CN112488563B - A method and device for determining computing power parameters - Google Patents

Abstract

The application discloses a method and a device for determining a calculation force parameter, which relate to the field of network communication and are used for improving the efficiency of determining the calculation force parameter. The determining method comprises the following steps: acquiring a calculation service request of a user; determining the parameter type of the power calculation service request according to the power calculation service request; determining a computational power parameter based on the computational power service request and the current neural network model; the current neural network model corresponds to the type of parameters of the power calculation service request. Compared with the method for manually determining the calculation force parameters corresponding to the service requests in the prior art, the method for determining the calculation force parameters is low in efficiency and needs a certain professional ability of a user, and the method for determining the calculation force parameters is based on the neural network model, so that the user does not need a certain professional ability, and the efficiency of determining the calculation force parameters can be improved.

Description

A method and device for determining computing power parameters

Technical Field

The present application relates to the field of network communications, and in particular to a method and device for determining computing power parameters.

Background Art

In recent years, artificial intelligence (AI) has become a universal technology in contemporary society and an indispensable technology for the future intelligent society. Computing power, algorithms, and data are the three elements of AI. Computing power, as the foundation of AI, directly affects the application and deployment of AI services. The currently emerging AI industry and potential intelligent services have put forward high requirements for computing power.

Different business requests correspond to different computing parameters, and the content and form of business requests vary. Therefore, when the description of the business request is relatively abstract, the user needs to manually analyze the content of the business request to determine the computing parameters corresponding to the business request. However, this method of manually determining the computing parameters corresponding to the business request requires the user to have certain professional capabilities and is inefficient.

Summary of the invention

The present application provides a method and device for determining computing power parameters, which are used to improve the efficiency of determining computing power parameters.

In order to achieve the above purpose, this application adopts the following technical solutions:

In a first aspect, the present application provides a method for determining a computing power parameter. The determination method includes: obtaining a computing power service request of a user. Then, according to the computing power service request, determining the type of the computing power service request, and determining a target computing power parameter based on the computing power service request and a current neural network model corresponding to the type of the computing power service request.

The method for determining computing power parameters provided in the present application obtains the computing power service request of the user and determines the computing power parameters based on the computing power service request and the current neural network model. Compared with the method of manually determining the computing power parameters corresponding to the service request in the prior art, which requires the user to have certain professional capabilities and is inefficient, the method for determining the computing power parameters provided in the present application is based on the neural network model to determine the computing power parameters, thereby not requiring the user to have certain professional capabilities, and thus being able to improve the efficiency of determining the computing power parameters.

In the second aspect, the present application provides a device for determining computing power parameters. The device includes: an acquisition unit, which acquires a computing power service request of a user; a determination unit, which is used to determine the type of the computing power service request according to the computing power service request acquired by the acquisition unit; the determination unit is also used to determine the computing power parameters based on the computing power service request acquired by the acquisition unit and the current neural network model; the current neural network model corresponds to the type of the computing power service request.

In a third aspect, the present application provides a device for determining a computing power parameter, the device for determining a computing power parameter comprising a memory and a processor. The memory and the processor are coupled. The memory is used to store a computer program code, the computer program code comprising a computer instruction. When the processor executes the computer instruction, the device for determining a computing power parameter performs the method for determining a computing power parameter as described in the first aspect and any possible design thereof.

In a fourth aspect, the present application provides a computer-readable storage medium, which stores instructions. When the computer-readable storage medium is run on a device for determining computing power parameters, the device executes the method for determining computing power parameters as described in the first aspect and any possible design method thereof.

In a fifth aspect, the present application provides a computer program product, which includes computer instructions. When the computer instructions are run on a device for determining computing power parameters, the device for determining computing power parameters executes the method for determining computing power parameters as described in the first aspect and any possible design method thereof.

For the specific description of the second to fifth aspects and their various implementations in this application, reference can be made to the detailed description in the first aspect and its various implementations; and for the beneficial effects of the second to fifth aspects and their various implementations, reference can be made to the beneficial effects analysis in the first aspect and its various implementations, which will not be repeated here.

These and other aspects of the present application will become more apparent from the following description.

