terminals such as mobile phones, tablet computers, and the like. As the functions of game applications become more rich, the requirements for terminal configuration become higher and higher.
In the related art, based on a large amount of user history data and manual experience, recommended game configuration parameters and recommended terminal configuration parameters are provided to a game application, for example, the recommended game configuration parameters may include recommended picture quality, frame number setting, anti-aliasing, and other parameters, and for example, the recommended terminal configuration parameters may include recommended CPU (Central Processing Unit) shift, GPU (Graphics Processing Unit) shift, maximum frame rate, and other parameters. And in the process of running the game application program, the terminal can run according to the recommended game configuration parameters and the recommended terminal configuration parameters so as to improve the game performance as much as possible.
The scheme provided by the related technology determines the recommended configuration parameters based on a large amount of user historical data and manual experience, and the accuracy is not high.
Disclosure of Invention
The embodiment of the application provides a recommendation method, a recommendation device, a terminal and a storage medium for configuration parameters, which can be used for solving the technical problem of low accuracy in the prior art that recommended configuration parameters are determined based on a large amount of user historical data and artificial experience. The technical scheme is as follows:
in one aspect, an embodiment of the present application provides a method for recommending configuration parameters, where the method includes:
in the process of running a game application program by a terminal, obtaining game configuration parameters, terminal configuration parameters and game running data in real time;
determining a recommended game configuration parameter and a recommended terminal configuration parameter according to the game configuration parameter, the terminal configuration parameter and the game running data through a configuration recommendation model;
and sending the recommended game configuration parameters to the game application program, and sending the recommended terminal configuration parameters to an operating system.
On the other hand, an embodiment of the present application provides a device for recommending configuration parameters, where the device includes:
the data acquisition module is used for acquiring game configuration parameters, terminal configuration parameters and game running data in real time in the process of running the game application program by the terminal;
the parameter determination module is used for determining recommended game configuration parameters and recommended terminal configuration parameters according to the game configuration parameters, the terminal configuration parameters and the game running data through a configuration recommendation model;
and the parameter sending module is used for sending the recommended game configuration parameters to the game application program and sending the recommended terminal configuration parameters to an operating system.
In another aspect, an embodiment of the present application provides a terminal, where the terminal includes a processor and a memory, where the memory stores at least one instruction, at least one program, a code set, or an instruction set, and the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the recommendation method for configuration parameters.
In yet another aspect, an embodiment of the present application provides a computer-readable storage medium, in which at least one instruction, at least one program, a code set, or a set of instructions is stored, and the at least one instruction, the at least one program, the code set, or the set of instructions is loaded and executed by a processor to implement the recommended method for configuring parameters.
In a further aspect, an embodiment of the present application provides a computer program product, which, when running on a computer device, causes the computer device to execute the method for recommending configuration parameters.
The technical scheme provided by the embodiment of the application can bring the following beneficial effects:
the method comprises the steps that game configuration parameters, terminal configuration parameters and game running data are obtained in real time in the process that a terminal runs a game application program, recommended game configuration parameters and recommended terminal configuration parameters are determined according to the data through a configuration recommendation model, then the recommended game configuration parameters are sent to the game application program, and the recommended terminal configuration parameters are sent to an operating system; the game configuration parameters and the terminal configuration parameters are dynamically recommended in real time in the game running process, and compared with the method for statically determining the configuration parameters in the related technology, the technical scheme provided by the embodiment of the application can dynamically recommend the configuration parameters in real time, so that the flexibility and the accuracy of recommending the configuration parameters are improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic diagram of a perceptron model provided in one embodiment of the present application;
FIG. 2 is a schematic diagram of a neural network model provided by one embodiment of the present application;
FIG. 3 is a schematic diagram of a neural network model provided in another embodiment of the present application;
FIG. 4 is a schematic diagram of a DNN model provided by one embodiment of the present application;
FIGS. 5 and 6 are schematic diagrams of a 3-layer DNN model provided by one embodiment of the present application;
FIG. 7 is a schematic diagram of a system architecture provided by one embodiment of the present application;
FIG. 8 is a flowchart of a method for recommending configuration parameters according to an embodiment of the present application;
FIG. 9 is a schematic diagram of a model training process provided by one embodiment of the present application;
FIG. 10 is a schematic illustration of the overall flow of the solution provided by one embodiment of the present application;
FIG. 11 is a schematic illustration of a configuration interface provided by one embodiment of the present application;
FIG. 12 is a block diagram of an apparatus for recommending configuration parameters according to an embodiment of the present application;
FIG. 13 is a block diagram of an apparatus for recommending configuration parameters according to another embodiment of the present application;
fig. 14 is a block diagram of a terminal according to an embodiment of the present application.
Detailed Description
To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
AI (Artificial Intelligence) is a theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human Intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technique of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence. Artificial intelligence is the research of the design principle and the realization method of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.
The artificial intelligence technology is a comprehensive subject and relates to the field of extensive technology, namely the technology of a hardware level and the technology of a software level. The artificial intelligence infrastructure generally includes technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and the like.
ML (Machine Learning) is a multi-domain cross discipline, and relates to a plurality of disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and the like. The special research on how a computer simulates or realizes the learning behavior of human beings so as to acquire new knowledge or skills and reorganize the existing knowledge structure to continuously improve the performance of the computer. Machine learning is the core of artificial intelligence, is the fundamental approach for computers to have intelligence, and is applied to all fields of artificial intelligence. Machine learning and deep learning generally include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and teaching learning.
With the research and progress of artificial intelligence technology, the artificial intelligence technology is developed and applied in a plurality of fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned driving, automatic driving, unmanned aerial vehicles, robots, smart medical care, smart customer service, and the like.
The scheme provided by the embodiment of the application relates to an artificial intelligence machine learning technology, and provides a configuration parameter recommendation method.
The configuration recommendation model related in the embodiment of the present application may be a machine learning model constructed based on a Neural network, such as DNN (Deep Neural Networks).
Neural network technology originated in the fifth and sixty years of the last century, when called perceptron (perceptron), which has an input layer, an output layer and a hidden layer (or called hidden layer). The input feature vectors reach an output layer through hidden layer transformation, and classification results are obtained at the output layer. However, the Rosenblatt single-layer perceptron has a problem that it is not as serious as possible, i.e., it does not perform as well for somewhat more complex functions (such as the most typical XOR operation). Most current learning methods such as classification and regression are shallow structure algorithms, and the method has the limitation that the expression capability of complex functions is limited under the condition of limited samples and computing units, and the generalization capability of the method is limited to a certain extent aiming at the problem of complex classification.
With the development of mathematics, this drawback was not overcome until the eighties of the last century by the multi-layer perceptron (multilayer perceptron) invented by Rumelhart, Williams, Hinton, LeCun et al. The multi-layer perceptron, as the name implies, is a perceptron with a plurality of hidden layers.
