CN114201246A - Data prediction method and related equipment - Google Patents
Data prediction method and related equipment Download PDFInfo
- Publication number
- CN114201246A CN114201246A CN202210149141.4A CN202210149141A CN114201246A CN 114201246 A CN114201246 A CN 114201246A CN 202210149141 A CN202210149141 A CN 202210149141A CN 114201246 A CN114201246 A CN 114201246A
- Authority
- CN
- China
- Prior art keywords
- prediction model
- prediction
- plug
- data
- model
- 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.)
- Pending
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
- G06F9/448—Execution paradigms, e.g. implementations of programming paradigms
- G06F9/4482—Procedural
-
- 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
- G06F9/4451—User profiles; Roaming
-
- 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
- G06F9/451—Execution arrangements for user interfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
Abstract
The invention provides a data prediction method and related equipment, comprising the following steps: determining a target prediction model plug-in corresponding to a business object to be predicted, wherein a prediction algorithm is arranged in the target prediction model plug-in; starting a parameter configuration interface, binding a data source for parameters in a prediction algorithm in the parameter configuration interface, and generating an initial prediction model; training the initial prediction model to obtain a first prediction model corresponding to the initial prediction model; reading real-time data corresponding to a data source when an instruction for predicting a business object to be predicted is received; and inputting the real-time data into the first prediction model for data prediction, and outputting a prediction result. According to the method, for different business objects to be predicted, an initial prediction model of the business objects to be predicted is built, after training and verification, data prediction is carried out according to current real-time data of the business objects to be predicted, and prediction results are output.
Description
Technical Field
The invention relates to the technical field of data prediction, in particular to a data prediction method and related equipment.
Background
With the development of big data information, data prediction technology is widely applied. In each business field, business data generated by business objects in the business field are analyzed and calculated through corresponding prediction models and prediction algorithms to infer the operation rules of the business objects. And further, the operation trend of the business object can be known as early as possible, and potential risks in the operation process of the business object are avoided, so that the normal operation of the business object is guaranteed.
Data prediction relates to a plurality of business fields, such as prediction of time series data, prediction of faults, prediction of quality, prediction of energy consumption and the like. In the existing data prediction process, aiming at different service fields, the adopted prediction models and prediction methods are different. The data prediction methods in various service fields cannot be used mutually, so that the reusability of the data prediction method is low, and the cost of data prediction is high.
Disclosure of Invention
In view of this, the embodiment of the present invention provides a data prediction method to solve the problems that data prediction methods in the existing service fields cannot be used mutually, and the reusability of the data prediction method is low.
The embodiment of the invention also provides a data prediction device which is used for ensuring the actual realization and application of the method.
In order to achieve the above purpose, the embodiments of the present invention provide the following technical solutions:
a method of data prediction, comprising:
determining a target prediction model plug-in corresponding to a current business object to be predicted, wherein a prediction algorithm is arranged in the target prediction model plug-in;
starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameters in the prediction algorithm in the parameter configuration interface to generate an initial prediction model corresponding to the prediction algorithm;
obtaining a first historical data set, carrying out model training on the initial prediction model, and obtaining a first prediction model corresponding to the initial prediction model;
acquiring a second historical data set, and performing model verification on the first prediction model;
when the first prediction model passes model verification, storing the first prediction model into a preset prediction model storage unit;
when a prediction instruction for performing data prediction on the current business object to be predicted is received, acquiring the first prediction model in the prediction model storage unit, and reading real-time data corresponding to the data source in the current business object to be predicted;
and inputting the real-time data into the first prediction model for data prediction, and outputting a data prediction result corresponding to the current business object to be predicted.
Optionally, the method for determining the target prediction model plug-in corresponding to the current business object to be predicted includes:
determining the object attribute of the current business object to be predicted;
writing a plurality of prediction model plug-ins corresponding to the object attributes, and registering the prediction model plug-ins into a pre-established plug-in storage unit;
and determining a target prediction model plug-in corresponding to the current business object to be predicted in each prediction model plug-in registered in the plug-in storage unit.
Optionally, the starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameter in the prediction algorithm in the parameter configuration interface includes:
calling a set editing interface, and starting a parameter configuration interface corresponding to the target prediction model plug-in;
binding the acquired data source corresponding to each parameter in the prediction algorithm for each parameter in the parameter configuration interface; and each data source corresponds to the current business object to be predicted.
The method described above, optionally, the acquiring a first historical data set includes:
configuring a first historical data interval participating in model training;
and reading each first historical data corresponding to the data source in the first historical data interval, and forming each first historical data into the first historical data set.
The above method, optionally, further includes:
and performing model training on the first prediction model by using the real-time data to obtain a second prediction model corresponding to the first prediction model, and replacing the first prediction model stored in the prediction model storage unit with the second prediction model.
A data prediction apparatus comprising:
the system comprises a determining unit, a predicting unit and a predicting unit, wherein the determining unit is used for determining a target predicting model plug-in corresponding to a current business object to be predicted, and a predicting algorithm is arranged in the target predicting model plug-in;
the configuration unit is used for starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameters in the prediction algorithm in the parameter configuration interface to generate an initial prediction model corresponding to the prediction algorithm;
the training unit is used for acquiring a first historical data set, carrying out model training on the initial prediction model and acquiring a first prediction model corresponding to the initial prediction model;
the verification unit is used for acquiring a second historical data set, carrying out model verification on the first prediction model, and storing the first prediction model into a preset prediction model storage unit when the first prediction model passes the model verification;
the prediction unit is used for acquiring the first prediction model in the prediction model storage unit when a prediction instruction for performing data prediction on the current business object to be predicted is received, and reading real-time data corresponding to the data source in the current business object to be predicted; and inputting the real-time data into the first prediction model for data prediction, and outputting a data prediction result corresponding to the current business object to be predicted.