BRIEF DESCRIPTION OF THE DRAWINGS

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 paying any creative work.

FIG1 is a flow chart of a method for determining a computing power parameter provided by the present application;

FIG2 is a schematic diagram of the structure of a first neural network expert system provided by the present application;

FIG3 is a schematic diagram of the structure of a second neural network expert system provided by the present application;

FIG4 is a second flow chart of a method for determining a computing power parameter provided by the present application;

FIG5 is a schematic diagram of the hardware structure of a device for determining computing power parameters provided in an embodiment of the present application;

FIG6 is a schematic diagram of the structure of a device for determining computing power parameters provided in an embodiment of the present application.

DETAILED DESCRIPTION

The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

In recent years, artificial intelligence (AI) has become a universal technology in contemporary society and an indispensable technology for the future intelligent society. Computing power, algorithms, and data are the three elements of AI. Computing power, as the foundation of AI, directly affects the application and deployment of AI services. The currently emerging AI industry and potential intelligent services have put forward high requirements for computing power.

Different business requests correspond to different computing parameters, and the content and form of business requests vary. Therefore, when the description of the business request is relatively abstract, the user needs to manually analyze the content of the business request to determine the computing parameters corresponding to the business request. However, this method of manually determining the computing parameters corresponding to the business request requires the user to have certain professional capabilities and is inefficient.

In response to the above problems, the present application provides a method for determining computing power parameters, which obtains the computing power service request of the user and determines the computing power parameters based on the computing power service request and the current neural network model. Compared with the method of manually determining the computing power parameters corresponding to the service request in the prior art, which requires the user to have certain professional capabilities and is inefficient, the method for determining the computing power parameters provided by the present application is based on the neural network model to determine the computing power parameters, thereby eliminating the need for the user to have certain professional capabilities, and thus improving the efficiency of determining the computing power parameters.

The execution subject of the method for determining computing power parameters provided in the embodiment of the present application is a computing power parameter determination device (hereinafter referred to as: determination device). The determination device can be integrated with the computing power network system or can be independently set, and the embodiment of the present application does not limit this.

The computing power network system is a network system that can publish the current computing power status and network status as routing information to the network. The network routes the computing task message to the corresponding computing node to achieve the best user experience, optimal computing resource utilization, and optimal network efficiency. Through the built-in computing task dynamic routing capability of the computing power network system, computing tasks can be dynamically and flexibly scheduled according to business needs based on real-time computing resource performance, network performance, cost and other multi-dimensional factors, thereby improving resource utilization, network utilization efficiency, and business user experience. For edge computing scenarios, the computing power network system can be used to realize edge computing networking and edge-to-edge collaboration. The multi-instance and multi-copy characteristics of the service can be used to achieve user access nearby and service load balancing to solve problems such as complex deployment, low efficiency, and low resource reuse rate, and help scale deployment of edge computing.

The following describes a method for determining computing power parameters provided in an embodiment of the present application.

As shown in FIG1 , the method for determining the computing power parameter includes:

S101. A determination device obtains a computing power service request from a user.

Optionally, the determination device may receive a computing power service request input by a user.

Optionally, the computing power service request may include rendering service, training task, inference service, video playback, supercomputing or other services.

The computing power service request includes computing power parameters, user business requirement parameters, or user abstract requirement parameters.

The computing power parameters are the parameters corresponding to the underlying hardware resources required by the computing power network system to meet the computing power business.

Optionally, the computing power parameters may include parameters of a central processing unit (CPU), parameters of a graphics processing unit (GPU), storage parameters, parameters of network indicators, etc.

The user's business demand parameters are the user's specific demands for computing power services.

Exemplarily, the user's business requirement parameters may include the user's business requirement parameters, model image requirement parameters, algorithm requirement parameters, time requirement parameters, and running times requirement parameters.

The user's abstract demand parameter is an abstract description of the user's request for computing power service.

Exemplarily, the user's abstract requirement parameter is to train a neural network model for face recognition, or the user's abstract requirement parameter is to perform subsequent rendering on a movie.

S102. The determining device determines the type of the computing power service request according to the computing power service request.