The multilayer perceptron can get rid of the constraint of an early discrete transfer function, the response of continuous function analog neurons such as sigmoid or tanh to excitation is used, and a BP (Back Propagation) algorithm invented by Werbos is used on a training algorithm. The deep neural network can realize complex function approximation by learning a deep nonlinear network structure, represent input data distributed representation, and show strong capability of learning essential characteristics of a data set from a few sample sets.
The multilayer perceptron overcomes the defect that the XOR logic cannot be simulated before, and simultaneously, the network can better depict the complex situation in the real world due to the more layers. The multilayer perceptron suggests us that the number of layers in a neural network directly determines its portrayal of reality-fitting a more complex function with fewer neurons per layer.
As the number of layers of the neural network increases, the optimization function is more likely to fall into a local optimal solution, and the "trap" is more and more deviated from a true global optimal solution. Deep networks trained with limited data do not perform as well as shallower networks. Meanwhile, another non-negligible problem is that the "gradient vanishing" phenomenon becomes more serious as the number of network layers increases. In particular, we often use sigmoid as an input-output function for neurons. For a signal with amplitude 1, the gradient decays to 0.25 per layer of propagation in the BP counter-propagating gradient. The number of layers is more than one, and the lower layer basically cannot receive effective training signals after the gradient exponential decay.
The essence of deep learning is to learn more useful features by constructing a machine learning model with many hidden layers and massive training data, thereby finally improving the accuracy of classification or prediction. Thus, "depth model" is a means and "feature learning" is a goal. Different from the traditional shallow learning, the deep learning is different in that: 1) emphasizes the depth of the model structure, and usually has hidden layer nodes of 5 layers, 6 layers and even 10 layers; 2) the importance of feature learning is clearly highlighted, that is, the feature representation of the sample in the original space is transformed into a new feature space through layer-by-layer feature transformation, so that the classification or the prediction is easier. Compared with a method for constructing the features by using manual rules, the method for constructing the features by using the big data to learn the features can depict rich intrinsic information of the data.
In the following, the mathematical principle of DNN is described, and we describe it from perceptron to neural network:
the model of the perceptron, which is a model with several inputs and one output, as shown in fig. 1, learns a linear relationship between the outputs and the inputs, and obtains an intermediate output result:
where m denotes the number of inputs, xiDenotes the ith input, wiRepresenting the ith input xiThe corresponding linear relation coefficient, b is the bias, z represents the output, and i is a positive integer less than or equal to m.
This is followed by a neuron activation function:
thereby obtaining the desired result of 1 or-1.
This model can only be used for binary classification and cannot learn more complicated non-linear models, and therefore cannot be used in the industry. The neural network is expanded on a model of a perception machine, and the following three main points are summarized:
(1) and a hidden layer is added, the hidden layer can have multiple layers, the expression capability of the model is enhanced, and as shown in FIG. 2, 2 hidden layers are included between an input layer and an output layer. Of course the complexity of such a number of hidden layer models increases considerably.
(2) The neuron of the output layer can have more than one output, and the model can be flexibly applied to classification regression and other machine learning fields, such as dimension reduction, clustering and the like. An example of an output layer of a plurality of neuron outputs is shown in fig. 3, where the output layer includes 4 neurons.
(3) The activation function is expanded, the activation function of the perceptron is sign (z), although simple, the processing capability is limited, so other activation functions are generally used in the neural network, such as the Sigmoid function we used in the logistic regression, namely:
there are also tan x, softmax and ReLU, etc. that appear later. The expressive power of the neural network is further enhanced by using different activation functions.
Next, the basic structure of DNN will be described.
While neural networks are based on extensions of the perceptron, DNN can be understood as neural networks with many hidden layers. Multilayer neural networks and DNNs are also really just one thing that is referred to, DNNs sometimes also being called MLPs (Multi-Layer Perceptron).
From the DNN, which is divided by the positions of different layers, the neural network layers inside the DNN can be divided into three types, an input layer, a hidden layer and an output layer, as shown in fig. 4, generally, the first layer is an input layer, the last layer is an output layer, and the middle layers are hidden layers.
The layers are all connected, that is, any neuron of the ith layer is necessarily connected with any neuron of the (i + 1) th layer. Although DNN appears to be complex, it is still a sense-machine from the small local modelSame, i.e. a linear relationship
Plus an activation function sigma (z).
Due to the large number of DNN layers, the coefficient w of the linear relation and the number of the bias b are large. The definitions of the two parameters mentioned above in DNN are described below.
First look at the definition of the linear relationship coefficient w. Taking the DNN of one three layer as an example shown in FIG. 5, the linear relationship from the 4 th neuron of the second layer to the 2 nd neuron of the third layer is defined as
The superscript 3 represents the number of layers in which the linear coefficient w is located, while the subscripts correspond to the third layer index 2 of the output and the second layer index 4 of the input. You may ask why it is not
Is there? This is primarily to facilitate the use of the model for matrix representation operations, if any
And each time the matrix operation is performed is ωTx + b, the transposition is required. If the output index is placed in front, the linear operation is directly ω x + b without transposition. The linear coefficient from the kth neuron of the l-1 layer to the jth neuron of the l layer is defined as
Note that the input layer is without the ω parameter.
See again the definition of bias b. Also for example with three layers of DNN, as shown in FIG. 6, the bias corresponding to the third neuron in the second layer is defined as
Where the superscript 2 represents the number of layers present and the subscript 3 represents the index of the neuron on which the bias is present. For the same reason, the first neuron in the third layerShould be expressed as
Likewise, the output layer is free of bias parameters.
In the method provided by the embodiment of the present application, the execution subject of each step may be a terminal, and the terminal may be an electronic device such as a mobile phone, a tablet Computer, a game device, a multimedia playing device, a wearable device, and a PC (Personal Computer). In the embodiment of the present application, the type of the Game application is not limited, and may be, for example, a shooting Game, a MOBA (Multiplayer Online Battle Arena) Game, an RPG (Role-playing Game), and the like.
Referring to fig. 7, a schematic diagram of a system architecture according to an embodiment of the present application is shown. As shown in fig. 7, the system architecture may include: a large data platform 10 and a terminal 20.
The big data platform 10 may include: a data platform 11 and a model training platform 12. The data platform 11 is a platform responsible for data storage, calculation and analysis, and is responsible for cleaning and storing data reported by the game client to a data warehouse. The data warehouse may be an HDFS (Hadoop Distributed File System). The model training platform 12 is configured to periodically extract data from the data warehouse of the data platform 11 to construct training samples, and perform offline training on a configuration recommendation model, which may be a DNN model, through the training samples.
The processor 21 of the terminal 20 runs a game application (i.e., a client of the game application), and the game application may include a performance optimization SDK (Software Development Kit) module 22, and the performance optimization SDK module 22 is a functional module for implementing the flow of the method of the present application.