The above apparatus, optionally, the determining unit includes:
the first determining subunit is used for determining the object attribute of the current business object to be predicted;
the compiling subunit is used for compiling a plurality of prediction model plug-ins corresponding to the object attributes and registering the prediction model plug-ins into a pre-established plug-in storage unit;
and the second determining subunit is configured to determine, in each prediction model plug-in registered in the plug-in storage unit, a target prediction model plug-in corresponding to the current business object to be predicted.
A data prediction apparatus comprising:
the device comprises a prediction model expansion device, a prediction model construction device and a data real-time prediction device;
the prediction model extension device comprises: the system comprises a plug-in development unit, a plug-in registration unit and a plug-in storage unit;
the plug-in development unit is used for developing a plurality of prediction model plug-ins according to the defined standard interface;
the plug-in registration unit is used for registering the plurality of prediction model plug-ins into the plug-in storage unit;
the prediction model construction device comprises: the device comprises a prediction model plug-in selection unit, a prediction model construction unit, a prediction model training unit and a prediction model verification unit;
the prediction model plug-in selection unit is used for determining a target prediction model plug-in the plug-in storage unit, and a prediction algorithm is arranged in the target prediction model plug-in;
the prediction model construction unit is used for starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameters in the prediction algorithm in the parameter configuration interface to generate an initial prediction model corresponding to the prediction algorithm;
the prediction model training unit is used for reading a plurality of first historical data used for model training, applying the plurality of first historical data to carry out model training on the initial prediction model, and obtaining a first prediction model corresponding to the initial prediction model;
the prediction model verification unit is used for reading a plurality of second historical data used for model verification and applying the plurality of second historical data to carry out model verification on the first prediction model;
the data real-time prediction device comprises: the real-time prediction system comprises a real-time data acquisition unit and a real-time prediction unit;
the real-time data acquisition unit is used for reading real-time data corresponding to the data source;
and the real-time prediction unit is used for inputting the real-time data into the first prediction model which passes model verification to perform data prediction and outputting a data prediction result corresponding to the current business object to be predicted.
A storage medium comprising stored instructions, wherein the instructions, when executed, control a device in which the storage medium is located to perform the above-described data prediction method.
An electronic device comprises at least one processor, at least one memory connected with the processor, and a bus; the processor and the memory complete mutual communication through a bus; the processor is used for calling the program instructions in the memory to execute the data prediction method.
Based on the data prediction method provided by the embodiment of the invention, the method comprises the following steps: determining a target prediction model plug-in corresponding to a current business object to be predicted, wherein a prediction algorithm is arranged in the target prediction model plug-in; starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameters in the prediction algorithm in the parameter configuration interface to generate an initial prediction model corresponding to the prediction algorithm; obtaining a first historical data set, carrying out model training on the initial prediction model, and obtaining a first prediction model corresponding to the initial prediction model; acquiring a second historical data set, and performing model verification on the first prediction model; when the first prediction model passes model verification, storing the first prediction model into a preset prediction model storage unit; when a prediction instruction for performing data prediction on the current business object to be predicted is received, acquiring the first prediction model in the prediction model storage unit, and reading real-time data corresponding to the data source in the current business object to be predicted; and inputting the real-time data into the first prediction model for data prediction, and outputting a data prediction result corresponding to the current business object to be predicted. By applying the data prediction method provided by the embodiment of the invention, the initial prediction model corresponding to the business object to be predicted is built in real time for different business objects to be predicted, the first prediction model corresponding to the initial prediction model is determined after the training and verification of the initial prediction model, and when the data prediction is needed to be carried out on the business object to be predicted, the data prediction is carried out in the first prediction model by combining the current real-time data of the business object to be predicted, and the prediction result is output.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart of a method for predicting data according to an embodiment of the present invention;
FIG. 2 is a flowchart of another method of a data prediction method according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a data prediction apparatus according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of another structure of a data prediction apparatus according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a data prediction apparatus according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In this application, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
In the present application, the terms "first" and "second" do not denote an order of arrangement, but merely distinguish between names.
An embodiment of the present invention provides a data prediction method, where the method may be applied to various system platforms, and an execution subject of the method may be a processor in the system platform, and a flowchart of the method is shown in fig. 1, where the method includes:
s101: determining a target prediction model plug-in corresponding to a current business object to be predicted, wherein a prediction algorithm is arranged in the target prediction model plug-in;
in the method provided by the embodiment of the invention, when data prediction needs to be performed on a current business object to be predicted, a processor determines a target prediction model plug-in corresponding to the current business object to be predicted, the target prediction model plug-in is provided with a plurality of data interfaces, and the plurality of data interfaces comprise an interface for opening a parameter configuration interface, an interface for training a prediction model, an interface for verifying the prediction model, an interface for data prediction and the like. The target prediction model plug-in is also provided with a prediction algorithm, the prediction algorithm corresponds to the current business object to be predicted, and when the target prediction model plug-in is constructed, the prediction algorithm is set in the target prediction model plug-in advance according to the relevant attributes of the business object to be predicted.
In the method provided by the embodiment of the invention, the business object to be predicted can be understood as any business process in the specific business field, for example, in the field of fans, the operation process of the fans can be understood as the object to be predicted when the faults of the fans are predicted.
