CN111796925A - Method and device for screening algorithm model, storage medium and electronic equipment - Google Patents
Method and device for screening algorithm model, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN111796925A CN111796925A CN201910282138.8A CN201910282138A CN111796925A CN 111796925 A CN111796925 A CN 111796925A CN 201910282138 A CN201910282138 A CN 201910282138A CN 111796925 A CN111796925 A CN 111796925A
- Authority
- CN
- China
- Prior art keywords
- information
- description information
- algorithm model
- screening
- environment
- 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.)
- Withdrawn
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 189
- 238000012216 screening Methods 0.000 title claims abstract description 90
- 238000000034 method Methods 0.000 title claims abstract description 70
- 238000003860 storage Methods 0.000 title claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 42
- 239000013598 vector Substances 0.000 claims description 37
- 238000004590 computer program Methods 0.000 claims description 17
- 230000015654 memory Effects 0.000 claims description 17
- 238000013507 mapping Methods 0.000 claims description 13
- 238000007635 classification algorithm Methods 0.000 claims description 11
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 19
- 230000000694 effects Effects 0.000 abstract description 7
- 238000000605 extraction Methods 0.000 description 8
- 230000008447 perception Effects 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000010354 integration Effects 0.000 description 5
- 230000001133 acceleration Effects 0.000 description 4
- 230000003190 augmentative effect Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 3
- 238000004140 cleaning Methods 0.000 description 3
- 238000013500 data storage Methods 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 239000011521 glass Substances 0.000 description 3
- 230000005484 gravity Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000013501 data transformation Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000000691 measurement method Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000005355 Hall effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
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
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The embodiment of the application discloses a screening method and device of an algorithm model, a storage medium and electronic equipment. The method comprises the following steps: acquiring device environment information, wherein the device environment information at least comprises: the service type, the equipment resource information and the application use information of the service to be processed; acquiring corresponding environment description information according to the equipment environment information; screening a target algorithm model from a preset model library based on the environment description information; and processing the service to be processed through the target algorithm model. According to the scheme, a more appropriate algorithm model can be dynamically matched for the service to be processed according to different current equipment environment information so as to process data, so that dynamic selection of the algorithm model is realized, and the processing effect of the algorithm model on the task is improved.
Description
Technical Field
The application relates to the field of electronic equipment, in particular to a method and a device for screening algorithm models, a storage medium and electronic equipment.
Background
With the development of electronic technology, electronic devices such as smart phones have become more and more intelligent. The electronic device may perform data processing through various algorithmic models to provide various functions to the user. For example, the electronic device may learn behavior characteristics of the user according to the algorithm model, thereby providing personalized services to the user.
Disclosure of Invention
The embodiment of the application provides a screening method and device of an algorithm model, a storage medium and electronic equipment, which can realize dynamic selection of the algorithm model and improve the task processing effect of the algorithm model.
In a first aspect, an embodiment of the present application provides a method for screening an algorithm model, including:
acquiring device environment information, wherein the device environment information at least comprises: the service type, the equipment resource information and the application use information of the service to be processed;
acquiring corresponding environment description information according to the equipment environment information;
screening a target algorithm model from a preset model library based on the environment description information;
and processing the service to be processed through the target algorithm model.
In a second aspect, an embodiment of the present application further provides a screening apparatus for an algorithm model, including:
a first obtaining module, configured to obtain device environment information, where the device environment information at least includes: the service type, the equipment resource information and the application use information of the service to be processed;
the second acquisition module is used for acquiring corresponding environment description information according to the equipment environment information;
the screening module is used for screening a target algorithm model from a preset model library based on the environment description information;
and the processing module is used for processing the service to be processed through the target algorithm model.
In a third aspect, an embodiment of the present application further provides a storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the method for screening an algorithm model.
In a fourth aspect, an embodiment of the present application further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the method for screening an algorithm model described above when executing the program.
According to the screening method of the algorithm model provided by the embodiment of the application, the equipment environment information is obtained and at least comprises the following steps: the service type, the equipment resource information and the application use information of the service to be processed; acquiring corresponding environment description information according to the equipment environment information; screening a target algorithm model from a preset model library based on the environment description information; and processing the service to be processed through the target algorithm model. According to the scheme, a more appropriate algorithm model can be dynamically matched for the service to be processed according to different current equipment environment information so as to process data, so that dynamic selection of the algorithm model is realized, and the processing effect of the algorithm model on the task is improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings used in the description of the embodiments will be briefly introduced below. It is obvious that the drawings in the following description are only some embodiments of the application, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
Fig. 1 is a schematic view of a panoramic sensing architecture provided in an embodiment of the present application.
Fig. 2 is a first flowchart of a method for screening an algorithm model according to an embodiment of the present disclosure.
Fig. 3 is a second flowchart of the algorithm model screening method according to the embodiment of the present application.
Fig. 4 is a third flowchart illustrating a screening method of an algorithm model according to an embodiment of the present application.
Fig. 5 is a fourth flowchart illustrating a screening method of an algorithm model according to an embodiment of the present disclosure.
Fig. 6 is a schematic structural diagram of a screening apparatus for an algorithm model provided in an embodiment of the present application.
Fig. 7 is a schematic structural diagram of a first electronic device according to an embodiment of the present application.
Fig. 8 is a schematic structural diagram of a second electronic device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without inventive step, are within the scope of the present application.