As shown in fig. 7, the performance optimization SDK module 22 may include a real-time calculation module 221 and a data acquisition module 222. The real-time computing module 221 obtains the configuration recommendation model trained offline from the background (i.e., the model training platform 12), computes the game recommendation configuration parameters and the terminal recommendation configuration parameters in real time through the configuration recommendation model, sends the game recommendation configuration parameters to the game application program, and sends the terminal recommendation configuration parameters to the operating system of the terminal. The data acquisition module 222 is configured to acquire game configuration parameters, terminal configuration parameters, and game running data in real time during the process of running the game application program by the terminal, and report the data to the data platform 11, so that the model training platform 12 updates the configuration recommendation model in an offline manner. In addition, the real-time computing module 221 is further configured to update the local configuration recommendation model of the terminal in real time by using the data.
Alternatively, the model training platform 12 may be a TensorFlow platform, which is an end-to-end open source machine learning platform. It has a comprehensive and flexible ecosystem containing various tools, libraries and community resources, allowing researchers to drive the development of advanced technologies in the field of machine learning, and allowing developers to easily build and deploy applications supported by machine learning. Accordingly, the performance optimization SDK module is integrated with a tensrflow Lite, which is an open source deep learning framework for device-side inference. TensorFlow Lite is a set of tools used to help developers run TensorFlow models on mobile, embedded, and Internet of things devices. It supports machine learning reasoning on the device with low latency and small binary size.
Referring to fig. 8, a flowchart of a method for recommending configuration parameters according to an embodiment of the present application is shown, where an execution subject of the method may be a processor of a terminal, and the method may include the following steps (801 to 803):
Game configuration parameters refer to configurable parameters of the game application, including but not limited to at least one of: picture quality, frame number settings, picture style, anti-aliasing, shading, etc.
The terminal configuration parameter refers to a configurable parameter of the terminal, including but not limited to at least one of the following: CPU gear, GPU gear, game scene holding time, game scene importance level, maximum frame rate and other parameters.
The game running data refers to real-time related data of the game application program in the running process, and includes but is not limited to at least one of the following: scene identification, target frame rate, current frame rate, map identification, the number of people in the same screen of the game, memory utilization rate, CPU utilization rate, terminal temperature and the like. The game play data may change dynamically in real time as the game progresses. For example, a switch may be made from scene A to scene B, the number of people on the same screen of the game may be changed from 2 to 3, and so on.
In the embodiment of the present application, the acquisition periods of the game configuration parameters, the terminal configuration parameters, and the game operation data are not limited. For example, the game configuration parameters may be acquired after each update, such as if the game configuration parameters are updated every 5 seconds, then the acquisition period of the game configuration parameters is 5 seconds; the terminal configuration parameter may also be acquired after each update, for example, if the terminal configuration parameter is updated every 100 milliseconds, the acquisition frequency of the terminal configuration parameter is 100 milliseconds; the game running data is dynamically changed in real time, so that the obtaining period of the game running data is shorter than the obtaining periods of the game configuration parameters and the terminal configuration parameters, for example, the obtaining period of the game running data can be 5 milliseconds. Of course, the above description of the acquisition period of the game configuration parameters, the terminal configuration parameters and the game operation data is only exemplary and explanatory, and may be reasonably set according to actual requirements in practical applications, which is not limited in the embodiment of the present application.
The configuration recommendation model is a machine learning model used for calculating the configuration parameters of the recommended games and the configuration parameters of the recommended terminals. Optionally, the configuration recommendation model may be a DNN model, i.e. the configuration recommendation model may be built based on DNN. The configuration recommendation model is trained to have the capability of calculating the configuration parameters of the recommended games and the configuration parameters of the recommended terminals, so that the configuration parameters of the recommended games and the configuration parameters of the recommended terminals can be calculated according to the configuration parameters of the games, the configuration parameters of the terminals and the game running data.
The recommended game configuration parameters refer to recommended (or referred to as suggested) game configuration parameters, and as such, the recommended game configuration parameters include, but are not limited to, at least one of: picture quality, frame number settings, picture style, anti-aliasing, shading, etc.
The recommended terminal configuration parameters refer to recommended (or referred to as suggested) terminal configuration parameters, and as such, the recommended terminal configuration parameters include, but are not limited to, at least one of the following: CPU gear, GPU gear, game scene holding time, game scene importance level, maximum frame rate and other parameters.
Because the game configuration parameters and the terminal configuration parameters can be adjusted along with the time dimension, the recommended game configuration parameters and the recommended terminal configuration parameters in a certain target time period in the future can be calculated based on the existing game configuration parameters, terminal configuration parameters and game running data in a historical time period.
It should be noted that, in the embodiment of the present application, since the game configuration parameters, the terminal configuration parameters, and the game running data are obtained by the processor in real time during the process of running the game application program by the terminal, the processor may also determine the recommended game configuration parameters and the recommended terminal configuration parameters in real time by configuring the recommendation model. That is, in the process of running the game application program, new recommended game configuration parameters and new recommended terminal configuration parameters are continuously calculated according to continuously updated data, so that real-time dynamic adjustment of the game configuration parameters and the terminal configuration parameters is realized.
After determining the recommended game configuration parameters and the recommended terminal configuration parameters, the processor may send the recommended game configuration parameters to the game application program, and may also send the recommended terminal configuration parameters to the operating system. After receiving the recommended game configuration parameters, the game application program may determine the actually adopted game configuration parameters according to the recommended game configuration parameters, and operate according to the actually adopted game configuration parameters. After receiving the recommended terminal configuration parameters, the operating system may determine actually-used terminal configuration parameters according to the recommended terminal configuration parameters, and operate according to the actually-used terminal configuration parameters.
It should be noted that the game configuration parameters actually adopted by the game application program may be the same as or different from the received recommended game configuration parameters, and the recommended game configuration parameters serve as a data reference for the game application program in determining the game configuration parameters actually adopted, so that the game application program is more reasonable and accurate in determining the game configuration parameters actually adopted, and the game experience of the user is better improved. Similarly, the terminal configuration parameters actually adopted by the operating system of the terminal may be the same as or different from the received recommended terminal configuration parameters, and the recommended terminal configuration parameters serve as a data reference for the operating system when determining the actually adopted terminal configuration parameters, so that the operating system is more reasonable and accurate when determining the actually adopted terminal configuration parameters, and the game experience of the user is better improved. The improvement of the game experience includes, but is not limited to, reducing the jamming, reducing the time delay, improving the picture quality, and so on.
To sum up, in the technical solution provided in the embodiment of the present application, in the process of running a game application program at a terminal, a game configuration parameter, a terminal configuration parameter, and game running data are obtained in real time, a recommended game configuration parameter and a recommended terminal configuration parameter are determined according to the data by a configuration recommendation model, then the recommended game configuration parameter is sent to the game application program, and the recommended terminal configuration parameter is sent to an operating system; the game configuration parameters and the terminal configuration parameters are dynamically recommended in real time in the game running process, and compared with the method for statically determining the configuration parameters in the related technology, the technical scheme provided by the embodiment of the application can dynamically recommend the configuration parameters in real time, so that the flexibility and the accuracy of recommending the configuration parameters are improved.