S102: starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameters in the prediction algorithm in the parameter configuration interface to generate an initial prediction model corresponding to the prediction algorithm;
in the method provided by the embodiment of the invention, the processor can open the parameter configuration interface corresponding to the target prediction model plug-in through the corresponding data interface in the target prediction model plug-in, and all algorithm parameters in the prediction algorithm can be displayed in the parameter configuration interface.
In the method provided by the embodiment of the invention, the data source corresponding to the business object to be predicted is obtained in real time according to the relevant characteristics of the business object to be predicted, and the data source is bound for the parameters in the prediction algorithm in the parameter configuration interface, so that the prediction algorithm can be understood as an algorithm framework, wherein the parameters are not assigned in the construction process, and the process of binding the data source for the parameters can be understood as the process of actually assigning the parameters.
In the method provided by the embodiment of the invention, the corresponding data source can be read from the preset data source storage unit.
In the method provided by the embodiment of the invention, after the data source is bound for the parameters in the prediction algorithm, the initial prediction model corresponding to the prediction algorithm, namely the initial prediction model for performing data prediction on the business object to be predicted, can be obtained.
S103: obtaining a first historical data set, carrying out model training on the initial prediction model, and obtaining a first prediction model corresponding to the initial prediction model;
in the method provided by the embodiment of the invention, after the initial prediction model is constructed, a first historical data set is obtained, the first historical data set comprises a plurality of first historical data, and each first historical data in the first historical data set is actual data generated in the past operation process of a business object to be predicted.
And performing model training on the initial prediction model by using each first historical data in the first historical data set, and taking the currently trained prediction model as a first prediction model corresponding to the initial prediction model when the initial prediction model meets a preset model training condition.
S104: acquiring a second historical data set, and performing model verification on the first prediction model;
in the method provided by the embodiment of the present invention, after a first prediction model is obtained, a second historical data set is obtained, where the second historical data set includes a plurality of second historical data, and each of the second historical data in the second historical data set is actual data of a to-be-predicted business object generated in a past operation process.
And verifying the first prediction model by using each second historical data in the second historical data set, and if the first prediction model is not verified, continuing to train the current first prediction model by using the first historical data set until the verification is passed.
In the method provided by the embodiment of the invention, if the verification fails, the algorithm parameters can be readjusted or configured, or a new target prediction model plug-in is reselected, and the training process is executed again.
In the method provided by the embodiment of the invention, the data prediction can not be carried out immediately after the first prediction model passes the model verification, so that the model training and the real-time prediction are two events with causal relationship, but not continuous events. And storing the prediction model obtained by training, and predicting a plurality of objects by using the same prediction model.
S105: when the first prediction model passes model verification, storing the first prediction model into a preset prediction model storage unit;
in the method provided by the embodiment of the invention, when the first prediction model passes the model verification, the first prediction model is stored in the prediction model storage unit, so that the first prediction model can be directly called in the subsequent data prediction process.
S106: when a prediction instruction for performing data prediction on the current business object to be predicted is received, acquiring the first prediction model in the prediction model storage unit, and reading real-time data corresponding to the data source in the current business object to be predicted;
in the method provided by the embodiment of the present invention, when data prediction is required, real-time data corresponding to the data source in a current business object to be predicted is read.
It can be understood that the data source is an actual assignment of a parameter of the prediction algorithm, the data source corresponds to the current business object to be predicted, and represents a relevant attribute of the business object to be predicted, and in the actual data prediction process, the real-time data corresponding to the data source is read, so that the data prediction of the business object to be predicted is realized by applying the real-time data.
S107: and inputting the real-time data into the first prediction model for data prediction, and outputting a data prediction result corresponding to the current business object to be predicted.
In the method provided by the embodiment of the invention, the read real-time data is input into the first prediction model, and after being processed by the first prediction model, the corresponding data prediction result can be output, so that the data prediction of the business object to be predicted is realized.
According to the data prediction method provided by the embodiment of the invention, the initial prediction model corresponding to the business object to be predicted is built in real time for different business objects to be predicted, the first prediction model corresponding to the initial prediction model is determined after the training and verification of the initial prediction model, and when the data prediction is needed to be carried out on the business object to be predicted, the data prediction is carried out in the first prediction model by combining the current real-time data of the business object to be predicted, and the prediction result is output.
Referring to fig. 2, a specific process of determining a target prediction model plug-in corresponding to a current business object to be predicted in the method provided by the embodiment of the present invention is shown, where the specific process includes:
s201: determining the object attribute of the current business object to be predicted;
s202: writing a plurality of prediction model plug-ins corresponding to the object attributes, and registering the prediction model plug-ins into a pre-established plug-in storage unit;
s203: and determining a target prediction model plug-in corresponding to the current business object to be predicted in each prediction model plug-in registered in the plug-in storage unit.
In the method provided by the embodiment of the present invention, when data prediction is required, for a current business object to be predicted, a plurality of prediction model plug-ins can be written according to object attributes of the current business object to be predicted, and it can be understood that different business objects to be predicted have corresponding object attributes, and the object attributes may be data information such as operating parameters and object characteristics. For example, the operation process of the fan needs to be predicted, the operation process of the fan serves as a current business object to be predicted, and relevant parameters of the fan and data parameters generated in the operation process can serve as object attributes.
In the method provided by the embodiment of the invention, each written prediction model plug-in can be combined into a plug-in package, and then the plug-in package is registered in the plug-in storage unit.