The types of computing power service requests include computing power parameter types, service requirement types, or abstract requirement types.

Optionally, the determining device may determine the type of the computing power service request based on keywords included in the computing power service request.

Keywords include computing power parameter keywords, business requirement keywords, and abstract requirement keywords.

Optional, computing power parameter keywords include CPU, GPU, tensor operation, storage, codec, and network indicators.

For example, as shown in Table 1, when the computing power service request includes the keyword CPU, the parameter 10 tera operations per second (TOPS) of the CPU is determined as the computing power parameter; when the computing power service request includes the keyword GPU, the parameter 12 (floating point operations per second, TFLOPs) of the GPU is determined as the computing power parameter; when the computing power service request includes the keyword tensor operation, the parameter 8 TFLOPS of the tensor operation embedded neural network processor (neural-network processing unit, NPU)/tensor processing unit (tensor processing unit, TPU) is determined as the computing power parameter; when the computing power service request includes the keyword storage, the parameter 10 gigabytes of storage is determined as the computing power parameter; when the computing power service request includes the keyword network indicator, the parameter delay of the network indicator is less than 50 milliseconds (millisecond, ms) is determined as the computing power parameter; when the computing power service request includes the keyword codec, the codec is determined to meet the standard as the computing power parameter.

Table 1

If the number of computing power parameters included in the computing power service request is greater than or equal to a first preset threshold, it is determined that the type of the computing power service request is a computing power parameter type.

The first preset threshold can be determined according to actual conditions, and this application does not limit this.

Further, if the number of key computing power parameters in the computing power service request is greater than or equal to the second preset threshold, the type of the computing power service request is determined to be a computing power parameter type.

The second preset threshold can be determined according to actual conditions, and this application does not limit this.

Key computing power parameters are the computing power parameters that are necessary to meet the computing power business.

Exemplarily, the key computing power parameters may be CPU parameters, GPU parameters, storage parameters, and network indicator parameters.

By determining that the type of the computing power service request is a computing power parameter type through key computing power parameters, the accuracy of determining that the type of the computing power service request is a computing power parameter type can be improved.

Optionally, business requirement keywords include business, model image, algorithm, time, number of runs, and data volume.

For example, as shown in Table 2, when it is determined that the computing power service request includes the keyword business, the demand parameter for the service, video rendering or model training, is determined as the business demand parameter; when it is determined that the computing power service request includes the keyword model image, the demand parameter for the model image, ray tracing and global light rendering KeyShot or residual network ResNet, is determined as the business demand parameter; when it is determined that the computing power service request includes the keyword algorithm, the demand parameter for the algorithm, resolution 1080 progressive scanning (progressive scanning, p) or 152-layer network and 6,000 parameters, are determined as the business demand parameters; when it is determined that the computing power service request includes the keyword time, the demand parameter for time, one month or one week, is determined as the business demand parameter; when it is determined that the computing power service request includes the keyword number of runs, the demand parameter for the number of runs, one time or one hundred thousand times, is determined as the business demand parameter; when it is determined that the computing power service request includes the keyword data volume, the demand parameter for the data volume, 10G video or 10G video, is determined as the business demand parameter.

Table 2

When it is determined that the number of service requirement parameters included in the computing power service request is greater than or equal to the third preset threshold, the type of the computing power service request is determined to be a service requirement type.

The third preset threshold can be determined according to actual conditions, and this application does not limit this.

Further, if the number of key business requirement parameters in the computing power business request is greater than or equal to a fourth preset threshold, the type of the computing power business request is determined to be a business requirement parameter type.

The fourth preset threshold can be determined based on actual conditions, and this application does not limit this.

Key business requirement parameters are the business requirement parameters in the computing power parameters that are necessary to meet the computing power business.

The key business requirement parameters may be requirement parameters for business, requirement parameters for model images, requirement parameters for algorithms, requirement for time, and requirement for the number of runs.

By determining that the type of the computing power business request is a business requirement parameter type through key business requirement parameters, the accuracy of determining that the type of the computing power business request is a business requirement parameter type can be improved.