In an exemplary embodiment, configuring the recommendation model includes: a game configuration recommendation model and a terminal configuration recommendation model. Wherein the game configuration recommendation model is a machine learning model for calculating recommended game configuration parameters, and the game configuration recommendation model may be a DNN model; the terminal configuration recommendation model is a machine learning model used for calculating and recommending terminal configuration parameters, and the terminal configuration recommendation model can also be a DNN model.
The input characteristics of the game configuration recommendation model can comprise terminal configuration parameters and game running data, and the game configuration recommendation model outputs recommended game configuration parameters according to the input characteristics. Optionally, the processor generates, by the game configuration recommendation model, a recommended game configuration parameter in the first target time period according to the terminal configuration parameter and the game running data in the first historical time period. The first history period refers to a time period before the current time, and assuming that the current time is t0, the first history period may be a [ t1, t0] time period or a [ t2, t3] time period; here, t1 is a time before the current time t0, t3 is a time before the current time t0, and t2 is a time before t 3. The duration of the first history period may be preset, which is not limited in this embodiment of the application, and for example, the duration of the first history period may be 5 s. The first target time interval is a time interval after the current time, and assuming that the current time is t0, the first target time interval may be a time interval [ t0, t4], or a time interval [ t5, t6 ]; where t4 is a time after the current time t0, t5 is a time after the current time t0, and t6 is a time after t 5. The duration of the first target time period may be preset, and the durations of the first target time period and the first history time period may be the same or different.
The input characteristics of the terminal configuration recommendation model can comprise game configuration parameters and game running data, and the terminal configuration recommendation model outputs the recommended terminal configuration parameters according to the input characteristics. Optionally, the processor generates, by the terminal configuration recommendation model, recommended terminal configuration parameters in a second target time period according to the game configuration parameters and the game running data in the second historical time period. The second history period refers to a time period before the current time, and assuming that the current time is t0, the second history period may be a time period [ t7, t0] or a time period [ t8, t9 ]; here, t7 is a time before the current time t0, t9 is a time before the current time t0, and t8 is a time before t 9. The duration of the second history period may be preset, and this is not limited in this embodiment of the application, and the duration of the second history period may be, for example, 1s or 100 ms. The second target time period is a time period after the current time, and assuming that the current time is t0, the second target time period may be a time period [ t0, t10] or a time period [ t11, t12 ]; where t10 is a time after the current time t0, t11 is a time after the current time t0, and t12 is a time after t 11. The duration of the second target period may be preset, and the durations of the second target period and the second historical period may be the same or different, which is not limited in this application embodiment, and for example, the duration of the second target period is the same as the duration of the second historical period, and is both 100 ms.
In addition, in the embodiment of the present application, the period for calculating the recommended game configuration parameter and the period for calculating the recommended terminal configuration parameter may be the same or different, which may be preset in combination with the actual situation (such as the requirement of the computing capability of the terminal and the update frequency of the two configuration parameters), and the embodiment of the present application does not limit this.
Optionally, the processor may also adjust parameters of the game configuration recommendation model and the terminal configuration recommendation model locally at the terminal, that is, continuously train and update the two models locally at the terminal, so that the two models can better adapt to personalized requirements of the terminal user, and provide recommended configuration parameters according with habits of the user.
Taking training and updating of the game configuration recommendation model locally at the terminal as an example, the processor acquires actual game configuration parameters in a first target time period, and adjusts the parameters of the game configuration recommendation model according to the terminal configuration parameters and game running data in a first historical time period and the actual game configuration parameters in the first target time period. The actual game configuration parameters in the first target time period refer to game configuration parameters actually adopted by the game application program in the first target time period. Based on the terminal configuration parameters and the game running data in the first historical period and the actual game configuration parameters in the first target period, the processor constructs a training sample and trains the game configuration recommendation model, so that the game configuration recommendation model can learn the latest use requirements of the terminal user, and recommended game configuration parameters which are more in line with the use requirements of the user are provided subsequently.
Taking training and updating of the terminal configuration recommendation model locally at the terminal as an example, the processor acquires actual terminal configuration parameters in a second target time period, and adjusts the parameters of the terminal configuration recommendation model according to the game configuration parameters and the game running data in a second historical time period and the actual terminal configuration parameters in the second target time period. The actual terminal configuration parameters in the second target time period refer to the terminal configuration parameters actually adopted by the operating system in the second target time period. Based on the game configuration parameters and the game running data in the second historical period and the actual terminal configuration parameters in the second target period, the processor constructs a training sample and trains the terminal configuration recommendation model, so that the terminal configuration recommendation model can learn the latest use requirements of the terminal user, and recommended terminal configuration parameters which are more in line with the use requirements of the user are provided subsequently.
In summary, in the technical solution provided in the embodiment of the present application, the game configuration recommendation model and the terminal configuration recommendation model are used to calculate the recommended game configuration parameters and the recommended terminal configuration parameters, so that the calculation of the two configuration parameters is decoupled, and thus, the game configuration parameter recommendation and the terminal configuration parameter recommendation can be performed at different time granularities, and flexibility is improved.
In addition, the two models are trained and updated locally at the terminal through the data acquired in real time, so that the models can learn the latest configuration requirements, and more accurate configuration parameters are provided subsequently. The terminal performance index is also considered while the user personalized game experience is effectively improved, so that the terminal resources are more effectively utilized, and the resource relation between the game side and the terminal side is efficiently coordinated.
In the embodiment of the present application, the configuration recommendation models (including the game configuration recommendation model and the terminal configuration recommendation model introduced above) may be trained offline in the background (e.g., the model training platform 12 shown in fig. 7), and the offline-trained configuration recommendation model is provided to the terminal, and the terminal performs configuration recommendation using the model.
The training data of the game configuration recommendation model comprises a first input feature and a first label; wherein the first input feature may include a terminal configuration parameter and game play data, and the terminal configuration parameter may include at least one of: the game playing data comprises parameters such as a CPU gear, a GPU gear, game scene holding time, game scene importance level, a maximum frame rate and the like, and the game running data can comprise at least one of the following parameters: scene identification, target frame rate, current frame rate, map identification, the number of people on the same screen of the game, memory utilization rate, CPU utilization rate, terminal temperature and other parameters. The first tag is a game configuration parameter, and the game configuration parameter may include at least one of the following: picture quality, frame number settings, picture style, anti-aliasing, shading, etc.
Illustratively, if the model training platform 12 is a TensorFlow platform, the Kera module of the TensorFlow platform may be employed to construct a DNN model as the game configuration recommendation model. In addition, the configuration parameters of the game configuration recommendation model may include parameters such as the number of neural network layers, the number of neurons, weights, activation functions, optimization functions, evaluation functions, training periods, and minimum training data sizes.