In the method provided by the embodiment of the invention, the plurality of compiled prediction model plug-ins can be model plug-ins with the same structure, and after the model plug-ins are registered in the plug-in storage unit, one prediction model plug-in is randomly selected from the plug-in storage unit as a target prediction model plug-in.
In the method provided by the embodiment of the invention, the plug-in storage unit can be a file, a database or a physical storage medium.
In the method provided by the embodiment of the invention, a plurality of compiled prediction model plug-ins, preferably, a certain difference exists among the prediction model plug-ins, when data prediction is carried out on the current business object to be predicted, a prediction algorithm of the data prediction process of the current business object to be predicted can be realized, a plurality of prediction model plug-ins can be provided, all the prediction algorithms capable of realizing the data prediction process of the business object to be predicted can be respectively packaged in the prediction model plug-ins, then the prediction model plug-ins are registered in a plug-in storage unit, and in the actual model construction process, according to actual needs, a more appropriate prediction model plug-in is selected as a target prediction model plug-in each prediction model plug-in.
In the method provided by the embodiment of the invention, each prediction model plug-in is provided with a plurality of data interfaces for reading data and other operations. And each prediction model plug-in is provided with a corresponding prediction algorithm, and each prediction algorithm corresponds to a business object to be predicted.
In the method provided by the embodiment of the invention, the process of determining the target prediction model plug-in can be as described in the above scheme, firstly, the corresponding prediction model plug-in is written, then, each prediction model plug-in is registered, and when data prediction is required, the target prediction model plug-in is determined from each registered prediction model plug-in.
In the method provided by the embodiment of the invention, when data prediction is required, a corresponding prediction model plug-in can be directly selected from a preset plug-in system as the target prediction model plug-in.
In the method provided by the embodiment of the present invention, the specific process of starting the parameter configuration interface corresponding to the target prediction model plug-in and binding the acquired data source for the parameter in the prediction algorithm in the parameter configuration interface includes:
calling a set editing interface, and starting a parameter configuration interface corresponding to the target prediction model plug-in;
binding the acquired data source corresponding to each parameter in the prediction algorithm for each parameter in the parameter configuration interface; and each data source corresponds to the current business object to be predicted.
In the method provided by the embodiment of the invention, the plurality of data interfaces of the target prediction model plug-in include an editing interface of a parameter configuration interface, the processor can open the parameter configuration interface corresponding to the target prediction model plug-in through the editing interface of the target prediction model plug-in, each algorithm parameter of the prediction algorithm is displayed in the parameter configuration interface, and data source binding is carried out on each algorithm parameter in the parameter configuration interface so as to establish the initial prediction model.
In the method provided by the embodiment of the present invention, the process of acquiring the first historical data set includes:
configuring a first historical data interval participating in model training;
and reading each first historical data corresponding to the data source in the first historical data interval, and forming each first historical data into the first historical data set.
In the method provided by the embodiment of the present invention, each first history data corresponding to the data source may be read in a preset file and a corresponding database, or a corresponding system interface may be called to read from a business system, or may be read in other realizable manners.
In the method provided by the embodiment of the invention, all historical data generated by the current business object to be predicted in the operation process are stored in the historical data storage unit, and when a processor needs to acquire a first historical data set, a first historical data interval participating in model training is configured in each historical data in the historical data storage unit, and the first historical data interval is configured according to the data prediction requirement of the current business object to be predicted.
In the configured first historical data interval, reading a plurality of first historical data corresponding to the data source to form a first historical data set.
In the method provided by the embodiment of the invention, the acquisition process of the second historical data set is the same as that of the first historical data set in principle, a corresponding second historical data interval for model verification is configured according to the data prediction requirement of the current business object to be predicted, and a plurality of second historical data are selected from the second historical data interval for model verification.
In the method provided by the embodiment of the invention, the first history interval and the second history interval, preferably the time interval, can select corresponding history data by setting the starting point and the ending point of the history interval.
The method provided by the embodiment of the invention further comprises the following steps:
and performing model training on the first prediction model by using the real-time data to obtain a second prediction model corresponding to the first prediction model, and replacing the first prediction model stored in a preset prediction model storage unit with the second prediction model.
In the method provided by the embodiment of the invention, when the real-time data is applied to carry out data prediction in the first prediction model, model training may also continue on the first predictive model using the acquired real-time data, the prediction model is continuously perfected through the corresponding learning unit, because the business object to be predicted is interfered by external factors in the actual operation process, the actual operation process has parameter variation, and the obtaining process of the first prediction model is obtained by training according to the historical data of the business object to be predicted, therefore, in the method provided by the embodiment of the present invention, after the first prediction model is obtained, according to the current real-time data, and the first prediction model is perfected, and a prediction model with higher accuracy is obtained through the self-learning process of the model.
Corresponding to the data prediction method shown in fig. 1, an embodiment of the present invention further provides a data prediction apparatus, which is used for implementing the method shown in fig. 1 specifically, and a schematic structural diagram of the data prediction apparatus is shown in fig. 3, where the data prediction apparatus includes:
a determining unit 301, configured to determine a target prediction model plug-in corresponding to a current business object to be predicted, where a prediction algorithm is set in the target prediction model plug-in;
a configuration unit 302, configured to start a parameter configuration interface corresponding to the target prediction model plug-in, and bind the acquired data source for the parameter in the prediction algorithm in the parameter configuration interface, so as to generate an initial prediction model corresponding to the prediction algorithm;
a training unit 303, configured to obtain a first historical data set, perform model training on the initial prediction model, and obtain a first prediction model corresponding to the initial prediction model;
a verification unit 304, configured to obtain a second historical data set, perform model verification on the first prediction model, and store the first prediction model in a preset prediction model storage unit when the first prediction model passes the model verification;
a prediction unit 305, configured to, when a prediction instruction for performing data prediction on the current business object to be predicted is received, obtain the first prediction model in the prediction model storage unit, and read real-time data corresponding to the data source in the current business object to be predicted; and inputting the real-time data into the first prediction model for data prediction, and outputting a data prediction result corresponding to the current business object to be predicted.