Referring to fig. 1, fig. 1 is a schematic view of a panoramic sensing architecture provided in an embodiment of the present application. The method for screening the algorithm model is applied to electronic equipment. A panoramic perception framework is arranged in the electronic equipment. The panoramic perception architecture is an integration of hardware and software used for implementing the screening method of the algorithm model in the electronic equipment.
The panoramic perception architecture comprises an information perception layer, a data processing layer, a feature extraction layer, a scene modeling layer and an intelligent service layer.
The information perception layer is used for acquiring information of the electronic equipment or information in an external environment. The information-perceiving layer may include a plurality of sensors. For example, the information sensing layer includes a plurality of sensors such as a distance sensor, a magnetic field sensor, a light sensor, an acceleration sensor, a fingerprint sensor, a hall sensor, a position sensor, a gyroscope, an inertial sensor, an attitude sensor, a barometer, and a heart rate sensor.
Among other things, a distance sensor may be used to detect a distance between the electronic device and an external object. The magnetic field sensor may be used to detect magnetic field information of the environment in which the electronic device is located. The light sensor can be used for detecting light information of the environment where the electronic equipment is located. The acceleration sensor may be used to detect acceleration data of the electronic device. The fingerprint sensor may be used to collect fingerprint information of a user. The Hall sensor is a magnetic field sensor manufactured according to the Hall effect, and can be used for realizing automatic control of electronic equipment. The location sensor may be used to detect the geographic location where the electronic device is currently located. Gyroscopes may be used to detect angular velocity of an electronic device in various directions. Inertial sensors may be used to detect motion data of an electronic device. The gesture sensor may be used to sense gesture information of the electronic device. A barometer may be used to detect the barometric pressure of the environment in which the electronic device is located. The heart rate sensor may be used to detect heart rate information of the user.
And the data processing layer is used for processing the data acquired by the information perception layer. For example, the data processing layer may perform data cleaning, data integration, data transformation, data reduction, and the like on the data acquired by the information sensing layer.
The data cleaning refers to cleaning a large amount of data acquired by the information sensing layer to remove invalid data and repeated data. The data integration refers to integrating a plurality of single-dimensional data acquired by the information perception layer into a higher or more abstract dimension so as to comprehensively process the data of the plurality of single dimensions. The data transformation refers to performing data type conversion or format conversion on the data acquired by the information sensing layer so that the transformed data can meet the processing requirement. The data reduction means that the data volume is reduced to the maximum extent on the premise of keeping the original appearance of the data as much as possible.
The characteristic extraction layer is used for extracting characteristics of the data processed by the data processing layer so as to extract the characteristics included in the data. The extracted features may reflect the state of the electronic device itself or the state of the user or the environmental state of the environment in which the electronic device is located, etc.
The feature extraction layer may extract features or process the extracted features by a method such as a filtering method, a packing method, or an integration method.
The filtering method is to filter the extracted features to remove redundant feature data. Packaging methods are used to screen the extracted features. The integration method is to integrate a plurality of feature extraction methods together to construct a more efficient and more accurate feature extraction method for extracting features.
The scene modeling layer is used for building a model according to the features extracted by the feature extraction layer, and the obtained model can be used for representing the state of the electronic equipment, the state of a user, the environment state and the like. For example, the scenario modeling layer may construct a key value model, a pattern identification model, a graph model, an entity relation model, an object-oriented model, and the like according to the features extracted by the feature extraction layer.
The intelligent service layer is used for providing intelligent services for the user according to the model constructed by the scene modeling layer. For example, the intelligent service layer can provide basic application services for users, perform system intelligent optimization for electronic equipment, and provide personalized intelligent services for users.
In addition, the panoramic perception architecture can further comprise a plurality of algorithms, each algorithm can be used for analyzing and processing data, and the plurality of algorithms can form an algorithm library. For example, the algorithm library may include algorithms such as a markov algorithm, a hidden dirichlet distribution algorithm, a bayesian classification algorithm, a support vector machine, a K-means clustering algorithm, a K-nearest neighbor algorithm, a conditional random field, a residual error network, a long-short term memory network, a convolutional neural network, and a cyclic neural network.
The embodiment of the application provides a screening method of an algorithm model, which can be applied to electronic equipment. The electronic device may be a smartphone, a tablet computer, a gaming device, an AR (augmented reality) device, an automobile, a data storage device, an audio playback device, a video playback device, a notebook, a desktop computing device, a wearable device such as a watch, glasses, a helmet, an electronic bracelet, an electronic necklace, an electronic garment, or the like.
Referring to fig. 2, fig. 2 is a schematic flowchart of a first method for screening an algorithm model according to an embodiment of the present application. The screening method of the algorithm model comprises the following steps:
and 110, acquiring device environment information, wherein the device environment information at least comprises: service type, device resource information, and application usage information of the pending service.
The service type mainly refers to what type of service needs to be provided for the user. The service type may be divided based on the functions specifically provided by the service, such as navigation class, driving class, game class, video class, music recommendation, and the like.
The device resource information refers to internal resource information of the electronic device, such as electric quantity, CPU (Central Processing Unit) occupation amount, GPU (Graphics Processing Unit) occupation amount, storage space, network speed, and the like.
The application use information refers to the use of the application installed in the current device. The application program mentioned in this embodiment may be any application installed on the electronic device, such as an office application, a social application, a game application, a shopping application, and the like. For example, a user is currently playing a game using a game application, listening to a song using a music playing application, or the like.