After the game configuration recommendation model is constructed, the game configuration recommendation model is trained by adopting the training samples generated based on the training data, and the training process of the model is actually the tuning process of the model parameters (namely the weights of all layers), so that the model can calculate and accurately output the parameters according to the input characteristics. The output parameters of the game configuration recommendation model are the recommended game configuration parameters, and may include parameters such as picture quality, frame number setting, picture style, anti-aliasing, shading, and the like.
The training data of the terminal configuration recommendation model comprise a second input feature and a second label; wherein the second input characteristic may include game configuration parameters and game play data, and the game configuration parameters may include at least one of: picture quality, frame number setting, picture style, anti-aliasing, shading, etc., and the game play data may include at least one of: scene identification, target frame rate, current frame rate, map identification, the number of people on the same screen of the game, memory utilization rate, CPU utilization rate, terminal temperature and other parameters. The second tag is a terminal configuration parameter, and the terminal configuration parameter may include at least one of the following: CPU gear, GPU gear, game scene holding time, game scene importance level, maximum frame rate and other parameters.
Illustratively, if the model training platform 12 is a tensrflow platform, the DNN model may be constructed using a kera module of the tensrflow platform as the terminal configuration recommendation model. In addition, the configuration parameters of the terminal configuration recommendation model may also include parameters such as the number of neural network layers, the number of neurons, weights, activation functions, optimization functions, evaluation functions, training periods, and minimum training data scale.
After the terminal configuration recommendation model is constructed, the terminal configuration recommendation model is trained by adopting a training sample generated based on the training data, and the training process of the model is actually a tuning process of model parameters (namely weights of all layers), so that the model can calculate and accurately output parameters according to input characteristics. The output parameters of the terminal configuration recommendation model are the recommended terminal configuration parameters, and may include parameters such as a CPU gear, a GPU gear, a game scene retention time, a game scene importance level, a maximum frame rate, and the like.
Illustratively, as shown in FIG. 9, a schematic diagram of a model training process is shown. The model training platform 12 obtains user data including game configuration parameters, terminal configuration parameters and game operation data, then performs preprocessing and feature engineering on the data to generate training samples, trains the constructed configuration recommendation model by using the training samples, and then evaluates the trained model. And if the evaluation result is not in accordance with the expectation, adjusting the parameters of the model, and then retraining the model. And stopping training until the evaluation result of the model is in accordance with the expectation, and storing the model.
In an exemplary embodiment, the terminal may send a data reporting request to the server according to a reporting period, where the data reporting request includes the game configuration parameters, the terminal configuration parameters, and the game running data acquired in the reporting period. The reporting period may be preset in combination with an actual situation, for example, 1 day, 12 hours, 1 hour, and the like, which is not limited in this application. The server refers to a server in the big data platform 10 of the system architecture shown in fig. 7. Illustratively, after receiving the user data, the server in the data platform 11 may clean the user data and store the user data in the data warehouse. The server in the model training platform 12 may obtain the user data from the data warehouse, generate a training sample according to the user data, and train the configuration recommendation model using the training sample. The terminal can also obtain the latest configuration recommendation model from the server according to the synchronization period, wherein the latest configuration recommendation model is used for recommending the configuration parameters in real time. The synchronization period may be preset according to practical situations, such as 1 day, 12 hours, 1 hour, and the like, which is not limited in the embodiments of the present application. And after acquiring the latest configuration recommendation model from the server, the terminal calculates the real-time recommended configuration parameters by adopting the latest configuration recommendation model.
In the embodiment of the application, the time granularity of updating the configuration recommendation model by the server may be larger than the time granularity of updating the configuration recommendation model locally by the terminal, for example, the configuration recommendation model is updated by the server once a day, but the configuration recommendation model locally stored by the terminal can be trained and updated locally according to the user data acquired in real time, so that the model can learn the latest configuration requirement, and more accurate configuration parameters are provided subsequently.
Exemplarily, as shown in fig. 10, which shows a schematic diagram of an overall flow of a solution provided by an embodiment of the present application, the method may include the following steps:
101, a real-time computing module computes configuration parameters of a recommended game and configuration parameters of a recommended terminal in real time through a configuration recommendation model;
102, a real-time computing module sends recommended terminal configuration parameters to an operating system;
103, the operating system sends the actually adopted terminal configuration parameters to the real-time computing module;
104, the real-time computing module sends recommended game configuration parameters to the game application program;
105, the game application program carries out compliance check on the recommended game configuration parameters;
step 106, the game application program displays recommended game configuration parameters to the user;
step 107, the game application program obtains game configuration parameters actually adopted by the user;
step 108, the game application program sends the actually adopted game configuration parameters to the real-time computing module;
and step 109, the real-time computing module carries out parameter tuning on the configuration recommendation model according to the game configuration parameters and the terminal configuration parameters which are actually adopted.
In the embodiment of the present application, the execution order of the steps 102 to 103 and the steps 104 to 108 is not limited.
For example, as shown in fig. 11, the recommended game configuration parameters provided by the processor to the game application in real time may include parameters such as picture quality, integer setting, picture style, anti-aliasing, and shading, and after the game application acquires the recommended game configuration parameters, the game application may display the recommended game configuration parameters in the configuration interface 110, for example, the recommended picture quality is high, the recommended frame number is set to be high, the recommended picture style is classic, and the like, so that the user may select game configuration parameters to be actually adopted according to the recommended game configuration parameters and by combining actual needs of the user.
In addition, the recommended terminal configuration parameters provided by the processor to the operating system in real time may include parameters such as a CPU gear, a GPU gear, a game scene holding time, a game scene importance level, and a maximum frame rate. Illustratively, the processor provides the recommended terminal configuration parameters to the operating system in the following format:
description of fields:
mobile: aiming at the configuration data of the model, wherein the data is the identification value of the model;
scene id: the game is identified according to a predefined scene;
MapID, game identification according to a predefined map;
the number of people on the same screen: the number of players on the same screen at the current moment of the game;
limited frame rate: the highest level that the game-limited FPS (Frames Per Second) can reach;
80: desired CPU frequency level, -1 means ignore;
81: desired GPU frequency level, -1 represents ignore;
82, representing the frame rate level the game is expected to reach in this scenario;
83: representing scene importance levels of the game, 1, 2, 3, 4 and 5, wherein the larger the number, the higher the level;
84: the scene keeping time of the game is represented, and the unit is wonderful;
configuration meaning: and (3) detecting 4 variables of scene id, MapID, number of people on screen and target frame rate when the game runs and the 4 variables of the user model change at a certain moment, and if a combined and matched comprehensive scene 'scene id | MapID | number on screen | target frame rate' exists in JSON (JavaScript Object Notation) configuration, sending the corresponding JSON content to an operating system. (if the game does not issue the number of people on the same screen, the number of people on the same screen field is set to be 1 by default, and if the game does not issue the MapID, the MapID field is set to be 0 by default.)
After receiving the recommended terminal configuration parameters, the operating system may select the actually adopted terminal configuration parameters according to the recommended terminal configuration parameters and in combination with its own policy.