By applying the data prediction device provided by the embodiment of the invention, the initial prediction model corresponding to the business object to be predicted is built in real time for different business objects to be predicted, the first prediction model corresponding to the initial prediction model is determined after the training and verification of the initial prediction model, and when the data prediction is needed to be carried out on the business object to be predicted, the data prediction is carried out in the first prediction model by combining the current real-time data of the business object to be predicted, and the prediction result is output.
On the basis of fig. 3, referring to fig. 4, a schematic diagram of another structure of the data prediction apparatus provided in the embodiment of the present invention is shown, where the determining unit 301 includes:
a first determining subunit 307, configured to determine an object attribute of the current business object to be predicted;
a writing subunit 308, configured to write a plurality of prediction model plug-ins corresponding to the object attributes, and register the plurality of prediction model plug-ins in a pre-established plug-in storage unit;
a second determining subunit 309, configured to determine, in each prediction model plug-in registered in the plug-in storage unit, a target prediction model plug-in corresponding to the current business object to be predicted.
The data prediction device provided by the embodiment of the invention comprises a processor and a memory, wherein each unit is stored in the memory as a program unit, and the processor executes the program unit stored in the memory to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel may set one or more, dynamically performing the data prediction process by adjusting kernel parameters.
Corresponding to the data prediction method shown in fig. 1, an embodiment of the present invention further provides a data prediction apparatus, which is an instantiation of the data prediction method shown in fig. 1, and fig. 5 shows a schematic structural diagram of the data prediction apparatus, and specifically includes:
the device comprises a prediction model expansion device, a prediction model construction device and a data real-time prediction device;
the prediction model extension device comprises: the system comprises a plug-in development unit, a plug-in registration unit and a plug-in storage unit;
the plug-in development unit is used for developing a plurality of prediction model plug-ins according to the defined standard interface;
the plug-in registration unit is used for registering the plurality of prediction model plug-ins into the plug-in storage unit;
the prediction model construction device comprises: the device comprises a prediction model plug-in selection unit, a prediction model construction unit, a prediction model training unit and a prediction model verification unit;
the prediction model plug-in selection unit is used for determining a target prediction model plug-in the plug-in storage unit, and a prediction algorithm is arranged in the target prediction model plug-in;
the prediction model construction unit is used for starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameters in the prediction algorithm in the parameter configuration interface to generate an initial prediction model corresponding to the prediction algorithm;
the prediction model training unit is used for reading a plurality of first historical data used for model training, applying the plurality of first historical data to carry out model training on the initial prediction model, and obtaining a first prediction model corresponding to the initial prediction model;
the prediction model verification unit is used for reading a plurality of second historical data used for model verification and applying the plurality of second historical data to carry out model verification on the first prediction model;
the data real-time prediction device comprises: the real-time prediction system comprises a real-time data acquisition unit and a real-time prediction unit;
the real-time data acquisition unit is used for reading real-time data corresponding to the data source;
and the real-time prediction unit is used for inputting the real-time data into the first prediction model which passes model verification to perform data prediction and outputting a data prediction result corresponding to the current business object to be predicted.
The data prediction device provided by the embodiment of the invention is a specific implementation of the data prediction method in fig. 1, is a universal data prediction device, and can unify different data prediction methods in the practical application process, thereby solving the problem that the product has a single data prediction function.
The data prediction equipment provided by the embodiment of the invention has strong expandability and can customize a prediction algorithm model according to a specific service scene.
The equipment comprises three parts: the device comprises a prediction model expansion device, a prediction model construction device and a data real-time prediction device.
And the prediction model extension device is used for developing a new prediction model plug-in package and registering the plug-in package in the plug-in storage unit.
The prediction model extension device specifically comprises a plug-in development unit, a plug-in registration unit and a plug-in storage unit. The plug-in development unit is used for developing the prediction model plug-ins according to the defined standard interfaces and packaging the developed prediction model plug-ins into a plug-in package.
The plug-in development unit may be an existing product, or may be a building tool set (e.g., Visual Studio), and a plug-in package includes one or more prediction model plug-ins, and an interface that a prediction model plug-in needs to provide includes:
displaying an interface of a prediction model parameter editing interface, and establishing a new prediction model, modifying parameters of the established prediction model and binding a data source for algorithm parameters in a prediction algorithm through the interface;
an interface for training a prediction model, through which the model can be trained according to historical data corresponding to a data source;
the prediction model verification interface can verify the trained prediction model and provide a verification result;
the real-time prediction interface can predict future data according to real-time data corresponding to the data source and output a prediction result;
the predictive model presents an interface (optional) through which the predicted results are presented graphically or graphically.
And a plug-in registration unit for registering the developed plug-in package in the plug-in storage unit, wherein only the prediction model plug-in registered in the plug-in storage unit can be used. The plug-in storage unit stores various resources of the prediction model plug-in and provides an interface for reading and distributing the plug-in resources.