In some embodiments, the application usage information may also be derived based on application usage record predictions for the user's usage of the application program over historical time periods. In practical application, the use information of each installed application can be recorded after the application is installed, and the use information is converted into corresponding data to be stored in a preset storage area. When the use information of one or some applications is needed to be used, the data corresponding to the one or some applications in a certain time period can be called from the storage area, and the obtained data is analyzed to obtain corresponding information which is used as the use information of the one or some applications in a historical time period. Wherein the historical time period may be the past month, the last week, etc. Then, based on the obtained application usage record, the current application usage information can be predicted and obtained through a preset algorithm model.
And 120, acquiring corresponding environment description information according to the device environment information.
The environment description information may be text type information, image type information, etc. for describing the content or attribute of the device environment information.
In some embodiments, the step of "obtaining corresponding environment description information according to the device environment information" may include the following processes:
acquiring first description information of the service type, second description information of the equipment resource information and third description information of the application use information;
preprocessing the first description information, the second description information and the third description information;
and generating the environment description information according to the preprocessed first description information, second description information and third description information.
In this embodiment, the first description information may be used to describe an operation index of the service type service. For example, the service to be processed currently is a navigation service, and a route needs to be planned by using the service prediction, and the first description information corresponding to the navigation service may include a precision requirement for route planning, a delay requirement for real-time feedback, and the like. The second description information may be information for describing whether resources are sufficient, such as whether power resources are sufficient, whether a network speed is fast, and the like. The third description information may be information for describing an application operation index, such as a data reading speed, a screen display frame rate, and the like that need to be reached when a certain application is applied.
In some embodiments, the device resource information includes resource amounts of a plurality of resources. Then, when obtaining the second description information of the device resource information, the following process may be included:
comparing the resource quantity with a corresponding preset threshold value to obtain a plurality of comparison results;
and generating second description information according to the comparison results.
Specifically, the plurality of resources may include power resources, CPU resources, storage space resources, network speed, other resources, and the like in the electronic device. Because the measurement standards of different resources are different, the resource amount of each resource needs to be acquired, and then the resource amount of each resource is compared with the corresponding preset threshold value to obtain the comparison result of multiple resources. For example, for the electricity resource, if the current electricity is 90%, and the corresponding preset threshold may be 60%, the comparison result may be that the current electricity is higher than the preset threshold; for another example, taking the CPU resource as an example, if the current CPU occupancy is 80%, and the corresponding preset threshold may be 50%, the comparison result may be that the current CPU occupancy is higher than the preset threshold.
Then, second descriptive information is generated based on the obtained plurality of comparison results. Still taking the above-mentioned electric quantity resources and CPU resources as examples, the comparison result includes: if the current power is higher than the preset threshold and the current CPU occupancy is higher than the preset threshold, the second description information may be generated as follows: sufficient electric quantity and overhigh CPU occupation.
In some embodiments, when obtaining the third description information of the application usage information, the following process may be included:
determining a target application currently used according to the application use information;
determining attribute information of the target application;
third description information is determined according to the attribute information.
In this embodiment, the running information of all current applications may be acquired, and then the target application currently used may be determined according to the running information of the applications. The currently used target application may be a foreground running application. For example, if a user is currently playing a game using a game application, the game application is used as a target application.
It should be noted that, in this embodiment, the attribute information may include the type of the application, the data size of the application, configuration information of the application, and indexes (such as frame rate, network speed, and reading speed) in normal operation. Then, the third description information may be determined based on the attribute information of the target application. For example, if the target application is a relatively large network sports game application, the real-time performance requirement for normal operation of the target application is relatively high, and the operation index may include a requirement for network speed, a requirement for reading speed, and the like. Therefore, the third description information generated based on the attribute information of the game application may be information for describing a required size of the network speed, a required size of the reading speed.
And 130, screening a target algorithm model from a preset model library based on the environment description information.
In the embodiment of the present application, a plurality of algorithm models may be set in an algorithm model library preset in the electronic device. For example, the algorithm model library may include: the method comprises an algorithm model such as a Markov algorithm, a hidden Dirichlet distribution algorithm, a Bayesian classification algorithm, a support vector machine, a K-means clustering algorithm, a K neighbor algorithm, a conditional random field, a residual error network, a long-short term memory network, a convolutional neural network and a cyclic neural network. Different algorithm models have different functional characteristics, such as an LSTM (Long short-Term Memory) model, which is a time recurrent neural network and is suitable for processing and predicting important events with relatively long interval and delay in a time sequence; for another example, an RNN (Recurrent Neural Network) model, whose internal state can exhibit dynamic time sequence behavior, can process an input sequence with any time sequence by using internal memory, which makes it easier to process non-segmented handwriting recognition, speech recognition, etc.
Specifically, based on the acquired environment description information, a target algorithm model suitable for processing the service to be processed in the current equipment environment is screened from a preset algorithm model library.
And 140, processing the service to be processed through the target algorithm model.
In practical application, the algorithm model can be configured according to tasks (i.e. to-be-processed services) required to be executed by the electronic equipment, so that the processing effect is improved. The algorithm model comprises a configuration file which declares the algorithm module to be called. When the electronic equipment executes the task corresponding to the algorithm model, the main program reads in the configuration file to generate an algorithm model graph. The algorithm model graph comprises algorithm modules needing to be called and data input and output relations among the algorithm modules. Then, the main program calls the declared algorithm module according to the algorithm model graph and generates an executable program to process the task.
In the embodiment of the present application, the manner of screening the target algorithm model may be various. For example, in some embodiments, referring to fig. 3, fig. 3 is a second flowchart of a screening method of an algorithm model provided in the embodiments of the present application.