The following are embodiments of the apparatus of the present application that may be used to perform embodiments of the method of the present application. For details which are not disclosed in the embodiments of the apparatus of the present application, reference is made to the embodiments of the method of the present application.
Referring to fig. 12, a block diagram of a device for recommending configuration parameters according to an embodiment of the present application is shown. The device has the functions of realizing the method examples, and the functions can be realized by hardware or by hardware executing corresponding software. The device may be a terminal or may be provided in a terminal. The apparatus 120 may include: a data acquisition module 121, a parameter determination module 122 and a parameter transmission module 123.
The data obtaining module 121 is configured to obtain the game configuration parameters, the terminal configuration parameters, and the game running data in real time during the process of running the game application program by the terminal.
And the parameter determining module 122 is configured to determine a recommended game configuration parameter and a recommended terminal configuration parameter according to the game configuration parameter, the terminal configuration parameter, and the game running data through a configuration recommendation model.
A parameter sending module 123, configured to send the recommended game configuration parameter to the game application, and send the recommended terminal configuration parameter to an operating system.
In an exemplary embodiment, the configuration recommendation model includes: a game configuration recommendation model and a terminal configuration recommendation model;
as shown in fig. 13, the parameter determining module 122 includes:
a first determining unit 1221, configured to generate, by using the game configuration recommendation model, a recommended game configuration parameter in a first target time period according to the terminal configuration parameter and the game running data in a first historical time period;
a second determining unit 1222, configured to generate, by the terminal configuration recommendation model, recommended terminal configuration parameters in a second target time period according to the game configuration parameters and the game running data in a second historical time period.
In an exemplary embodiment, as shown in fig. 13, the parameter determining module 122 further includes:
a first updating unit 1223, configured to obtain an actual game configuration parameter in the first target time period; and adjusting parameters of the game configuration recommendation model according to the terminal configuration parameters and the game running data in the first historical time period and the actual game configuration parameters in the first target time period.
In an exemplary embodiment, as shown in fig. 13, the parameter determining module 122 further includes:
a second updating unit 1224, configured to obtain actual terminal configuration parameters in the second target time period; and adjusting parameters of the terminal configuration recommendation model according to the game configuration parameters and the game running data in the second historical time period and the actual terminal configuration parameters in the second target time period.
In an exemplary embodiment, the training data of the game configuration recommendation model includes a first input feature and a first label;
wherein the first input feature comprises at least one of: CPU gear, GPU gear, game scene holding time, game scene importance level, maximum frame rate, scene identification, target frame rate, current frame rate, map identification, number of people in the same screen of the game, memory utilization rate, CPU utilization rate and terminal temperature;
the first tag includes at least one of: picture quality, frame number settings, picture style, anti-aliasing, shading.
In an exemplary embodiment, the training data of the terminal configuration recommendation model comprises a second input feature and a second label;
wherein the second input features include at least one of: picture quality, frame number setting, picture style, anti-aliasing, shadow, scene identification, target frame rate, current frame rate, map identification, number of people on the same screen of the game, memory utilization rate, CPU utilization rate and terminal temperature;
the second label includes at least one of: CPU gear, GPU gear, game scene holding time, game scene importance level and maximum frame rate.
In an exemplary embodiment, as shown in fig. 13, the apparatus 120 further includes:
a data reporting module 124, configured to send a data reporting request to a server according to a reporting period, where the data reporting request includes game configuration parameters, terminal configuration parameters, and game running data acquired in the reporting period; the server is used for generating a training sample, and training the configuration recommendation model by adopting the training sample;
a model obtaining module 125, configured to obtain a latest configuration recommendation model from the server according to a synchronization cycle; wherein the latest configuration recommendation model is used for recommending configuration parameters in real time.
To sum up, in the technical scheme provided by the embodiment of the application, the game configuration parameters, the terminal configuration parameters and the game running data are obtained in real time in the process of running the game application program by the terminal, the recommended game configuration parameters and the recommended terminal configuration parameters are determined according to the data by the configuration recommendation model, then the recommended game configuration parameters are sent to the game application program, and the recommended terminal configuration parameters are sent to the operating system; the game configuration parameters and the terminal configuration parameters are dynamically recommended in real time in the game running process, and compared with the method for statically determining the configuration parameters in the related technology, the technical scheme provided by the embodiment of the application can dynamically recommend the configuration parameters in real time, so that the flexibility and the accuracy of recommending the configuration parameters are improved.
It should be noted that, when the apparatus provided in the foregoing embodiment implements the functions thereof, only the division of the functional modules is illustrated, and in practical applications, the functions may be distributed by different functional modules according to needs, that is, the internal structure of the apparatus may be divided into different functional modules to implement all or part of the functions described above. In addition, the apparatus and method embodiments provided by the above embodiments belong to the same concept, and specific implementation processes thereof are described in the method embodiments for details, which are not described herein again.
Referring to fig. 14, a block diagram of a terminal according to an embodiment of the present application is shown. In general, terminal 1400 includes: a processor 1401, and a memory 1402.
In some embodiments, terminal 1400 may further optionally include: a peripheral device interface 1403 and at least one peripheral device. The processor 1401, the memory 1402, and the peripheral device interface 1403 may be connected by buses or signal lines. Each peripheral device may be connected to the peripheral device interface 1403 via a bus, signal line, or circuit board. Specifically, the peripheral device may include: at least one of radio frequency circuitry 1404, a display 1405, a camera assembly 1406, audio circuitry 1407, a positioning assembly 1408, and a power supply 1409.
Those skilled in the art will appreciate that the configuration shown in fig. 14 is not intended to be limiting with respect to terminal 1400 and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components may be employed.
In an exemplary embodiment, there is also provided a computer-readable storage medium having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions, which when executed by a processor of a terminal, implements the above-described recommendation method for configuration parameters.
Optionally, the computer-readable storage medium may include: a Read Only Memory (ROM), a Random Access Memory (RAM), a Solid State Drive (SSD), or an optical disc. The Random Access Memory may include a resistive Random Access Memory (ReRAM) and a Dynamic Random Access Memory (DRAM).
In an exemplary embodiment, a computer program product is also provided, which when executed by a processor of a terminal, is configured to implement the above recommendation method for configuration parameters.
It should be understood that reference to "a plurality" herein means two or more. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. In addition, the step numbers described herein only exemplarily show one possible execution sequence among the steps, and in some other embodiments, the steps may also be executed out of the numbering sequence, for example, two steps with different numbers are executed simultaneously, or two steps with different numbers are executed in a reverse order to the order shown in the figure, which is not limited by the embodiment of the present application.
The above description is only exemplary of the present application and should not be taken as limiting the present application, and any modifications, equivalents, improvements and the like that are made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims (14)
1. A method for recommending configuration parameters, the method comprising:
in the process of running a game application program by a terminal, obtaining game configuration parameters, terminal configuration parameters and game running data in real time;
determining a recommended game configuration parameter and a recommended terminal configuration parameter according to the game configuration parameter, the terminal configuration parameter and the game running data through a configuration recommendation model;
and sending the recommended game configuration parameters to the game application program, and sending the recommended terminal configuration parameters to an operating system.