And the prediction model construction device is used for constructing an initial prediction model according to the selected target prediction model plug-in and the data source, training and verifying the initial prediction model according to the historical data corresponding to the data source, and storing the trained and verified first prediction model into the prediction model storage unit.
The prediction model plug-in selection unit is used for reading a prediction model plug-in list from the plug-in storage unit and selecting a proper plug-in as a target prediction model plug-in;
the data source selection unit is used for reading the data source list from the data source storage unit and selecting a proper data source;
the data source storage unit is used for storing the definition of the target data source and providing an interface for reading the data source; a prediction model construction unit for opening parameter configuration boundary according to the selected target prediction model plug-in
Binding the data source selected from the data source selection unit for the algorithm parameter of the prediction algorithm to generate an initial prediction model;
the historical data storage unit is used for storing historical data corresponding to the data source and providing a reading interface of the historical data;
a prediction model training unit for training an initial prediction model using the initial prediction model output from the prediction model construction unit and the historical data read from the historical data storage unit to obtain a corresponding first prediction model;
a prediction model verification unit operable to verify the first prediction model using the history data stored from the history data storage unit and store the first prediction model that has passed the verification to the prediction model storage unit;
and the prediction model storage unit is used for storing the prediction model, the mark indicating whether the prediction model is trained or not, the verification result and other data and providing an interface for reading the prediction model.
And the data real-time prediction device is used for predicting future data according to the trained prediction model and real-time data and continuously perfecting the prediction model through the prediction model learning unit.
The real-time data acquisition unit is used for acquiring real-time data of a data source and providing an interface for reading the real-time data of the data source;
the real-time prediction unit is used for calling a real-time prediction interface of the prediction model and outputting prediction result data;
the prediction model learning unit is used for retraining according to the real-time data of the data source and updating the prediction model into the prediction model storage unit;
and the prediction data display unit is used for displaying the prediction result in a graphical or charting mode.
The prediction process of the data prediction device, the actual execution steps can be as follows:
building a prediction model plug-in a plug-in storage unit;
(optional) developing a prediction model plug-in package, and registering the prediction model plug-in package in a plug-in storage unit;
selecting a proper target prediction model plug-in from a plug-in storage unit through a prediction model plug-in selection unit;
opening a parameter configuration interface corresponding to the selected target prediction model plug-in, and configuring corresponding parameters;
selecting a proper data source through a data source selection unit, binding the selected data source for the algorithm parameter of the prediction algorithm in the target prediction model plug-in unit, and generating an initial prediction model;
configuring a historical data interval participating in training, reading corresponding historical data of a data source in the interval, training an initial prediction model, and generating a first prediction model corresponding to the initial prediction model;
configuring a historical data interval participating in verification, reading corresponding historical data of a data source in the interval, and verifying a first prediction model;
storing the first prediction model in a prediction model storage unit;
reading the first prediction model, reading real-time data corresponding to the data source, predicting and outputting future data;
iteratively training a prediction model, and updating the self-learning prediction model into a prediction model storage unit;
and displaying the prediction result in a prediction data display unit.
An embodiment of the present invention provides a storage medium on which a program is stored, the program implementing the above data prediction method when executed by a processor.
The embodiment of the invention provides a processor, which is used for running a program, wherein the data prediction method is executed when the program runs.
As shown in fig. 6, an embodiment of the present invention provides an electronic device, where the electronic device 40 includes at least one processor 401, at least one memory 402 connected to the processor 401, and a bus 403; the processor 401 and the memory 402 complete communication with each other through the bus 403; the processor 401 is used to call program instructions in the memory 402 to perform the data prediction method described above. The electronic device herein may be a server, a PC, or the like.
The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device, comprising:
determining a target prediction model plug-in corresponding to a current business object to be predicted, wherein a prediction algorithm is arranged in the target prediction model plug-in;
starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameters in the prediction algorithm in the parameter configuration interface to generate an initial prediction model corresponding to the prediction algorithm;
obtaining a first historical data set, carrying out model training on the initial prediction model, and obtaining a first prediction model corresponding to the initial prediction model;
acquiring a second historical data set, and performing model verification on the first prediction model;
when the first prediction model passes model verification, storing the first prediction model into a preset prediction model storage unit;
when a prediction instruction for performing data prediction on the current business object to be predicted is received, acquiring the first prediction model in the prediction model storage unit, and reading real-time data corresponding to the data source in the current business object to be predicted;
and inputting the real-time data into the first prediction model for data prediction, and outputting a data prediction result corresponding to the current business object to be predicted.
Optionally, the method for determining the target prediction model plug-in corresponding to the current business object to be predicted includes:
determining the object attribute of the current business object to be predicted;
writing a plurality of prediction model plug-ins corresponding to the object attributes, and registering the prediction model plug-ins into a pre-established plug-in storage unit;
and determining a target prediction model plug-in corresponding to the current business object to be predicted in each prediction model plug-in registered in the plug-in storage unit.
Optionally, the starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameter in the prediction algorithm in the parameter configuration interface includes:
calling a set editing interface, and starting a parameter configuration interface corresponding to the target prediction model plug-in;
binding the acquired data source corresponding to each parameter in the prediction algorithm for each parameter in the parameter configuration interface; and each data source corresponds to the current business object to be predicted.
The method described above, optionally, the acquiring a first historical data set includes:
configuring a first historical data interval participating in model training;
and reading each first historical data corresponding to the data source in the first historical data interval, and forming each first historical data into the first historical data set.