In some embodiments, the step of "screening the target algorithm model from the preset model library based on the environment description information" may include the following procedures:
131a, matching corresponding modeling indexes for the environment description information based on a preset classification algorithm;
132a, screening the target algorithm model from the plurality of sample algorithm models in the preset model library according to the modeling index.
Specifically, the sensed environment description information may be processed according to a preset classification algorithm, so as to obtain a current modeling gravity center as a modeling index. In some embodiments, the modeling center of gravity may be divided into: the method has the advantages that the method can be customized in a targeted manner according to task requirements in actual work, and the method is energy-saving, high in real-time requirement, low in real-time requirement, high in accuracy, low in accuracy, interference-avoiding (namely, interference-avoiding of the current behavior of a user) and the like. And then, based on the obtained modeling indexes, matching a corresponding sample algorithm model from a preset model base to serve as an optimal algorithm model for processing the tasks to be processed under the current equipment environment.
In some embodiments, the environment description information may be text information, and the determining of the modeling barycenter may be based on matching corresponding modeling indicators according to key words in the text information. That is, the step "matching the corresponding modeling index for the environment description information based on the preset classification algorithm" may include the following steps:
extracting a plurality of first keywords from the environment description information;
obtaining a modeling index matched with the first keyword according to a preset mapping relation, wherein the preset mapping relation comprises: and mapping relation between the sample keywords and the sample modeling indexes.
In the embodiment of the application, a mapping relationship between the sample keywords and the sample modeling indexes needs to be constructed in advance. For example, if the sample keyword is "low power", the corresponding matched sample modeling index may be set to "low power consumption", and if the sample keyword is "short time delay", the corresponding matched sample modeling index may be set to "high processing speed".
In some embodiments, the step of "screening the target algorithm model from a plurality of sample algorithm models in the preset model library according to the modeling index" may include the following steps:
acquiring the number of first keywords which are correspondingly matched with each modeling index;
determining the priority of the modeling index according to the number of the first keywords;
and screening a target algorithm model from the preset model library according to the priority of the modeling index.
Specifically, a corresponding modeling index is matched for each keyword, and the modeling indexes are prioritized according to the number of keywords matched by the modeling indexes, so as to determine the importance degree of each modeling index matched by the modeling indexes. When the algorithm models are screened, the algorithm models corresponding to the modeling indexes with higher priority tend to be selected as target algorithm models according to the importance degree of each modeling index.
In some embodiments, referring to fig. 4, fig. 4 is a third flowchart illustrating a screening method of an algorithm model provided in an embodiment of the present application.
In the embodiment of the application, a model library for panoramic view modeling needs to be constructed in advance, and meanwhile, corresponding labels are given to each algorithm model, for example, the labels of the panoramic view model based on the neural network are large in computation amount, low in real-time performance, high in precision and the like, while the labels of the panoramic view model based on the classification algorithm are small in computation amount, high in real-time performance, low in precision and the like.
Specifically, the target algorithm model can be screened from a preset model library based on a semantic similarity measurement method for keyword matching. That is, in some embodiments, the step "screening a target algorithm model from a plurality of sample algorithm models in a preset model library according to the modeling index" may include the following steps:
131b, extracting a second keyword from the text description information corresponding to the modeling index;
132b, determining a label corresponding to each sample algorithm model, and extracting a third key word from the text description information corresponding to the label;
133b, calculating the matching degree of the second keyword and the third keyword;
134b, screening the target algorithm model from the preset model library according to the matching degree.
Specifically, each modeling index (i.e., modeling center of gravity) has a corresponding description text. Similarly, each tag has a corresponding description text. In practical application, a second keyword can be extracted from the text description information corresponding to the modeling index, and then a text vector is directly constructed based on the second keyword; similarly, a third key word can be extracted from the description text of the label corresponding to each sample algorithm model, and then a text vector can be directly constructed based on the third key word. And then, directly performing text matching on the two text vectors, and counting the number of texts which commonly appear. This value is used as a semantic similarity (i.e., degree of match) between the modeling indicator and the text label based on the number of co-occurring texts divided by the length of the text vector. And finally, screening the sample algorithm model with the maximum matching degree from the preset model library as a target algorithm model according to the matching degree of the modeling index and the corresponding label of each sample algorithm model.
In some embodiments, referring to fig. 5, fig. 5 is a fourth flowchart illustrating a screening method of an algorithm model provided in an embodiment of the present application.
Specifically, a target algorithm model can be screened from a preset algorithm model library based on a semantic similarity measurement method of word vectors. That is, the step of "screening a target algorithm model from a plurality of sample algorithm models in a preset model library according to the modeling index" may include the following steps:
131c, constructing a first word vector corresponding to the modeling index;
131c, constructing a second word vector of each sample algorithm model corresponding to the label;
131c, calculating a vector distance between the first word vector and the second word vector;
131c, screening the target algorithm model from the preset model library based on the vector distance.
Specifically, the modeling indexes and the algorithm labels can be learned through the word vectors, and the word vectors of all the labels and the modeling indexes are obtained. Then, based on the word vectors of the algorithm labels and the word vectors of the modeling indexes, the vector distance (or similarity) between the two vectors is calculated by adopting cosine similarity, and the distance between the modeling indexes and each label of the algorithm is obtained, namely the semantic similarity is obtained. And finally, screening the sample algorithm model with the minimum vector distance from the preset model library as a target algorithm model according to the semantic vector distance between the modeling index obtained by calculation and the corresponding label of each sample algorithm model.
It is to be understood that the terms "first," "second," and the like in the embodiments of the present application are used merely for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order, such that the described elements may be interchanged under appropriate circumstances.
In particular implementation, the present application is not limited by the execution sequence of the described steps, and some steps may be performed in other sequences or simultaneously without conflict.
As can be seen from the above, in the method for screening an algorithm model provided in the embodiment of the present application, by obtaining the device environment information, the device environment information at least includes: the service type, the equipment resource information and the application use information of the service to be processed; acquiring corresponding environment description information according to the equipment environment information; screening a target algorithm model from a preset model library based on the environment description information; and processing the service to be processed through the target algorithm model. According to the scheme, a more appropriate algorithm model can be dynamically matched for the service to be processed according to different current equipment environment information so as to process data, so that dynamic selection of the algorithm model is realized, and the processing effect of the algorithm model on the task is improved.
In some embodiments, based on the algorithm model screening method of the embodiments of the present application, first, device environment information of an electronic device of a user, such as a service type of a service to be processed, device resource information, application use information, and the like, is obtained through an information sensing layer. Then, processing (such as invalid data deletion) is carried out through the environment information of the data processing layer, then required environment description information is extracted from the information processed by the data processing layer through the characteristic extraction layer, then the environment description information is input into the scene modeling layer, the scene modeling layer comprises a pre-stored prediction model, and the prediction model of the scene modeling layer is trained according to the environment description information to obtain the trained prediction model. And finally, when the intelligent service layer processes the task, the trained prediction model can be used for identifying the equipment environment information of the current electronic equipment, and an appropriate model is screened from the model base based on the equipment environment information to process the current task, so that the intelligent service efficiency is improved.
The embodiment of the application also provides a screening device of the algorithm model. The screening means of the algorithm model may be integrated in the electronic device. The electronic device may be a smartphone, a tablet computer, a gaming device, an AR (Augmented Reality) device, an automobile, a data storage device, an audio playback device, a video playback device, a notebook, a desktop computing device, a wearable device such as a watch, glasses, a helmet, an electronic bracelet, an electronic necklace, an electronic garment, or the like.
Referring to fig. 6, fig. 6 is a schematic structural diagram of a screening apparatus of an algorithm model according to an embodiment of the present application. The screening apparatus 200 for algorithm models provided in the embodiment of the present application may include:
a first obtaining module 201, configured to obtain device environment information, where the device environment information at least includes: the service type, the equipment resource information and the application use information of the service to be processed;
a second obtaining module 202, configured to obtain corresponding environment description information according to the device environment information;
the screening module 203 is used for screening a target algorithm model from a preset model library based on the environment description information;
and the processing module 204 is configured to process the service to be processed through the target algorithm model.
In some embodiments, the second obtaining module 202 may include:
the obtaining submodule is used for obtaining first description information of the service type, second description information of the equipment resource information and third description information of the application use information;
the processing submodule is used for preprocessing the first description information, the second description information and the third description information;
and the generation submodule is used for generating the environment description information according to the preprocessed first description information, second description information and third description information.
In some embodiments, the device resource information includes resource amounts of a plurality of resources; the acquisition sub-module may be further operable to:
comparing the resource quantity with a corresponding preset threshold value to obtain a plurality of comparison results;
and generating second description information according to the comparison results.
In some embodiments, the acquisition sub-module is further operable to:
determining a target application currently used according to the application use information;
determining attribute information of the target application;
third description information is determined according to the attribute information.
In some embodiments, the screening module 203 may include:
the matching sub-module is used for matching corresponding modeling indexes for the environment description information based on a preset classification algorithm;
and the screening submodule is used for screening a target algorithm model from a plurality of sample algorithm models in a preset model library according to the modeling index.
In some embodiments, the matching sub-module may be configured to:
extracting a plurality of first keywords from the environment description information;
obtaining a modeling index matched with the first keyword according to a preset mapping relation, wherein the preset mapping relation comprises: and mapping relation between the sample keywords and the sample modeling indexes.
In some embodiments, the screening submodule may be configured to:
acquiring the number of first keywords which are correspondingly matched with each modeling index;
determining the priority of the modeling index according to the number of the first keywords;
and screening a target algorithm model from the preset model library according to the priority of the modeling index.
In some embodiments, the screening submodule may be configured to:
extracting a second keyword from the text description information corresponding to the modeling index;
determining a label corresponding to each sample algorithm model, and extracting a third key word from text description information corresponding to the label;
calculating the matching degree of the second key words and the third key words;
and screening a target algorithm model from the preset model library according to the matching degree.
In some embodiments, the screening submodule may be configured to:
constructing a first word vector corresponding to the modeling index;
constructing a second word vector of the label corresponding to each sample algorithm model;
calculating a vector distance between the first word vector and the second word vector;
and screening a target algorithm model from the preset model library based on the vector distance.
As can be seen from the above, the screening apparatus 200 for algorithm models provided in the embodiment of the present application includes: a first obtaining module 201, configured to obtain device environment information, where the device environment information at least includes: the service type, the equipment resource information and the application use information of the service to be processed; a second obtaining module 202, configured to obtain corresponding environment description information according to the device environment information; the screening module 203 is used for screening a target algorithm model from a preset model library based on the environment description information; and the processing module 204 is configured to process the service to be processed through the target algorithm model. According to the scheme, a more appropriate algorithm model can be dynamically matched for the service to be processed according to different current equipment environment information so as to process data, so that dynamic selection of the algorithm model is realized, and the processing effect of the algorithm model on the task is improved.
The embodiment of the application also provides the electronic equipment. The electronic device may be a smartphone, a tablet computer, a gaming device, an AR (Augmented Reality) device, an automobile, a data storage device, an audio playback device, a video playback device, a notebook, a desktop computing device, a wearable device such as a watch, glasses, a helmet, an electronic bracelet, an electronic necklace, an electronic garment, or the like. The electronic equipment is provided with an algorithm model, the algorithm model comprises a first algorithm module, and the first algorithm module is used for processing a preset task.
Referring to fig. 7, fig. 7 is a schematic view of a first structure of an electronic device 300 according to an embodiment of the present disclosure. Electronic device 300 includes, among other things, a processor 301 and a memory 302. The processor 301 is electrically connected to the memory 302.
The processor 301 is a control center of the electronic device 300, connects various parts of the entire electronic device using various interfaces and lines, and performs various functions of the electronic device and processes data by running or calling a computer program stored in the memory 302 and calling data stored in the memory 302, thereby performing overall monitoring of the electronic device.
In this embodiment, the processor 301 in the electronic device 300 loads instructions corresponding to one or more processes of the computer program into the memory 302 according to the following steps, and the processor 301 runs the computer program stored in the memory 302, so as to implement various functions:
acquiring device environment information, wherein the device environment information at least comprises: the service type, the equipment resource information and the application use information of the service to be processed;
acquiring corresponding environment description information according to the equipment environment information;
screening a target algorithm model from a preset model library based on the environment description information;
and processing the service to be processed through the target algorithm model.
In some embodiments, when acquiring the corresponding environment description information according to the device environment information, the processor 301 performs the following steps:
acquiring first description information of the service type, second description information of the equipment resource information and third description information of the application use information;
preprocessing the first description information, the second description information and the third description information;
and generating the environment description information according to the preprocessed first description information, second description information and third description information.
In some embodiments, the device resource information includes resource amounts of a plurality of resources; when acquiring the second description information of the device resource information, the processor 301 further performs the following steps:
comparing the resource quantity with a corresponding preset threshold value to obtain a plurality of comparison results;
and generating second description information according to the comparison results.
In some embodiments, when obtaining the third description information of the application usage information, the processor 301 performs the following steps:
determining a target application currently used according to the application use information;
determining attribute information of the target application;
third description information is determined according to the attribute information.
In some embodiments, when the target algorithm model is screened from the preset model library based on the environment description information, the processor 301 performs the following steps:
matching corresponding modeling indexes for the environment description information based on a preset classification algorithm;
and screening a target algorithm model from a plurality of sample algorithm models in a preset model library according to the modeling index.
In some embodiments, when matching the corresponding modeling index for the environment description information based on a preset classification algorithm, the processor 301 performs the following steps:
extracting a plurality of first keywords from the environment description information;
obtaining a modeling index matched with the first keyword according to a preset mapping relation, wherein the preset mapping relation comprises: and mapping relation between the sample keywords and the sample modeling indexes.
In some embodiments, when the target algorithm model is screened from a plurality of sample algorithm models in the preset model library according to the modeling index, the processor 301 performs the following steps:
acquiring the number of first keywords which are correspondingly matched with each modeling index;
determining the priority of the modeling index according to the number of the first keywords;
and screening a target algorithm model from the preset model library according to the priority of the modeling index.
In some embodiments, when the target algorithm model is screened from a plurality of sample algorithm models in the preset model library according to the modeling index, the processor 301 performs the following steps:
extracting a second keyword from the text description information corresponding to the modeling index;
determining a label corresponding to each sample algorithm model, and extracting a third key word from text description information corresponding to the label;
calculating the matching degree of the second key words and the third key words;
and screening a target algorithm model from the preset model library according to the matching degree.
In some embodiments, when the target algorithm model is screened from a plurality of sample algorithm models in the preset model library according to the modeling index, the processor 301 performs the following steps:
constructing a first word vector corresponding to the modeling index;
constructing a second word vector of the label corresponding to each sample algorithm model;
calculating a vector distance between the first word vector and the second word vector;
and screening a target algorithm model from the preset model library based on the vector distance.
In some embodiments, referring to fig. 8, fig. 8 is a schematic diagram of a second structure of an electronic device 300 according to an embodiment of the present disclosure.
Wherein, the electronic device 300 further comprises: a display 303, a control circuit 304, an input unit 305, a sensor 306, and a power supply 307. The processor 301 is electrically connected to the display 303, the control circuit 304, the input unit 305, the sensor 306, and the power source 307.
The display screen 303 may be used to display information entered by or provided to the user as well as various graphical user interfaces of the electronic device, which may be comprised of images, text, icons, video, and any combination thereof.
The control circuit 304 is electrically connected to the display 303, and is configured to control the display 303 to display information.
The input unit 305 may be used to receive input numbers, character information, or user characteristic information (e.g., fingerprint), and generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function control. Wherein, the input unit 305 may include a fingerprint recognition module.
The sensor 306 is used to collect information of the electronic device itself or information of the user or external environment information. For example, the sensor 306 may include a plurality of sensors such as a distance sensor, a magnetic field sensor, a light sensor, an acceleration sensor, a fingerprint sensor, a hall sensor, a position sensor, a gyroscope, an inertial sensor, an attitude sensor, a barometer, a heart rate sensor, and the like.
The power supply 307 is used to power the various components of the electronic device 300. In some embodiments, the power supply 307 may be logically coupled to the processor 301 through a power management system, such that functions of managing charging, discharging, and power consumption are performed through the power management system.
Although not shown in fig. 8, the electronic device 300 may further include a camera, a bluetooth module, and the like, which are not described in detail herein.
As can be seen from the above, an embodiment of the present application provides an electronic device, where the electronic device performs the following steps: acquiring device environment information, wherein the device environment information at least comprises: the service type, the equipment resource information and the application use information of the service to be processed; acquiring corresponding environment description information according to the equipment environment information; screening a target algorithm model from a preset model library based on the environment description information; and processing the service to be processed through the target algorithm model. According to the scheme, a more appropriate algorithm model can be dynamically matched for the service to be processed according to different current equipment environment information so as to process data, so that dynamic selection of the algorithm model is realized, and the processing effect of the algorithm model on the task is improved.
The embodiment of the present application further provides a storage medium, in which a computer program is stored, and when the computer program runs on a computer, the computer executes the method for screening an algorithm model according to any one of the above embodiments.
For example, in some embodiments, when the computer program is run on a computer, the computer performs the steps of:
acquiring device environment information, wherein the device environment information at least comprises: the service type, the equipment resource information and the application use information of the service to be processed;
acquiring corresponding environment description information according to the equipment environment information;
screening a target algorithm model from a preset model library based on the environment description information;
and processing the service to be processed through the target algorithm model.
It should be noted that, all or part of the steps in the methods of the above embodiments may be implemented by hardware related to instructions of a computer program, which may be stored in a computer-readable storage medium, which may include, but is not limited to: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.
The method, the apparatus, the storage medium, and the electronic device for screening algorithm models provided in the embodiments of the present application are described in detail above. The principle and the implementation of the present application are explained herein by applying specific examples, and the above description of the embodiments is only used to help understand the method and the core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (12)
1. A method for screening algorithm models is characterized by comprising the following steps:
acquiring device environment information, wherein the device environment information at least comprises: the service type, the equipment resource information and the application use information of the service to be processed;
acquiring corresponding environment description information according to the equipment environment information;
screening a target algorithm model from a preset model library based on the environment description information;
and processing the service to be processed through the target algorithm model.
2. The method for filtering algorithm models according to claim 1, wherein the obtaining of the corresponding environment description information according to the device environment information includes:
acquiring first description information of the service type, second description information of the equipment resource information and third description information of the application use information;
preprocessing the first description information, the second description information and the third description information;
and generating the environment description information according to the preprocessed first description information, second description information and third description information.
3. The algorithmic model screening method of claim 2, wherein the device resource information comprises resource amounts of a plurality of resources; acquiring second description information of the device resource information, including:
comparing the resource quantity with a corresponding preset threshold value to obtain a plurality of comparison results;
and generating second description information according to the comparison results.
4. The algorithmic model screening method of claim 2, wherein obtaining third description information of the application usage information comprises:
determining a target application currently used according to the application use information;
determining attribute information of the target application;
and determining third description information according to the attribute information.
5. The algorithm model screening method of claim 1, wherein the screening of the target algorithm model from a preset model library based on the environment description information comprises:
matching corresponding modeling indexes for the environment description information based on a preset classification algorithm;
and screening a target algorithm model from a plurality of sample algorithm models in a preset model library according to the modeling index.
6. The algorithm model screening method of claim 5, wherein the matching of the corresponding modeling index for the environment description information based on a preset classification algorithm comprises:
extracting a plurality of first keywords from the environment description information;
obtaining a modeling index matched with the first keyword according to a preset mapping relation, wherein the preset mapping relation comprises: and mapping relation between the sample keywords and the sample modeling indexes.
7. The algorithm model screening method according to claim 6, wherein the screening of the target algorithm model from a plurality of sample algorithm models in a preset model library according to the modeling index comprises:
acquiring the number of first keywords which are correspondingly matched with each modeling index;
determining the priority of the modeling index according to the number of the first keywords;
and screening a target algorithm model from the preset model library according to the priority of the modeling index.
8. The algorithm model screening method according to claim 5, wherein the screening of the target algorithm model from a plurality of sample algorithm models in a preset model library according to the modeling index comprises:
extracting a second keyword from the text description information corresponding to the modeling index;
determining a label corresponding to each sample algorithm model, and extracting a third key word from text description information corresponding to the label;
calculating the matching degree of the second key words and the third key words;
and screening a target algorithm model from the preset model library according to the matching degree.
9. The algorithm model screening method according to claim 5, wherein the screening of the target algorithm model from a plurality of sample algorithm models in a preset model library according to the modeling index comprises:
constructing a first word vector corresponding to the modeling index;
constructing a second word vector of the label corresponding to each sample algorithm model;
calculating a vector distance between the first word vector and the second word vector;
and screening a target algorithm model from the preset model library based on the vector distance.
10. An apparatus for screening an algorithm model, comprising:
a first obtaining module, configured to obtain device environment information, where the device environment information at least includes: the service type, the equipment resource information and the application use information of the service to be processed;
the second acquisition module is used for acquiring corresponding environment description information according to the equipment environment information;
the screening module is used for screening a target algorithm model from a preset model library based on the environment description information;
and the processing module is used for processing the service to be processed through the target algorithm model.
11. A storage medium having a computer program stored thereon, the computer program, when being executed by a processor, performing the steps of the method according to any of the claims 1-9.
12. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method according to any of claims 1-9 are implemented when the processor executes the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910282138.8A CN111796925A (en) | 2019-04-09 | 2019-04-09 | Method and device for screening algorithm model, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910282138.8A CN111796925A (en) | 2019-04-09 | 2019-04-09 | Method and device for screening algorithm model, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111796925A true CN111796925A (en) | 2020-10-20 |
Family
ID=72805325
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910282138.8A Withdrawn CN111796925A (en) | 2019-04-09 | 2019-04-09 | Method and device for screening algorithm model, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111796925A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113568740A (en) * | 2021-07-16 | 2021-10-29 | 开放智能机器（上海）有限公司 | Model aggregation method, system, device and medium based on federal learning |
| CN114528040A (en) * | 2022-01-26 | 2022-05-24 | 高新兴科技集团股份有限公司 | Environment self-adaption method, device, medium and roadside perception and calculation system |
| WO2022267066A1 (en) * | 2021-06-25 | 2022-12-29 | Oppo广东移动通信有限公司 | Model management method and communication device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105550746A (en) * | 2015-12-08 | 2016-05-04 | 北京旷视科技有限公司 | Training method and training device of machine learning model |
| US20180330002A1 (en) * | 2016-01-27 | 2018-11-15 | Alibaba Group Holding Limited | Service Processing Method, and Data Processing Method and Apparatus |
| CN109086680A (en) * | 2018-07-10 | 2018-12-25 | Oppo广东移动通信有限公司 | Image processing method, device, storage medium and electronic equipment |
| CN109165249A (en) * | 2018-08-07 | 2019-01-08 | 阿里巴巴集团控股有限公司 | Data processing model construction method, device, server and user terminal |
-
2019
- 2019-04-09 CN CN201910282138.8A patent/CN111796925A/en not_active Withdrawn
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105550746A (en) * | 2015-12-08 | 2016-05-04 | 北京旷视科技有限公司 | Training method and training device of machine learning model |
| US20180330002A1 (en) * | 2016-01-27 | 2018-11-15 | Alibaba Group Holding Limited | Service Processing Method, and Data Processing Method and Apparatus |
| CN109086680A (en) * | 2018-07-10 | 2018-12-25 | Oppo广东移动通信有限公司 | Image processing method, device, storage medium and electronic equipment |
| CN109165249A (en) * | 2018-08-07 | 2019-01-08 | 阿里巴巴集团控股有限公司 | Data processing model construction method, device, server and user terminal |
Non-Patent Citations (1)
| Title |
|---|
| 庞剑锋, 卜东波, 白硕: "基于向量空间模型的文本自动分类***的研究与实现", 计算机应用研究, no. 09, 28 September 2001 (2001-09-28) * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022267066A1 (en) * | 2021-06-25 | 2022-12-29 | Oppo广东移动通信有限公司 | Model management method and communication device |
| CN113568740A (en) * | 2021-07-16 | 2021-10-29 | 开放智能机器（上海）有限公司 | Model aggregation method, system, device and medium based on federal learning |
| CN114528040A (en) * | 2022-01-26 | 2022-05-24 | 高新兴科技集团股份有限公司 | Environment self-adaption method, device, medium and roadside perception and calculation system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8781991B2 (en) | Emotion recognition apparatus and method | |
| CN111797858A (en) | Model training method, behavior prediction method, device, storage medium and equipment | |
| CN111814475A (en) | User portrait construction method and device, storage medium and electronic equipment | |
| CN111796925A (en) | Method and device for screening algorithm model, storage medium and electronic equipment | |
| CN111797861A (en) | Information processing method, information processing apparatus, storage medium, and electronic device | |
| CN111798259A (en) | Application recommendation method and device, storage medium and electronic equipment | |
| CN111797854A (en) | Scene model establishing method and device, storage medium and electronic equipment | |
| CN111797870A (en) | Optimization method and device of algorithm model, storage medium and electronic equipment | |
| CN111797873A (en) | Scene recognition method and device, storage medium and electronic equipment | |
| CN111796926A (en) | Instruction execution method and device, storage medium and electronic equipment | |
| CN111798367A (en) | Image processing method, image processing device, storage medium and electronic equipment | |
| CN111797856B (en) | Modeling method and device, storage medium and electronic equipment | |
| CN111798019B (en) | Intention prediction method, intention prediction device, storage medium and electronic equipment | |
| CN111797874B (en) | Behavior prediction method and device, storage medium and electronic equipment | |
| CN111797867A (en) | System resource optimization method and device, storage medium and electronic equipment | |
| CN111797862A (en) | Task processing method and device, storage medium and electronic equipment | |
| CN111814812A (en) | Modeling method, modeling device, storage medium, electronic device and scene recognition method | |
| CN111796663B (en) | Scene recognition model updating method and device, storage medium and electronic equipment | |
| CN111797849A (en) | User activity identification method and device, storage medium and electronic equipment | |
| CN111797261A (en) | Feature extraction method and device, storage medium and electronic equipment | |
| CN111797986A (en) | Data processing method, data processing device, storage medium and electronic equipment | |
| CN114924876A (en) | Voiceprint recognition method and device based on distributed heterogeneous operation and storage medium | |
| CN111797869A (en) | Model training method and device, storage medium and electronic equipment | |
| CN111796924A (en) | Service processing method, device, storage medium and electronic equipment | |
| CN111797303A (en) | Information processing method, information processing apparatus, storage medium, and electronic 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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20201020 |