2. The method of claim 1, wherein configuring the recommendation model comprises: a game configuration recommendation model and a terminal configuration recommendation model;
the step of determining the recommended game configuration parameters and the recommended terminal configuration parameters according to the game configuration parameters, the terminal configuration parameters and the game running data through the configuration recommendation model comprises the following steps:
generating recommended game configuration parameters in a first target time period according to the terminal configuration parameters and the game running data in the first historical time period through the game configuration recommendation model;
and generating the recommended terminal configuration parameters in the second target time period according to the game configuration parameters and the game running data in the second historical time period through the terminal configuration recommendation model.
3. The method of claim 2, wherein after generating, by the game configuration recommendation model, the recommended game configuration parameters for the first target time period according to the terminal configuration parameters and the game running data in the first historical time period, the method further comprises:
acquiring actual game configuration parameters in the first target time period;
and adjusting parameters of the game configuration recommendation model according to the terminal configuration parameters and the game running data in the first historical time period and the actual game configuration parameters in the first target time period.
4. The method of claim 2, wherein after generating, by the terminal configuration recommendation model, the recommended terminal configuration parameters for the second target time period according to the game configuration parameters and the game running data in the second historical time period, the method further comprises:
acquiring actual terminal configuration parameters in the second target time period;
and adjusting parameters of the terminal configuration recommendation model according to the game configuration parameters and the game running data in the second historical time period and the actual terminal configuration parameters in the second target time period.
5. The method of claim 2, wherein the training data of the game configuration recommendation model comprises a first input feature and a first label;
wherein the first input feature comprises at least one of: CPU (central processing unit) gear, GPU (graphics processing unit) gear, game scene holding time, game scene importance level, maximum frame rate, scene identification, target frame rate, current frame rate, map identification, number of people on the same screen of a game, memory utilization rate, CPU utilization rate and terminal temperature;
the first tag includes at least one of: picture quality, frame number settings, picture style, anti-aliasing, shading.
6. The method of claim 2, wherein the training data of the terminal configuration recommendation model comprises a second input feature and a second label;
wherein the second input features include at least one of: picture quality, frame number setting, picture style, anti-aliasing, shadow, scene identification, target frame rate, current frame rate, map identification, number of people on the same screen of the game, memory utilization rate, CPU utilization rate and terminal temperature;
the second label includes at least one of: CPU gear, GPU gear, game scene holding time, game scene importance level and maximum frame rate.
7. The method according to any one of claims 1 to 6, further comprising:
sending a data reporting request to a server according to a reporting period, wherein the data reporting request comprises game configuration parameters, terminal configuration parameters and game running data acquired in the reporting period; the server is used for generating a training sample, and training the configuration recommendation model by adopting the training sample;
acquiring a latest configuration recommendation model from the server according to a synchronization period;
wherein the latest configuration recommendation model is used for recommending configuration parameters in real time.
8. An apparatus for recommending configuration parameters, the apparatus comprising:
the data acquisition module is used for acquiring game configuration parameters, terminal configuration parameters and game running data in real time in the process of running the game application program by the terminal;
the parameter determination module is used for determining recommended game configuration parameters and recommended terminal configuration parameters according to the game configuration parameters, the terminal configuration parameters and the game running data through a configuration recommendation model;
and the parameter sending module is used for sending the recommended game configuration parameters to the game application program and sending the recommended terminal configuration parameters to an operating system.
9. The apparatus of claim 8, wherein the configuration recommendation model comprises: a game configuration recommendation model and a terminal configuration recommendation model;
the parameter determination module comprises:
the first determining unit is used for generating recommended game configuration parameters in a first target time period according to the terminal configuration parameters and the game running data in a first historical time period through the game configuration recommendation model;
and the second determining unit is used for generating the recommended terminal configuration parameters in the second target time period according to the game configuration parameters and the game running data in the second historical time period through the terminal configuration recommendation model.
10. The apparatus of claim 9, wherein the parameter determination module further comprises:
the first updating unit is used for acquiring actual game configuration parameters in the first target time period; and adjusting parameters of the game configuration recommendation model according to the terminal configuration parameters and the game running data in the first historical time period and the actual game configuration parameters in the first target time period.
11. The apparatus of claim 9, wherein the parameter determination module further comprises:
the second updating unit is used for acquiring the actual terminal configuration parameters in the second target time interval; and adjusting parameters of the terminal configuration recommendation model according to the game configuration parameters and the game running data in the second historical time period and the actual terminal configuration parameters in the second target time period.
12. The apparatus of any one of claims 8 to 11, further comprising:
the data reporting module is used for sending a data reporting request to a server according to a reporting period, wherein the data reporting request comprises game configuration parameters, terminal configuration parameters and game running data acquired in the reporting period; the server is used for generating a training sample, and training the configuration recommendation model by adopting the training sample;
the model acquisition module is used for acquiring the latest configuration recommendation model from the server according to the synchronization period;
wherein the latest configuration recommendation model is used for recommending configuration parameters in real time.
13. A terminal, characterized in that it comprises a processor and a memory in which at least one instruction, at least one program, set of codes or set of instructions is stored, which is loaded and executed by the processor to implement the method according to any one of claims 1 to 7.