When it is determined that the number of computing power parameters included in the computing power service request is less than the first preset threshold, the number of service requirement parameters is less than the second preset threshold, and abstract requirement parameters are included, the type of the computing power service request is determined to be an abstract requirement type.

The abstract requirement parameter is an abstract description of the user's request for computing power service.

For example, the user's computing power service request is to train a neural network model for face recognition. The computing power service request does not include computing power parameters or service requirement parameters, and training a neural network model for face recognition is an abstract requirement parameter. In this case, the type of the computing power service request is an abstract requirement type.

It should be noted that when the computing power service request includes computing power parameters, business demand parameters and abstract demand parameters, and the number of computing power parameters is greater than or equal to the first preset threshold and the number of business demand parameters is greater than or equal to the third preset threshold, the type of the computing power service request can be a computing power parameter type, a business demand type, or an abstract demand type.

Optionally, when the type of the computing power service request of the user is a computing power parameter type, the determining device determines the computing power parameter included in the computing power service request as the target computing power parameter. When the computing power parameter type is a business requirement type or an abstract requirement type, continue to execute S103.

S103. The determination device determines the computing power parameters based on the computing power service request and the current neural network model.

The neural network model includes a first neural network model and a second neural network model.

The first neural network model corresponds to the type of computing power service request being a business demand type. That is, when the type of computing power service request is a business demand type, the current neural network model is the first neural network model.

The first neural network model is used to predict computing power parameters based on business demand parameters.

Optionally, the first neural network model may be a neural network model in a first neural network expert system.

As shown in FIG. 2 , the first neural network expert system 20 includes a first neural network model 21 , a first rule base 22 and a first output interpretation module 23 .

The input of the first neural network expert system 20 is the business demand parameter in the computing power business request, and the output of the first neural network expert system 20 is the computing power parameter.

The first rule base 22 is used to store expert knowledge corresponding to business demand parameters and computing power parameters, and to train the first neural network model 21.

Optionally, the first rule base 22 may store expert knowledge corresponding to computing power parameters and business demand parameters obtained from actual experience. Expert knowledge may be a large number of instance sets from actual operation, or a large number of rules obtained from experts or literature in the field, or a mixture of the two.

Exemplarily, as shown in Table 3, the first rule base 22 includes business demand parameters such as recommended classic AlexNet model configuration, 80w nodes, 6096w parameters, convolution layer parameters 3.8%, fully connected layer parameters 96.2%, recommended storage of 10G, recommended completion time of 1 week, and recommended network configuration as non-real-time, and the corresponding computing power parameters are logical operation of 10TOPS, matrix operation of 12TFLOPS, tensor operation of 8TFLOPS, storage of 10G, and network as non-real-time; the business demand parameters are 8-channel video, clarity of 1080p, storage of 2T, network delay of <200ms, upstream bandwidth of xxMB, downstream bandwidth of xxMb, and encoding and decoding capability of xxx standard, and the corresponding computing power parameters are logical operation of 3TO PS, matrix operation of 5TFLOPS, storage of 25T, network delay of <200ms, upstream bandwidth of xxMB, downstream bandwidth of xxMb, and encoding and decoding of 2 decoding engines and 4 encoding engines.

Table 3

The first output interpretation module 23 is used to translate the predicted result obtained by the first neural network model 21 to obtain the computing power parameter.

Exemplarily, when it is determined that the type of computing power business request is a business demand type, the business demand parameters in the computing power business request are substituted into the first neural network model to obtain a first prediction result, and the first prediction result is interpreted through a first output interpretation mechanism to obtain a computing power parameter.

The second neural network model corresponds to the type of computing power service request being an abstract demand type. That is, when the type of computing power service request is an abstract demand type, the current neural network model is the second neural network model.

The second neural network model is used to predict computing power parameters based on abstract demand parameters.

Optionally, the second neural network model may be a neural network model in a second neural network expert system.

As shown in FIG. 3 , the second neural network expert system 30 includes a second neural network model 31 , a second rule base 32 and a second output interpretation module 33 .

The input of the second neural network expert system 30 is the abstract demand parameter in the computing power service request, and the output of the second neural network expert system 30 is the computing power parameter.

The second rule base 32 is used to store expert knowledge corresponding to abstract demand parameters and computing power parameters, and to train the second neural network model 31.

Optionally, the second rule base 32 can store expert knowledge corresponding to computing power parameters and business demand parameters obtained from actual experience. Expert knowledge can be a large number of instance sets from actual operation, or a large number of rules obtained from experts or literature in the field, or a mixture of the two.

Exemplarily, as shown in Table 4, the second rule base 32 includes abstract requirement parameters for training a network model for image recognition, 8G of training data, and model uncertainty, and the corresponding computing power parameters are 10TOPS for logical operation, 12TFLOPS for matrix operation, 8TFLOPS for tensor operation, 10G for storage, and non-real-time network; the abstract requirement parameters are the deployment of cloud augmented reality (AR)/virtual reality (VR) games, 200G of all game content, and the corresponding computing power parameters are 3TOPS for logical operation, 5TFLOPS for matrix operation, 25T for storage, network with delay <200ms, xxMB for uplink bandwidth, xxMb for downlink bandwidth, and 2 decoding engines and 4 encoding engines for encoding and decoding.

Table 4

The second output interpretation module 33 is used to translate the predicted results obtained by the second neural network model 31 to obtain computing power parameters.

Exemplarily, when it is determined that the type of computing power service request is an abstract demand type, the abstract demand parameters in the computing power service request are substituted into the second neural network model to obtain a second prediction result, and the second prediction result is interpreted through a second output interpretation mechanism to obtain the computing power parameters.

Optionally, in combination with FIG1 , as shown in FIG4 , after S103 , the method for determining the computing power parameter further includes:

S104: The determining device determines that the difference between the computing power parameter and the target computing power parameter is less than or equal to a preset threshold.

The target computing power parameters are the computing power parameters that meet the computing power service request.

Optionally, after the determining device determines the computing power parameter, the computing power parameter can be compared with the target computing power parameter. When the difference between the computing power parameter and the target computing power parameter is less than or equal to a preset threshold, it is determined that the computing power parameter meets the user's computing power service request.

S105. Optimize the current neural network model based on the computing power parameters and computing power business request.

A feasible way is to optimize the current neural network model based on the computing power parameters and the business demand parameters in the computing power business request.

Exemplarily, when it is determined that the type of computing power business request is a business demand type, the computing power parameters are determined based on the business demand parameters in the computing power business request and the first neural network model, and when the difference between the computing power parameters and the target computing power parameters is less than or equal to a preset threshold, the first rule base is updated based on the determined computing power parameters and the business demand parameters in the computing power business request.

The first neural network model is trained using the updated first rule base to obtain a trained first neural network model.

For example, as shown in Table 5, the updated first rule base includes the user's abstract requirement parameters of 8G training data for training an image recognition network model, uncertain model, and business requirement parameters of recommended classic AlexNet model configuration, 80w nodes, 6096w parameters, convolution layer parameters 3.8%, fully connected layer parameters 96.2%, recommended storage of 10G, recommended completion time of 1 week, and recommended network configuration as non-real-time. The corresponding computing power parameters are 10TOPS for logical operations, 12TFLOPS for matrix operations, 8TFLOPS for tensor operations, 10G for storage, and non-real-time network, and the computing power parameters meet the user's needs.

Table 5

The next time the determination device determines the computing power parameters based on the business demand parameters in the computing power business request and the current neural network model, the first trained neural network model is the current neural network model.

Another feasible approach is to optimize the current neural network model based on the computing power parameters and the abstract demand parameters in the computing power business request.

Exemplarily, when it is determined that the type of the computing power service request is an abstract demand type, the computing power parameters are determined based on the abstract demand parameters in the computing power service request and the second neural network model.

If the difference between the computing power parameter and the target computing power parameter is less than or equal to the preset threshold, the second rule base is updated according to the determined computing power parameter and the abstract demand parameter in the computing power service request.

The second neural network model is trained using the updated second rule base to obtain a trained second neural network model.

The next time the determination device determines the computing power parameters based on the abstract demand parameters in the computing power service request and the current neural network model, the trained second neural network model is the current neural network model.

The method for determining computing power parameters provided in the present application obtains the computing power service request of the user, and determines the computing power parameters based on the computing power service request and the current neural network model. Compared with the method of manually determining the computing power parameters corresponding to the service request in the prior art, which requires the user to have certain professional capabilities and has low efficiency, the method for determining the computing power parameters provided in the present application is based on a neural network expert system to determine the computing power parameters, so that the user does not need to have certain professional capabilities, thereby improving the efficiency of determining the computing power parameters.

The above mainly introduces the solution provided by the embodiment of the present application from the perspective of the method. In order to achieve the above functions, it includes hardware structures and/or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

As shown in FIG5 , an embodiment of the present application provides a computing power parameter determination device 500. The computing power parameter determination device may include at least one processor 501, a communication line 502, a memory 503, and a communication interface 504.

Specifically, the processor 501 is used to execute the computer-executable instructions stored in the memory 503, so as to implement the steps or actions of the terminal.

The processor 501 may be a chip. For example, it may be a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on chip (SoC), a central processor unit (CPU), a network processor (NP), a digital signal processor (DSP), a microcontroller unit (MCU), a programmable logic device (PLD), or other integrated chips.

The communication line 502 is used to transmit information between the processor 501 and the memory 503 .

The memory 503 is used to store and execute computer-executable instructions, and the execution is controlled by the processor 501 .

The memory 503 may be independent and connected to the processor via the communication line 502. The memory 503 may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), and enhanced synchronous dynamic random access memory (ESDRAM). It should be noted that memory of the systems and devices described herein is intended to comprise, without being limited to, these and any other suitable types of memory.

The communication interface 504 is used to communicate with other devices or communication networks, where the communication network can be Ethernet, a radio access network (RAN), or a wireless local area network (WLAN).

It should be pointed out that the structure shown in FIG5 does not constitute a limitation on the device for determining the computing power parameters. In addition to the components shown in FIG5, the device for determining the computing power parameters may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

As shown in FIG6 , an embodiment of the present application provides a computing power parameter determination device 60. The computing power parameter determination device may include an acquisition unit 61 and a determination unit 62.

The acquisition unit 61 is used to acquire the computing power service request of the user. For example, in conjunction with FIG1 , the acquisition unit 61 can be used to execute S101.

The determining unit 62 is configured to determine the type of the computing power service request obtained by the obtaining unit 61. For example, in conjunction with FIG1 , the determining unit 62 may be configured to execute S102.

The determining unit 62 is further configured to determine the computing power parameter based on the computing power service request and the current neural network model acquired by the acquiring unit 61. For example, in conjunction with FIG1 , the determining unit 62 may be configured to execute S103.

It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

In actual implementation, the acquisition unit 61 and the determination unit 62 can be implemented by the processor 501 shown in Figure 5 calling the program code in the memory 503. The specific execution process can refer to the description of the determination method of the computing power parameter shown in Figure 1, which will not be repeated here.

Another embodiment of the present application further provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed on a device for determining computing power parameters, the device for determining computing power parameters executes each step executed by the device for determining computing power parameters in the method flow shown in the above method embodiment.

In another embodiment of the present application, a computer program product is also provided. The computer program product includes instructions. When the instructions are executed on a device for determining computing power parameters, the device for determining computing power parameters executes each step executed by the device for determining computing power parameters in the method flow shown in the above method embodiment.

Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of 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. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims (10)

1. A method for determining a computing power parameter, characterized in that the method for determining comprises: Obtain the user's computing power service request; According to the computing power service request, determine the type of the computing power service request; the computing power service request includes at least one of the following: computing power parameters, key computing power parameters, business demand parameters, key business demand parameters, abstract requirements parameter; the type of computing power service request includes at least one of the following: computing power parameter type, business requirement type, and abstract requirement type; the key computing power parameter is a computing power parameter satisfying the computing power business; the The key business demand parameters are business demand parameters that meet the computing power business; Wherein, when the number of the computing power parameters is greater than or equal to the first preset threshold; or the number of the key computing power parameters is greater than or equal to the second preset threshold, it is determined that the type of the computing power service request is The type of computing power parameter; When the number of the business demand parameters is greater than or equal to the third preset threshold; or the number of the key business demand parameters is greater than or equal to the fourth preset threshold, it is determined that the type of the computing power business request is the Type of business requirement; When the number of the computing power parameters is less than the first preset threshold, the number of the service demand parameters is less than the second preset threshold, and the computing power service request includes the abstract demand parameters , determining that the type of the computing power service request is the abstract demand type; A computing power parameter is determined based on the computing power service request and the current neural network model; the current neural network model corresponds to the type of the computing power service request. 2. The determination method according to claim 1, characterized in that, when the number of the computing power parameters is greater than or equal to the first preset threshold, the number of the business demand parameters is greater than or equal to the third preset threshold Set a threshold, and when the computing power service request includes the computing power parameter, the business demand parameter, and the abstract demand parameter, determine that the type of the computing power service request is the computing power parameter type , the business requirement type, and any one type of the abstract requirement type. 3. The determination method according to claim 2, wherein the determination of the type of the computing service request comprises: The type of the computing power service request is determined according to the keyword included in the computing power service request. 4. The determination method according to claim 1, wherein the method further comprises: determining that the difference between the computing power parameter and the target computing power parameter is less than or equal to a preset threshold; the target computing power parameter is a computing power parameter that satisfies the computing power service request; Optimizing the current neural network model based on the computing power parameter and the computing power service request. 5. A device for determining a computing power parameter, wherein the device for determining comprises: The obtaining unit is used to obtain the computing power service request of the user; A determining unit, configured to determine the type of the computing power service request according to the computing power service request acquired by the acquiring unit; the computing power service request includes at least one of the following: computing power parameters, key computing power parameters, Business demand parameters, key business demand parameters, and abstract demand parameters; the type of computing power business request includes at least one of the following: computing power parameter type, business demand type, and abstract demand type; the key computing power parameter is to meet the required The computing power parameter of the computing power business; the key business demand parameter is a business demand parameter satisfying the computing power business; Wherein, when the number of the computing power parameters is greater than or equal to the first preset threshold; or the number of the key computing power parameters is greater than or equal to the second preset threshold, it is determined that the type of the computing power service request is The type of computing power parameter; When the number of the business demand parameters is greater than or equal to the third preset threshold; or the number of the key business demand parameters is greater than or equal to the fourth preset threshold, it is determined that the type of the computing power business request is the Type of business requirement; When the number of the computing power parameters is less than the first preset threshold, the number of the service demand parameters is less than the second preset threshold, and the computing power service request includes the abstract demand parameters determining that the type of the computing power service request is the abstract demand type; the determining unit is further configured to determine a computing power parameter based on the computing power service request obtained by the obtaining unit and the current neural network model; The current neural network model corresponds to the type of computing power service request. 6. The determining device according to claim 5, wherein when the number of the computing power parameters is greater than or equal to the first preset threshold, the number of the business demand parameters is greater than or equal to the third preset threshold. Set a threshold, and when the computing power service request includes the computing power parameter, the business demand parameter, and the abstract demand parameter, determine that the type of the computing power service request is the computing power parameter type , the business requirement type, and any one type of the abstract requirement type. 7. The determining device according to claim 6, wherein the determining unit is specifically used for: Determine the type of the computing power service request according to the keywords included in the computing power service request. 8. The determining device according to claim 5, wherein the determining unit is further used for: determining that the difference between the computing power parameter and the target computing power parameter is less than or equal to a preset threshold; the target computing power parameter is a computing power parameter that satisfies the computing power service request; Optimizing the current neural network model based on the computing power parameter and the computing power service request. 9. A device for determining a computing power parameter, characterized in that the device for determining a computing power parameter includes a memory and a processor; the memory is coupled to the processor; the memory is used to store computer program codes, and the The computer program code includes computer instructions; when the processor executes the computer instructions, the computing power parameter determining device executes the computing power parameter determining method according to any one of claims 1-4. 10. A computer-readable storage medium, characterized in that, instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on a computing power parameter determination device, the device executes The method for determining computing power parameters described in any one of claims 1-4.

Images