The above method, optionally, further includes:
and performing model training on the first prediction model by using the real-time data to obtain a second prediction model corresponding to the first prediction model, and replacing the first prediction model stored in the prediction model storage unit with the second prediction model.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the system or system embodiments are substantially similar to the method embodiments and therefore are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for related points. The above-described system and system embodiments are only illustrative, wherein the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method of data prediction, comprising:
determining a target prediction model plug-in corresponding to a current business object to be predicted, wherein a prediction algorithm is arranged in the target prediction model plug-in;
starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameters in the prediction algorithm in the parameter configuration interface to generate an initial prediction model corresponding to the prediction algorithm;
obtaining a first historical data set, carrying out model training on the initial prediction model, and obtaining a first prediction model corresponding to the initial prediction model;
acquiring a second historical data set, and performing model verification on the first prediction model;
when the first prediction model passes model verification, storing the first prediction model into a preset prediction model storage unit;
when a prediction instruction for performing data prediction on the current business object to be predicted is received, acquiring the first prediction model in the prediction model storage unit, and reading real-time data corresponding to the data source in the current business object to be predicted;
and inputting the real-time data into the first prediction model for data prediction, and outputting a data prediction result corresponding to the current business object to be predicted.
2. The method according to claim 1, wherein the determining a target prediction model plug-in corresponding to the current business object to be predicted comprises:
determining the object attribute of the current business object to be predicted;
writing a plurality of prediction model plug-ins corresponding to the object attributes, and registering the prediction model plug-ins into a pre-established plug-in storage unit;
and determining a target prediction model plug-in corresponding to the current business object to be predicted in each prediction model plug-in registered in the plug-in storage unit.
3. The method according to claim 1, wherein the starting a parameter configuration interface corresponding to the target prediction model plug-in and binding the acquired data source for the parameter in the prediction algorithm in the parameter configuration interface comprises:
calling a set editing interface, and starting a parameter configuration interface corresponding to the target prediction model plug-in;
binding the acquired data source corresponding to each parameter in the prediction algorithm for each parameter in the parameter configuration interface; and each data source corresponds to the current business object to be predicted.
4. The method of claim 1, wherein said obtaining a first historical data set comprises:
configuring a first historical data interval participating in model training;
and reading each first historical data corresponding to the data source in the first historical data interval, and forming each first historical data into the first historical data set.
5. The method of claim 1, further comprising:
and performing model training on the first prediction model by using the real-time data to obtain a second prediction model corresponding to the first prediction model, and replacing the first prediction model stored in the prediction model storage unit with the second prediction model.
6. A data prediction apparatus, comprising:
the system comprises a determining unit, a predicting unit and a predicting unit, wherein the determining unit is used for determining a target predicting model plug-in corresponding to a current business object to be predicted, and a predicting algorithm is arranged in the target predicting model plug-in;
the configuration unit is used for starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameters in the prediction algorithm in the parameter configuration interface to generate an initial prediction model corresponding to the prediction algorithm;
the training unit is used for acquiring a first historical data set, carrying out model training on the initial prediction model and acquiring a first prediction model corresponding to the initial prediction model;
the verification unit is used for acquiring a second historical data set, carrying out model verification on the first prediction model, and storing the first prediction model into a preset prediction model storage unit when the first prediction model passes the model verification;
the prediction unit is used for acquiring the first prediction model in the prediction model storage unit when a prediction instruction for performing data prediction on the current business object to be predicted is received, and reading real-time data corresponding to the data source in the current business object to be predicted; and inputting the real-time data into the first prediction model for data prediction, and outputting a data prediction result corresponding to the current business object to be predicted.
7. The apparatus of claim 6, wherein the determining unit comprises:
the first determining subunit is used for determining the object attribute of the current business object to be predicted;
the compiling subunit is used for compiling a plurality of prediction model plug-ins corresponding to the object attributes and registering the prediction model plug-ins into a pre-established plug-in storage unit;
and the second determining subunit is configured to determine, in each prediction model plug-in registered in the plug-in storage unit, a target prediction model plug-in corresponding to the current business object to be predicted.
8. A data prediction apparatus, comprising:
the device comprises a prediction model expansion device, a prediction model construction device and a data real-time prediction device;
the prediction model extension device comprises: the system comprises a plug-in development unit, a plug-in registration unit and a plug-in storage unit;
the plug-in development unit is used for developing a plurality of prediction model plug-ins according to the defined standard interface;
the plug-in registration unit is used for registering the plurality of prediction model plug-ins into the plug-in storage unit;
the prediction model construction device comprises: the device comprises a prediction model plug-in selection unit, a prediction model construction unit, a prediction model training unit and a prediction model verification unit;
the prediction model plug-in selection unit is used for determining a target prediction model plug-in the plug-in storage unit, and a prediction algorithm is arranged in the target prediction model plug-in;
the prediction model construction unit is used for starting a parameter configuration interface corresponding to the target prediction model plug-in, and binding the acquired data source for the parameters in the prediction algorithm in the parameter configuration interface to generate an initial prediction model corresponding to the prediction algorithm;
the prediction model training unit is used for reading a plurality of first historical data used for model training, applying the plurality of first historical data to carry out model training on the initial prediction model, and obtaining a first prediction model corresponding to the initial prediction model;
the prediction model verification unit is used for reading a plurality of second historical data used for model verification and applying the plurality of second historical data to carry out model verification on the first prediction model;
the data real-time prediction device comprises: the real-time prediction system comprises a real-time data acquisition unit and a real-time prediction unit;
the real-time data acquisition unit is used for reading real-time data corresponding to the data source;
and the real-time prediction unit is used for inputting the real-time data into the first prediction model which passes model verification to perform data prediction and outputting a data prediction result corresponding to the current business object to be predicted.
9. A storage medium comprising stored instructions, wherein the instructions, when executed, control a device on which the storage medium resides to perform a data prediction method as claimed in any one of claims 1 to 5.
10. An electronic device comprising at least one processor, and at least one memory, bus connected to the processor; the processor and the memory complete mutual communication through a bus; the processor is used for calling program instructions in the memory to execute the data prediction method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210149141.4A CN114201246A (en) | 2022-02-18 | 2022-02-18 | Data prediction method and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210149141.4A CN114201246A (en) | 2022-02-18 | 2022-02-18 | Data prediction method and related equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114201246A true CN114201246A (en) | 2022-03-18 |
Family
ID=80645533
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210149141.4A Pending CN114201246A (en) | 2022-02-18 | 2022-02-18 | Data prediction method and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114201246A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111124840A (en) * | 2019-12-02 | 2020-05-08 | 北京天元创新科技有限公司 | Method and device for predicting alarm in business operation and maintenance and electronic equipment |
| CN111461379A (en) * | 2019-01-21 | 2020-07-28 | 北京嘀嘀无限科技发展有限公司 | Position prediction method and device |
| CN112633567A (en) * | 2020-12-16 | 2021-04-09 | 深圳前海微众银行股份有限公司 | Method and device for predicting waiting duration and storage medium |
| CN113013990A (en) * | 2021-03-18 | 2021-06-22 | 华润电力技术研究院有限公司 | Generator set fault early warning method, system and related equipment |
| CN113222057A (en) * | 2021-05-28 | 2021-08-06 | 中邮信息科技(北京)有限公司 | Data prediction model training method, data prediction device, data prediction equipment and data prediction medium |
| CN113610230A (en) * | 2021-08-16 | 2021-11-05 | 中国工商银行股份有限公司 | Training method, prediction method, device, computer system and storage medium |
| CN113780691A (en) * | 2020-06-09 | 2021-12-10 | 富泰华工业（深圳）有限公司 | Data testing method and device, electronic equipment and storage medium |
-
2022
- 2022-02-18 CN CN202210149141.4A patent/CN114201246A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111461379A (en) * | 2019-01-21 | 2020-07-28 | 北京嘀嘀无限科技发展有限公司 | Position prediction method and device |
| CN111124840A (en) * | 2019-12-02 | 2020-05-08 | 北京天元创新科技有限公司 | Method and device for predicting alarm in business operation and maintenance and electronic equipment |
| CN113780691A (en) * | 2020-06-09 | 2021-12-10 | 富泰华工业（深圳）有限公司 | Data testing method and device, electronic equipment and storage medium |
| CN112633567A (en) * | 2020-12-16 | 2021-04-09 | 深圳前海微众银行股份有限公司 | Method and device for predicting waiting duration and storage medium |
| CN113013990A (en) * | 2021-03-18 | 2021-06-22 | 华润电力技术研究院有限公司 | Generator set fault early warning method, system and related equipment |
| CN113222057A (en) * | 2021-05-28 | 2021-08-06 | 中邮信息科技(北京)有限公司 | Data prediction model training method, data prediction device, data prediction equipment and data prediction medium |
| CN113610230A (en) * | 2021-08-16 | 2021-11-05 | 中国工商银行股份有限公司 | Training method, prediction method, device, computer system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10671368B2 (en) | Automatic creation of delivery pipelines | |
| US20130298110A1 (en) | Software Visualization Using Code Coverage Information | |
| CN103501308A (en) | Distributed cloud application deployment systems and/or associated methods | |
| CN108710949A (en) | The method and system of template are modeled for creating machine learning | |
| CN105843873A (en) | System and method for managing data modeling | |
| CN112036577B (en) | Method and device for applying machine learning based on data form and electronic equipment | |
| US20100275186A1 (en) | Segmentation for static analysis | |
| CN106155806A (en) | A kind of multi-task scheduling method and server | |
| CN108270837A (en) | A kind of distributed task dispatching method and system using slack resources | |
| US8381170B2 (en) | Test driven architecture enabled process for open collaboration in global | |
| CN107391219A (en) | Function Compilation Method and device | |
| Sundas et al. | An introduction of CloudSim simulation tool for modelling and scheduling | |
| JP2023086678A (en) | Method and apparatus for generating and applying deep learning model based on deep learning framework | |
| CN108921407A (en) | A kind of task processing system and method | |
| CN108874520A (en) | Calculation method and device | |
| CN114595058A (en) | Model training method and device based on GPU (graphics processing Unit) resources, electronic equipment and storage medium | |
| CN109558146A (en) | Application issued method, apparatus and storage medium | |
| EP4152715A1 (en) | Method and apparatus for determining resource configuration of cloud service system | |
| CN116107869A (en) | Arrangement test method and device, electronic equipment and readable storage medium | |
| Araújo‐de‐Oliveira et al. | A procedural and flexible approach for specification, modeling, definition, and analysis for self‐adaptive systems | |
| CN114201246A (en) | Data prediction method and related equipment | |
| CN110362294A (en) | Development task executes method, apparatus, electronic equipment and storage medium | |
| CN114327779A (en) | Application operation environment building method and device and storage medium | |
| CN108804088A (en) | Protocol processes method and apparatus | |
| CN110351104A (en) | A kind of VIM selection method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220318 |