14. A computer readable storage medium having stored therein at least one instruction, at least one program, a set of codes, or a set of instructions, which is loaded and executed by a processor to implement the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911415084.4A CN111176758B (en) | 2019-12-31 | 2019-12-31 | Configuration parameter recommendation method and device, terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911415084.4A CN111176758B (en) | 2019-12-31 | 2019-12-31 | Configuration parameter recommendation method and device, terminal and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111176758A true CN111176758A (en) | 2020-05-19 |
| CN111176758B CN111176758B (en) | 2021-06-22 |
Family
ID=70647483
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911415084.4A Active CN111176758B (en) | 2019-12-31 | 2019-12-31 | Configuration parameter recommendation method and device, terminal and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111176758B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112316437A (en) * | 2020-11-23 | 2021-02-05 | 上海米哈游天命科技有限公司 | Resource switching method and device, electronic equipment and storage medium |
| CN113648658A (en) * | 2021-07-23 | 2021-11-16 | 广州三七互娱科技有限公司 | Game data processing method, system, device, computer equipment and storage medium |
| CN113769387A (en) * | 2021-09-18 | 2021-12-10 | 网易（杭州）网络有限公司 | Game graphic parameter configuration method and device and terminal equipment |
| CN115333941A (en) * | 2022-01-28 | 2022-11-11 | 花瓣云科技有限公司 | Method for acquiring application running condition and related equipment |
| WO2022247446A1 (en) * | 2021-05-28 | 2022-12-01 | 华为技术有限公司 | Performance optimization method and related device |
| WO2023142664A1 (en) * | 2022-01-27 | 2023-08-03 | 腾讯科技（深圳）有限公司 | Image display method, terminal device, server, electronic device, storage medium, and program product |
| WO2024036941A1 (en) * | 2022-08-17 | 2024-02-22 | 华为云计算技术有限公司 | Parameter management system and related method |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150126286A1 (en) * | 2013-09-27 | 2015-05-07 | Tencent Technology (Shenzhen) Company Limited | Method And System For Implementing Artificial Intelligence |
| CN104991775A (en) * | 2015-07-03 | 2015-10-21 | 北京奇虎科技有限公司 | Game acceleration method and apparatus |
| CN105677767A (en) * | 2015-12-30 | 2016-06-15 | 联想(北京)有限公司 | Equipment configuration recommending method and device |
| CN108211352A (en) * | 2017-12-28 | 2018-06-29 | 深圳市创梦天地科技股份有限公司 | A kind of method and terminal for adjusting image quality |
| CN108499108A (en) * | 2017-02-28 | 2018-09-07 | 电子技术公司 | The real-time dynamic that game parameter is played in video game application programs is changed and optimization |
| CN109754060A (en) * | 2017-11-06 | 2019-05-14 | 阿里巴巴集团控股有限公司 | A kind of training method and device of neural network machine learning model |
| CN109814933A (en) * | 2019-01-29 | 2019-05-28 | 腾讯科技（深圳）有限公司 | A kind of business data processing method and device |
| CN109831585A (en) * | 2019-02-22 | 2019-05-31 | 维沃移动通信有限公司 | A kind of operating parameter method of adjustment and mobile terminal |
| CN110531951A (en) * | 2019-08-30 | 2019-12-03 | 腾讯科技（深圳）有限公司 | Picture output mode method of adjustment, device, equipment, system and storage medium |
-
2019
- 2019-12-31 CN CN201911415084.4A patent/CN111176758B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150126286A1 (en) * | 2013-09-27 | 2015-05-07 | Tencent Technology (Shenzhen) Company Limited | Method And System For Implementing Artificial Intelligence |
| CN104991775A (en) * | 2015-07-03 | 2015-10-21 | 北京奇虎科技有限公司 | Game acceleration method and apparatus |
| CN105677767A (en) * | 2015-12-30 | 2016-06-15 | 联想(北京)有限公司 | Equipment configuration recommending method and device |
| CN108499108A (en) * | 2017-02-28 | 2018-09-07 | 电子技术公司 | The real-time dynamic that game parameter is played in video game application programs is changed and optimization |
| CN109754060A (en) * | 2017-11-06 | 2019-05-14 | 阿里巴巴集团控股有限公司 | A kind of training method and device of neural network machine learning model |
| CN108211352A (en) * | 2017-12-28 | 2018-06-29 | 深圳市创梦天地科技股份有限公司 | A kind of method and terminal for adjusting image quality |
| CN109814933A (en) * | 2019-01-29 | 2019-05-28 | 腾讯科技（深圳）有限公司 | A kind of business data processing method and device |
| CN109831585A (en) * | 2019-02-22 | 2019-05-31 | 维沃移动通信有限公司 | A kind of operating parameter method of adjustment and mobile terminal |
| CN110531951A (en) * | 2019-08-30 | 2019-12-03 | 腾讯科技（深圳）有限公司 | Picture output mode method of adjustment, device, equipment, system and storage medium |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112316437A (en) * | 2020-11-23 | 2021-02-05 | 上海米哈游天命科技有限公司 | Resource switching method and device, electronic equipment and storage medium |
| CN112316437B (en) * | 2020-11-23 | 2023-02-03 | 上海米哈游天命科技有限公司 | Resource switching method and device, electronic equipment and storage medium |
| WO2022247446A1 (en) * | 2021-05-28 | 2022-12-01 | 华为技术有限公司 | Performance optimization method and related device |
| CN113648658A (en) * | 2021-07-23 | 2021-11-16 | 广州三七互娱科技有限公司 | Game data processing method, system, device, computer equipment and storage medium |
| CN113648658B (en) * | 2021-07-23 | 2023-10-20 | 广州三七互娱科技有限公司 | Game data processing method, system, device, computer equipment and storage medium |
| CN113769387A (en) * | 2021-09-18 | 2021-12-10 | 网易（杭州）网络有限公司 | Game graphic parameter configuration method and device and terminal equipment |
| WO2023142664A1 (en) * | 2022-01-27 | 2023-08-03 | 腾讯科技（深圳）有限公司 | Image display method, terminal device, server, electronic device, storage medium, and program product |
| CN115333941A (en) * | 2022-01-28 | 2022-11-11 | 花瓣云科技有限公司 | Method for acquiring application running condition and related equipment |
| CN115333941B (en) * | 2022-01-28 | 2023-08-22 | 花瓣云科技有限公司 | Method for acquiring application running condition and related equipment |
| WO2024036941A1 (en) * | 2022-08-17 | 2024-02-22 | 华为云计算技术有限公司 | Parameter management system and related method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111176758B (en) | 2021-06-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111176758B (en) | Configuration parameter recommendation method and device, terminal and storage medium | |
| Jaafra et al. | Reinforcement learning for neural architecture search: A review | |
| CN111522962B (en) | Sequence recommendation method, device and computer readable storage medium | |
| CN110168578B (en) | Multi-tasking neural network with task-specific paths | |
| JP7264376B2 (en) | How to generate a general-purpose trained model | |
| Schölkopf | Learning to see and act | |
| CN110298443A (en) | Neural network computing device and method | |
| CN103279958B (en) | A kind of image partition method based on Spiking neural network | |
| Çatal et al. | Learning generative state space models for active inference | |
| CN106953862A (en) | The cognitive method and device and sensor model training method and device of network safety situation | |
| Jeon et al. | Artificial intelligence for traffic signal control based solely on video images | |
| JPH06509669A (en) | improved neural network | |
| CN109886422A (en) | Model configuration method, device, electronic equipment and read/write memory medium | |
| CN111178507A (en) | Atlas convolution neural network data processing method and device | |
| CN112329948A (en) | Multi-agent strategy prediction method and device | |
| CN109344969B (en) | Neural network system, training method thereof, and computer-readable medium | |
| CN109447096A (en) | A kind of pan path prediction technique and device based on machine learning | |
| CN112508190A (en) | Method, device and equipment for processing structured sparse parameters and storage medium | |
| TW201737202A (en) | Method and device for training model of quasi-Alexnet | |
| CN112365359A (en) | Training method, device, equipment and storage medium for planting decision determination model | |
| Nguyen et al. | A prioritized objective actor-critic method for deep reinforcement learning | |
| CN116128060A (en) | Chess game method based on opponent modeling and Monte Carlo reinforcement learning | |
| CN108985449A (en) | A kind of control method and device of pair of convolutional neural networks processor | |
| CN115168720A (en) | Content interaction prediction method and related equipment | |
| CN108229640A (en) | The method, apparatus and robot of emotion expression service |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |