CN109242043A - Method and apparatus for generating information prediction model - Google Patents
Method and apparatus for generating information prediction model Download PDFInfo
- Publication number
- CN109242043A CN109242043A CN201811145079.1A CN201811145079A CN109242043A CN 109242043 A CN109242043 A CN 109242043A CN 201811145079 A CN201811145079 A CN 201811145079A CN 109242043 A CN109242043 A CN 109242043A
- Authority
- CN
- China
- Prior art keywords
- information
- sample
- data
- feature
- initial 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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"
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Entrepreneurship & Innovation (AREA)
- Evolutionary Biology (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the present application discloses the method and apparatus for generating information prediction model.One specific embodiment of this method includes: acquisition sample set, therefrom chooses sample, executes training step: the first data of sample, the second data being inputted initial model, obtain the first information and the second information;The first information, the second information are analyzed with the sample first information, the second information of sample respectively, determine the penalty values of the first information and the penalty values of the second information;Relationship between the first information and the second information is analyzed, determines the penalty values between the first information and the second information;According to the relationship weight between preset first information weight, the second information weight and the first information and the second information, using the weighted results of three's penalty values as total losses value, total losses value is compared with target value;Determine whether initial model trains completion according to comparison result;In response to determining that initial model training is completed, using initial model as information prediction model.
Description
Technical field
The invention relates to field of computer technology, and in particular to for generating the method and dress of information prediction model
It sets.
Background technique
Urban area, the urban area of narrow sense refer to the cell divided in city by its function (function).Broad sense is appreciated that
There is a kind of specific region structure system between closely coupled surrounding area for urban development and therewith.
Summary of the invention
The embodiment of the present application proposes the method and apparatus for generating information prediction model.
In a first aspect, the embodiment of the present application provides a kind of method for generating information prediction model, this method comprises:
Obtain sample set, wherein sample include the first data of sample and the second data of sample and with the first data of sample and sample the
The corresponding sample first information of two data and the second information of sample;Sample is chosen from above-mentioned sample set, and executes following instruction
Practice step: the first data of sample of the sample of selection and the second data of sample are inputted into initial model, obtains the first letter of sample
Second information of breath and sample;The first information is analyzed with the corresponding sample first information, determines the loss of the first information
Value;Second information is analyzed with corresponding the second information of sample, determines the penalty values of the second information;By the first information and
Relationship between two information is analyzed, and determines the penalty values between the first information and the second information;According to preset first letter
Relationship weight between breath weight, the second information weight and the first information and the second information, by the loss of the first information of sample
The weighting knot of penalty values between value, the penalty values of the second information of sample and the first information of sample and the second information of sample
Total losses value of the fruit as sample, and the total losses value of sample is compared with target value;It is determined just according to comparison result
Whether beginning model trains completion;In response to determining that initial model training is completed, using initial model as information prediction model.
In some embodiments, above-mentioned initial model includes fisrt feature extract layer, fisrt feature extraction network, the second spy
Sign extracts network, second feature extract layer and output layer;And the first data of sample and sample of the above-mentioned sample by selection
Two data input initial model, obtain the first information of sample and the second information of sample, comprising: by the sample of the sample of selection
First data input above-mentioned fisrt feature extract layer, generate fisrt feature and second feature;By fisrt feature generated and
Two features input above-mentioned fisrt feature respectively and extract network and above-mentioned second feature extraction network, obtain third feature and the 4th spy
Sign;The second data of sample of the sample of selection are inputted into above-mentioned second feature extract layer, generate fifth feature;By obtained
Three features, fourth feature and fifth feature generated input above-mentioned output layer, generate the first information and the second information.
In some embodiments, this method further include: in response to determining that initial model not complete by training, adjusts initial model
In relevant parameter, and choose sample again from above-mentioned sample set, use initial model adjusted as initial model,
Continue to execute above-mentioned training step.
Second aspect, the embodiment of the present application provide a kind of for generating the device of information prediction model, which includes:
Acquiring unit is configured to obtain sample set, wherein sample includes the first data of sample and the second data of sample and and sample
First data and the corresponding sample first information of the second data of sample and the second information of sample;Training unit is configured to from upper
It states and chooses sample in sample set, and execute following training step: by the first data of sample and sample second of the sample of selection
Data input initial model, obtain the first information of sample and the second information of sample;By the first information and corresponding sample
One information is analyzed, and determines the penalty values of the first information;Second information is analyzed with corresponding the second information of sample, really
The penalty values of fixed second information;Relationship between the first information and the second information is analyzed, determines the first information and second
Penalty values between information;According between preset first information weight, the second information weight and the first information and the second information
Relationship weight, by the first information of the penalty values of the first information of sample, the penalty values of the second information of sample and sample with
Total losses value of the weighted results of penalty values between second information of sample as sample, and by the total losses value of sample with
Target value is compared;Determine whether initial model trains completion according to comparison result;In response to determining that initial model training is complete
At using initial model as information prediction model.
In some embodiments, above-mentioned initial model includes fisrt feature extract layer, fisrt feature extraction network, the second spy
Sign extracts network, second feature extract layer and output layer;And above-mentioned training unit is further configured to: by the sample of selection
The first data of sample input above-mentioned fisrt feature extract layer, generate fisrt feature and second feature;It is special by generated first
Second feature of seeking peace inputs that above-mentioned fisrt feature extracts network and above-mentioned second feature extracts network respectively, obtain third feature and
Fourth feature;The second data of sample of the sample of selection are inputted into above-mentioned second feature extract layer, generate fifth feature;By gained
Third feature, fourth feature and the fifth feature generated arrived inputs above-mentioned output layer, generates the first information and the second information.
In some embodiments, device further include: adjustment unit is configured in response to determine that initial model is not trained
It completes, adjusts the relevant parameter in initial model, and choose sample again from above-mentioned sample set, using adjusted initial
Model continues to execute above-mentioned training step as initial model.
The third aspect, the embodiment of the present application provide a kind of method for generating information, comprising: obtain target object
Track data, wherein above-mentioned track data is as obtained from recording the track of above-mentioned target object;Based on above-mentioned track number
According to determining the first data and the second data;By above-mentioned first data and the input of above-mentioned second data using such as above-mentioned first aspect
In the information prediction model that method described in middle any embodiment generates, the first information and the second information are generated.
Fourth aspect, the embodiment of the present application provide a kind of for generating the device of information, comprising: acquiring unit is matched
It is set to the track data for obtaining target object, wherein above-mentioned track data is obtained and recording the track of above-mentioned target object
It arrives;Determination unit is configured to determine the first data and the second data based on above-mentioned track data;Generation unit is configured
The method as described in any embodiment in above-mentioned first aspect is used at by above-mentioned first data and the input of above-mentioned second data
In the information prediction model of generation, the first information and the second information are generated.
5th aspect, the embodiment of the present application provide a kind of electronic equipment, comprising: one or more processors;Storage dress
It sets, for storing one or more programs;When one or more programs are executed by one or more processors, so that one or more
A processor realizes the method as described in any embodiment in above-mentioned first aspect and the third aspect.
6th aspect, the embodiment of the present application provide a kind of computer-readable medium, are stored thereon with computer program,
In, it is realized as described in any embodiment in above-mentioned first aspect and the third aspect when which is executed by processor
Method.
Method and apparatus provided by the embodiments of the present application for generating information prediction model can by obtaining sample set
To choose sample therefrom to carry out the training of initial model.Wherein, sample include the first data of sample and the second data of sample with
And the sample first information corresponding with the first data of sample and the second data of sample and the second information of sample.In this way, by selection
The first data of sample and the second data of sample of sample input initial model, obtain the first information of sample and the second letter of sample
Breath.Later, according to the second information of the sample first information and sample, the penalty values and the second information of the first information can be determined respectively
Penalty values.Relationship between the first information and the second information is analyzed, is determined between the first information and the second information
Penalty values;Then, according to the relationship between preset first information weight, the second information weight and the first information and the second information
Weight, by the penalty values of the first information of sample, the penalty values of the second information of sample and the first information of sample and sample
Total losses value of the weighted results of penalty values between second information as sample, and can be by the total losses value and target of sample
Value is compared.Finally, can determine whether initial model trains completion according to comparison result.If it is determined that initial model is instructed
Practice and complete, so that it may which the initial model for completing training is as information prediction model.It can be used for giving birth to so as to obtain one kind
At the model of information.And facilitate the generating mode of abundant model.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is that one embodiment of the application can be applied to exemplary system architecture figure therein;
Fig. 2 is the flow chart according to one embodiment of the method for generating information prediction model of the application;
Fig. 3 is the signal according to an application scenarios of the method for generating information prediction model of the embodiment of the present application
Figure;
Fig. 4 is the structural schematic diagram for being used to generate one embodiment of the device of information prediction model according to the application;
Fig. 5 is the flow chart for being used to generate one embodiment of the method for information according to the application;
Fig. 6 is the structural schematic diagram for being used to generate one embodiment of the device of information according to the application;
Fig. 7 is adapted for the structural schematic diagram for the computer system for realizing the electronic equipment of the embodiment of the present application.
Specific embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, part relevant to related invention is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is shown can believe using the method for generating information prediction model of the embodiment of the present application, for generating
Cease the exemplary system architecture 100 of the device of prediction model, the method for generating information or the device for generating information.
As shown in Figure 1, system architecture 100 may include terminal 101,102, network 103,104 kimonos of database server
Business device 105.Network 103 is to provide communication link in terminal 101,102 between database server 104 and server 105
Medium.Network 103 may include various connection types, such as wired, wireless communication link or fiber optic cables etc..
User 110 can be used terminal 101,102 and be interacted by network 103 with server 105, to receive or send
Message etc..Various client applications can be installed, such as the application of model training class, information prediction class are answered in terminal 101,102
With, generate info class application, the application of shopping class, news category application and immediate communication tool etc..
Here terminal 101,102 can be hardware, be also possible to software.When terminal 101,102 is hardware, can be
Various electronic equipments, including but not limited to smart phone, tablet computer, E-book reader, pocket computer on knee and platform
Formula computer etc..When terminal 101,102 is software, may be mounted in above-mentioned cited electronic equipment.It can be real
Ready-made multiple softwares or software module (such as providing Distributed Services), also may be implemented into single software or software mould
Block.It is not specifically limited herein.
Database server 104 can be to provide the database server of various services.Such as it can in database server
To be stored with sample set.It include a large amount of sample in sample set.Wherein, sample may include the first data of sample and sample
Two data and the sample first information corresponding with the first data of sample and the second data of sample and the second information of sample.User
110 can choose sample from the sample set that database server 104 is stored by terminal 101,102.
Server 105 can be to provide the server of various services, such as the various applications to showing in terminal 101,102
The background server supported is provided.Background server can use the sample in the sample set of the transmission of terminal 101,102, to initial
Model is trained, and training result (such as the information prediction model generated) can be sent to terminal 101,102.In this way, with
Family can obtain the information of push by terminal 101,102.
Here database server 104 and server 105 can be hardware, be also possible to software.When they are hardware
When, the distributed server cluster of multiple server compositions may be implemented into, individual server also may be implemented into.When they are
When software, multiple softwares or software module (such as providing Distributed Services) may be implemented into, also may be implemented into single
Software or software module.It is not specifically limited herein.
It should be noted that for generating the method for information prediction model or for generating provided by the embodiment of the present application
The method of information is generally executed by server 105.Correspondingly, for generating the device of information prediction model or for generating information
Device be generally also disposed in server 105.
It should be pointed out that being in the case where the correlation function of database server 104 may be implemented in server 105
Database server 104 can be not provided in system framework 100.
It should be understood that the number of terminal, network, database server and server in Fig. 1 is only schematical.Root
It factually now needs, can have any number of terminal, network, database server and server.
With continued reference to Fig. 2, it illustrates an implementations according to the method for generating information prediction model of the application
The process 200 of example.The method for being used to generate information prediction model may comprise steps of:
Step 201, sample set is obtained.
In the present embodiment, for generating executing subject (such as the server shown in FIG. 1 of the method for information prediction model
105) sample set can be obtained in several ways.For example, executing subject can pass through wired connection mode or wireless connection
Mode, from acquisition sample set in database server (such as database server 104 shown in FIG. 1).For another example user can be with
Sample is collected by terminal (such as terminal shown in FIG. 1 101,102).In this way, executing subject can receive collected by terminal
Sample, and these samples are stored in local, to obtain sample set.
It herein, may include at least one sample in sample set.Wherein, sample includes the first data of sample and sample
Two data and the sample first information corresponding with the first data of sample and the second data of sample and the second information of sample.Sample
The desired output of one data and sample the second data entry information prediction model is the second information of the sample first information and sample.
In the present embodiment, above-mentioned first data may include on the influential data of number inflow and outflow.Specifically, on
Stating the first data can be the track data of taxi of target object seating, be also possible to the shared bicycle that target object uses
Track data, be also possible to target object and swipe the card in subway data out of the station, can also be other to number inflow and outflow
Influential data, this is not restricted.Above-mentioned target object can occur from the people of predeterminable area.Second data can wrap
Include surface data influential on number inflow and outflow.Surface data can be festivals or holidays data, be also possible to day
Gas meteorological data can be temperature air speed data, can also be that other, on the influential data of number inflow and outflow, do not make herein
Limitation.
The above-mentioned first information can be within the specified period after above-mentioned preset time period into predeterminable area
The quantity of the quantity of people and the people for leaving predeterminable area.Above-mentioned second information can be specified after above-mentioned preset time period
Period in enter the people of predeterminable area from the information in specified region and leave the people of predeterminable area and be moved to specified region
Information.As an example, the above-mentioned first information may include the area within the specified period after above-mentioned preset time period
Flow of the people passes in and out data in domain, and the second information may include the region within the specified period after above-mentioned preset time period
Between flow of the people shift data.
It is understood that the first data of sample can be obtained by recording the track data of user.The number of sample first
According to can also be obtained by artificial setting in advance.The first data of sample can also be through executing subject or other equipment execution
Obtained from certain setting program.Above-mentioned track data may include the track data for the taxi that user takes, and also may include
The track data for the shared bicycle that user uses also may include that user swipes the card data out of the station in subway, can also include
Other track datas, this is not restricted.
Step 202, sample is chosen from sample set.
In the present embodiment, sample is chosen in the sample set that executing subject can be obtained from step 201, to execute step
203 to step 208 training step.Wherein, the selection mode of sample and selection quantity are not intended to limit in this application.Such as it can
To be to randomly select at least one sample.
Step 203, the first data of sample of the sample of selection and the second data of sample are inputted into initial model, obtains sample
The first information and sample the second information.
In the present embodiment, above-mentioned initial model is for characterizing the first data of sample and the second data of sample and the first information
With the corresponding relationship of the second information.The model that initial model can be unbred model or training is not completed.
In the present embodiment, above-mentioned initial model is also possible to the existing various minds created based on machine learning techniques
Through network model.The neural network model can have existing various neural network structures (such as convolutional neural networks, circulation
Neural network etc.).The storage location of initial model is in this application with no restriction.
In some optional implementations of the present embodiment, above-mentioned initial model includes fisrt feature extract layer, first
Feature extraction network, second feature extract network, second feature extract layer and output layer;And the sample of the above-mentioned sample by selection
This first data and the second data of sample input initial model, obtain the first information of sample and the second information of sample, comprising:
The first data of sample of the sample of selection and the second data of sample are inputted into above-mentioned fisrt feature extract layer, generate fisrt feature and
Second feature;Fisrt feature generated and above-mentioned second feature are inputted into above-mentioned fisrt feature respectively and extract network and above-mentioned the
Two feature extraction networks, obtain third feature and fourth feature;The second data of sample of the sample of selection are inputted above-mentioned second
Feature extraction layer generates fifth feature;Obtained third feature, fourth feature and fifth feature generated are inputted above-mentioned
Output layer generates the first information and the second information.
Above-mentioned fisrt feature extract layer is used to characterize the corresponding relationship of the first data Yu fisrt feature and second feature.It is above-mentioned
Fisrt feature can be the quantity of the people for entering predeterminable area within a preset period of time of executing subject statistics and leave preset areas
The quantity of the people in domain.Above-mentioned second feature can be executing subject statistics within a preset period of time by a region to another
The quantity of the people in region.Above-mentioned fisrt feature, which extracts network, can be the various function for having and extracting third feature from fisrt feature
The neural network of energy.Above-mentioned second feature, which extracts network, can be the various functions of having and extract fourth feature from second feature
Neural network.For example, fisrt feature, which extracts network, can be the Recognition with Recurrent Neural Network including at least one layer of neural network.First
Feature extraction network can also be the convolutional neural networks including at least one layer of neural network.Second feature is extracted network and be can be
Recognition with Recurrent Neural Network including at least one layer of neural network.Second feature is extracted network and be can also be including at least one layer of nerve net
The convolutional neural networks of network.
Above-mentioned second feature extract layer is used to characterize the corresponding relationship of the second data and fifth feature.For example, second feature
Extract layer can be the Recognition with Recurrent Neural Network including at least one layer of neural network.Above-mentioned output layer is for characterizing third feature, the
The corresponding relationship of four features and fifth feature and the first information and the second information.Optionally, above-mentioned output layer can be depth mind
Through network.
Step 204, the first information is analyzed with the corresponding sample first information, determines the penalty values of the first information.
In the present embodiment, executing subject can be by the first information of sample and first data of sample and the number of sample second
It is analyzed according to the corresponding sample first information, may thereby determine that the penalty values of the first information.Such as it can be by the first information
With the corresponding sample first information as parameter, input in specified loss function (loss function), so as to calculate
Obtain penalty values between the two.
In the present embodiment, loss function is usually for estimating the predicted value of model (such as first information) and true value
The inconsistent degree of (such as sample first information).It is a non-negative real-valued function.Under normal circumstances, loss function is smaller, mould
The robustness of type is better.Loss function can be arranged according to actual needs.
Step 205, the second information is analyzed with corresponding the second information of sample, determines the penalty values of the second information.
In the present embodiment, executing subject can also be by the second information of sample and first data of sample and sample second
Corresponding the second information of sample of data is analyzed, and may thereby determine that the second information loss value.It may refer to retouch in step 204
The correlation technique stated, details are not described herein again.
Step 206, the relationship between the first information and the second information is analyzed, determines the first information and the second information
Between penalty values.
In the present embodiment, executing subject can also be divided the second information of the first information of sample and the sample
Analysis, may thereby determine that the penalty values between the first information and the second information.It may refer to related side described in step 204
Method, details are not described herein again.
Step 207, according between preset first information weight, the second information weight and the first information and the second information
Relationship weight, by the penalty values of the first information of sample, the penalty values of the second information of sample and the first information of sample and sample
Total losses value of the weighted results of penalty values between this second information as sample, and by the total losses value and mesh of sample
Scale value is compared.
In the present embodiment, executing subject can be according to preset first information weight, the second information weight and the first letter
Cease the relationship weight between the second information, to the penalty values of the first information of same sample, the penalty values of the second information and the
Penalty values between one information and the second information are weighted processing.Preset first information weight is first information penalty values
Weight.Preset second information weight is the weight of the second information loss value.Between the preset first information and the second information
Relationship weight.Later, executing subject can be using the above-mentioned weighted sum of same sample as the total losses value of the sample.And it can be with
The total losses value of the sample of selection is compared with target value.
In the present embodiment, between preset first information weight, the second information weight and the first information and the second information
Relationship weight can be arranged according to the actual situation.And target value can be generally used for indicating predicted value (the i.e. first information, the
Two information) and true value (the sample first information, the second information of sample) between inconsistent degree ideal situation.That is,
When total losses value is less than target value, it is believed that predicted value nearly or approximately true value.Target value can be come according to actual needs
Setting.
It should be noted that if choosing in step 202 has multiple (at least two) samples, then executing subject can will be each
The total losses value of sample is compared with target value respectively.It may thereby determine that whether the total losses value of each sample reaches target
Value.
In some optional implementations of the present embodiment, preset first information weight and preset second information weight
Weight can be respectively a fixed weight value.
It is understood that using the method for weight, by first information penalty values, the penalty values of the second information and first
Penalty values between information and the second information blend, so as to adjust Optimized model.The information prediction that this method training obtains
Model can effectively improve the robustness of information prediction.
Step 208, determine whether initial model trains completion according to comparison result.
In the present embodiment, according to the comparison result in step 207, executing subject can determine whether initial model trains
It completes.As an example, reaching target in the total losses value of each sample if choosing in step 202 has multiple samples
In the case where value, executing subject can determine that initial model training is completed.As another example, executing subject can count total damage
The sample that mistake value reaches target value accounts for the ratio of the sample of selection.And reach default sample proportion (such as 95%) in the ratio,
It can determine that initial model training is completed.
In the present embodiment, if executing subject determines that initial model has trained completion, step 209 can be continued to execute.
If executing subject determines that initial model not complete by training, the relevant parameter in adjustable initial model.It is alternatively possible to adopt
With the weight in backpropagation techniques modification initial model in each neural net layer.And it can be with return step 202, from sample set
In choose sample again.So as to continue to execute above-mentioned training step.
It should be noted that selection mode here does not also limit in this application.Such as a large amount of samples are concentrated in sample
In the case where this, executing subject can therefrom choose the sample of unselected mistake.
Step 209, in response to determining that initial model training is completed, using initial model as information prediction model.
In the present embodiment, if executing subject determines that initial model training is completed, which (can be trained
The initial model of completion) it is used as information prediction model.
Optionally, the information prediction model of generation can be stored in local by executing subject, can also send it to end
End or database server.
With further reference to the application that Fig. 3, Fig. 3 are according to the method for generating information prediction model of the present embodiment
The schematic diagram of scene.In the application scenarios 300 of Fig. 3, model training class can be installed in terminal 31 used by a user and answered
With.When user opens the application, and after uploading the store path of sample set or sample set, the clothes of back-office support are provided to the application
Business device 32 can run the program for generating information prediction model, comprising:
It is possible, firstly, to obtain sample set.Wherein, the sample in sample set may include the first data of sample and sample second
Data 321 and the sample first information 322 corresponding with the first data of sample and sample the second data 321 and the second information of sample
323.Later, sample can be chosen from sample set, and executes following training step: by the number of sample first of the sample of selection
Initial model 320 is inputted according to the second data of sample 321, obtains 322 ' of the first information of sample and the second information of sample
323 ';322 ' of the first information of sample is analyzed with the corresponding sample first information 322, determines first information penalty values
324;Second information, 323 ' is analyzed with corresponding the second information of sample 323, determines the second information loss value 325;By
Relationship between 323 ' of one information, 322 ' and the second information is analyzed, and determines the penalty values between the first information and the second information
326;It, will according to the relationship weight between preset first information weight, the second information weight and the first information and the second information
The of the penalty values 324 of the first information of sample, the first information of the penalty values 325 of the second information of sample and sample and sample
Total losses value 327 of the weighted results of penalty values 326 between two information as sample, and by the total losses value 327 of sample
It is compared with target value;Determine whether initial model 320 trains completion according to comparison result;In response to determining initial model
320 training are completed, and regard initial model 320 as 320 ' of information prediction model.
At this point, server 32 can also send the prompt information for being used to indicate model training and completing to terminal 31.The prompt
Information can be voice and/or text information.In this way, user can get information prediction model in preset storage location.
Method in the present embodiment for generating information prediction model can therefrom choose sample by obtaining sample set
To carry out the training of initial model.Wherein, sample includes the first data of sample and the second data of sample and counts with sample first
According to the second information of the sample first information corresponding with the second data of sample and sample.In this way, by the sample first of the sample of selection
Data and the second data of sample input initial model, obtain the first information of sample and the second information of sample.Later, according to sample
The second information of this first information and sample can determine the penalty values of the first information and the penalty values of the second information respectively.By
Relationship between one information and the second information is analyzed, and determines the penalty values between the first information and the second information;Then, root
According to the relationship weight between preset first information weight, the second information weight and the first information and the second information, by sample
Between the penalty values of the first information, the penalty values of the second information of sample and the first information of sample and the second information of sample
Total losses value of the weighted results of penalty values as sample, and the total losses value of sample can be compared with target value.Most
Afterwards, it can determine whether initial model trains completion according to comparison result.If it is determined that initial model training is completed, so that it may
The initial model that training is completed is as information prediction model.So as to obtain a kind of model that can be used for generating information.
And facilitate the generating mode of abundant model.
With further reference to Fig. 4, as the realization to method shown in above-mentioned each figure, this application provides one kind for generating letter
Cease one embodiment of the device of prediction model.The Installation practice is corresponding with embodiment of the method shown in Fig. 2, device tool
Body can be applied in various electronic equipments.
As shown in figure 4, the present embodiment may include: acquiring unit for generating the device 400 of information prediction model
401, it is configured to obtain sample set, wherein the sample in above-mentioned sample set includes the second data of the first data of sample and sample
And the sample first information corresponding with the first data of sample and the second data of sample and the second information of sample;Training unit 402,
It is configured to choose sample from above-mentioned sample set, and executes following training step: by the number of sample first of the sample of selection
Initial model is inputted according to the second data of sample, obtains the first information of sample and the second information of sample;By the first information with
The corresponding sample first information is analyzed, and determines the penalty values of the first information;Second information is believed with corresponding sample second
Breath is analyzed, and determines the penalty values of the second information;Relationship between the first information and the second information is analyzed, determines
Penalty values between one information and the second information;According to preset first information weight, the second information weight and the first information with
Relationship weight between second information, by the penalty values of the first information of sample, the penalty values and sample of the second information of sample
The first information and sample the second information between penalty values total losses value of the weighted results as sample, and by sample
Total losses value be compared with target value;Determine whether initial model trains completion according to comparison result;In response to determining just
Beginning model training is completed, using initial model as information prediction model.
In some optional implementations of the present embodiment, above-mentioned initial model may include fisrt feature extract layer,
Fisrt feature extracts network, second feature extracts network, second feature extract layer and output layer;And above-mentioned training unit 402
It can also be further configured to: the first data of sample and the second data of sample the input introductory die of the above-mentioned sample by selection
Type obtains the first information of sample and the second information of sample, comprising: by the first data of sample and sample of the sample of selection
Two data input above-mentioned fisrt feature extract layer, generate fisrt feature and second feature;By above-mentioned fisrt feature and above-mentioned second
Feature inputs above-mentioned fisrt feature respectively and extracts network and above-mentioned second feature extraction network, obtains third feature and the 4th spy
Sign;The second data of sample of the sample of selection are inputted into above-mentioned second feature extract layer, generate fifth feature;By obtained
Three features, fourth feature and fifth feature generated input above-mentioned output layer, generate the first information and the second information.
Optionally, which can also include: adjustment unit 403, be configured in response to determine that initial model is not instructed
Practice and complete, adjusts the relevant parameter in initial model, and choose sample again from sample set, continue to execute training step.
It is understood that all units recorded in the device 400 and each step phase in the method with reference to Fig. 2 description
It is corresponding.The feature above with respect to method description and the beneficial effect of generation are equally applicable to device 400 and wherein include as a result,
Unit, details are not described herein.
Fig. 5 is referred to, it illustrates provided by the present application for generating the process of one embodiment of the method for information
500.The method for being used to generate information may comprise steps of:
Step 501, the track data of target object is obtained.
It in the present embodiment, can be with for generating the executing subject (such as server 105 shown in FIG. 1) of the method for information
Obtain the track data of target object in several ways.For example, executing subject can be by wired connection mode or wireless
Connection type is stored in track number therein from obtaining in database server (such as database server 104 shown in FIG. 1)
According to.For another example executing subject also can receive the rail of terminal (such as terminal shown in FIG. 1 101,102) or other equipment acquisition
Mark data.
In the present embodiment, above-mentioned target object can occur from the people of predeterminable area, and above-mentioned track data can be
The track data for the taxi that above-mentioned target object is taken, the track data for the shared bicycle that above-mentioned target object uses are above-mentioned
Target object is swiped the card data etc. out of the station in subway.Above-mentioned predeterminable area can be region specified according to actual needs, make
For example, above-mentioned predeterminable area can be the division result of the map area obtained from road net data.
Step 502, it is based on track data, determines the first data and the second data.
In the present embodiment, above-mentioned first data may include on the influential data of number inflow and outflow.Specifically,
One data can be the quantity of the people for entering predeterminable area within a preset period of time of executing subject statistics and leave predeterminable area
People quantity.Above-mentioned second data can be executing subject statistics within a preset period of time by a region to another area
The quantity of the people in domain.Second data may include surface data influential on number inflow and outflow.Surface data
It can be festivals or holidays data, be also possible to weather meteorological data, can be temperature air speed data, can also be other to number stream
Enter and flow out influential data, this is not restricted.
In the present embodiment, after the track data for obtaining target object, executing subject can determine the first data and
Two data.Executing subject can be by map partitioning at many irregular enclosed regions with latitude and longitude information, a rail
Mark may include time point and the sequence that longitude and latitude point is constituted, and count the variation of each point, it can be deduced that the first data, according to rail
Temporal information in mark data, corresponding second data of the available temporal information.
Step 503, the first information and the second letter will in the first data and the second data entry information prediction model, be generated
Breath.
In the present embodiment, executing subject can be by the first data obtained in step 502 and the second data entry information
In prediction model, to generate the first information and the second information.
In the present embodiment, above- mentioned information prediction model can be a pre-stored mapping table.The correspondence is closed
It is the corresponding relationship that can store multiple the first data of sample and sample the second data and the first information and the second information in table.
The mapping table can be technical staff be based on to a large amount of the first data of sample and the second data of sample and the first information and
The statistics of second information and formulate.Above-mentioned executing subject can be by the first data of target and the second data and above-mentioned corresponding relationship
The first data and the second data in table are compared.If first data and the second data and mesh in the mapping table
It marks the first data and the second data is same or similar, then it will be corresponding to first data and the second data in the mapping table
The first information and the second information as above-mentioned the first data of target and the second data of the first information and the second information.
In the present embodiment, information prediction model can also be using the method as described in above-mentioned Fig. 2 embodiment and give birth to
At.Specific generating process may refer to the associated description of Fig. 2 embodiment, and details are not described herein.
With continued reference to Fig. 6, as the realization to method shown in above-mentioned Fig. 5, this application provides one kind for generating information
Device one embodiment.The Installation practice is corresponding with embodiment of the method shown in fig. 5, which can specifically apply
In various electronic equipments.
As shown in fig. 6, the present embodiment may include: acquiring unit 601 for generating the device 600 of information, it is configured
At the track data for obtaining target object, wherein above-mentioned track data is obtained and recording the track of above-mentioned target object
's;Determination unit 602 is configured to determine the first data and the second data based on above-mentioned track data;Generation unit 603, quilt
It is configured to the letter for generating above-mentioned first data and the input of above-mentioned second data using the method as described in above-mentioned Fig. 2 embodiment
It ceases in prediction model, generates the first information and the second information.
It is understood that all units recorded in the device 600 and each step phase in the method with reference to Fig. 5 description
It is corresponding.The feature above with respect to method description and the beneficial effect of generation are equally applicable to device 600 and wherein include as a result,
Unit, details are not described herein.
Below with reference to Fig. 7, it illustrates the computer systems 700 for the electronic equipment for being suitable for being used to realize the embodiment of the present application
Structural schematic diagram.Electronic equipment shown in Fig. 7 is only an example, function to the embodiment of the present application and should not use model
Shroud carrys out any restrictions.
As shown in fig. 7, computer system 700 includes central processing unit (CPU) 701, it can be read-only according to being stored in
Program in memory (ROM) 702 or be loaded into the program in random access storage device (RAM) 703 from storage section 708 and
Execute various movements appropriate and processing.In RAM 703, also it is stored with system 700 and operates required various programs and data.
CPU 701, ROM 702 and RAM 703 are connected with each other by bus 704.Input/output (I/O) interface 705 is also connected to always
Line 704.
I/O interface 705 is connected to lower component: the importation 706 including touch screen, keyboard, mouse etc.;Including such as
The output par, c 707 of cathode-ray tube (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.;Storage unit including hard disk etc.
Divide 708;And the communications portion 709 of the network interface card including LAN card, modem etc..Communications portion 709 via
The network of such as internet executes communication process.Driver 710 is also connected to I/O interface 705 as needed.Detachable media
711, such as disk, CD, magneto-optic disk, semiconductor memory etc., are mounted on as needed on driver 710, in order to from
The computer program read thereon is mounted into storage section 708 as needed.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communications portion 709, and/or from detachable media
711 are mounted.When the computer program is executed by central processing unit (CPU) 701, limited in execution the present processes
Above-mentioned function.It should be noted that the computer-readable medium of the application can be computer-readable signal media or calculating
Machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be-- but it is unlimited
In system, device or the device of --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, or any above combination.It calculates
The more specific example of machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, portable of one or more conducting wires
Formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory
(EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or
The above-mentioned any appropriate combination of person.In this application, computer-readable medium, which can be, any includes or storage program has
Shape medium, the program can be commanded execution system, device or device use or in connection.And in the application
In, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal, wherein
Carry computer-readable program code.The data-signal of this propagation can take various forms, including but not limited to electric
Magnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer-readable and deposit
Any computer-readable medium other than storage media, the computer-readable medium can send, propagate or transmit for by referring to
Enable execution system, device or device use or program in connection.The program for including on computer-readable medium
Code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. or above-mentioned times
The suitable combination of meaning.
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the application, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard
The mode of part is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor packet
Include acquiring unit and training unit.For another example also can be described as: a kind of processor includes acquiring unit, determination unit, generation
Unit.Wherein, the title of these units does not constitute the restriction to the unit itself under certain conditions, for example, acquiring unit
It is also described as " obtaining the unit of sample set ".
As on the other hand, present invention also provides a kind of computer-readable medium, which be can be
Included in electronic equipment described in above-described embodiment;It is also possible to individualism, and without in the supplying electronic equipment.
Above-mentioned computer-readable medium carries one or more program, when said one or multiple programs are held by the electronic equipment
When row, so that the electronic equipment: obtain sample set, wherein sample include the first data of sample and the second data of sample and with
The first data of sample and the corresponding sample first information of the second data of sample and the second information of sample;It is chosen from above-mentioned sample set
Sample, and execute following training step: the first data of sample of the sample of selection and the second data of sample are inputted into introductory die
Type obtains the first information of sample and the second information of sample;The first information is analyzed with the corresponding sample first information,
Determine the penalty values of the first information;Second information is analyzed with corresponding the second information of sample, determines the damage of the second information
Mistake value;Relationship between the first information and the second information is analyzed, determines the loss between the first information and the second information
Value;According to the relationship weight between preset first information weight, the second information weight and the first information and the second information, by sample
Second information of this penalty values of the first information, the penalty values of the second information of sample and the first information of sample and sample it
Between penalty values total losses value of the weighted results as sample, and the total losses value of sample is compared with target value;
Determine whether initial model trains completion according to comparison result;In response to determining that initial model training is completed, initial model is made
For information prediction model.
In addition, when said one or multiple programs are executed by the electronic equipment, it is also possible that the electronic equipment: obtaining
Take the track data of target object, wherein above-mentioned track data is as obtained from recording the track of above-mentioned target object;Base
In above-mentioned track data, the first data and the second data are determined;Above-mentioned first data and above-mentioned second data entry information are pre-
It surveys in model, generates the first information and the second information.Wherein, information prediction model can be using as the various embodiments described above are retouched
That states generates for generating the method for information prediction model.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from foregoing invention design, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed herein
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (10)
1. a kind of method for generating information prediction model, comprising:
Obtain sample set, wherein sample include the first data of sample and the second data of sample and with the first data of sample and sample
The corresponding sample first information of this second data and the second information of sample;
Choose sample from the sample set, and execute following training step: by the first data of sample of the sample of selection and
The second data of sample input initial model, obtain the first information of sample and the second information of sample;By the first information with it is corresponding
The sample first information analyzed, determine the penalty values of the first information;By the second information and corresponding the second information of sample into
Row analysis, determines the penalty values of the second information;Relationship between the first information and the second information is analyzed, determines the first letter
Penalty values between breath and the second information;According to preset first information weight, the second information weight and the first information and second
Relationship weight between information, by the of the penalty values of the first information of sample, the penalty values of the second information of sample and sample
Total losses value of the weighted results of penalty values between one information and the second information of sample as sample, and by the total of sample
Penalty values are compared with target value;Determine whether initial model trains completion according to comparison result;In response to determining introductory die
Type training is completed, using initial model as information prediction model.
2. according to the method described in claim 1, wherein, the initial model includes that fisrt feature extract layer, fisrt feature mention
Network, second feature is taken to extract network, second feature extract layer and output layer;And
The first data of sample and the second data of sample of the sample by selection input initial model, obtain the first letter of sample
Second information of breath and sample, comprising:
The first data of sample of the sample of selection are inputted into the fisrt feature extract layer, generate fisrt feature and second feature;
Fisrt feature generated and second feature are inputted into the fisrt feature extraction network respectively and the second feature mentions
Network is taken, third feature and fourth feature are obtained;
The second data of sample of the sample of selection are inputted into the second feature extract layer, generate fifth feature;
Obtained third feature, fourth feature and fifth feature generated are inputted into the output layer, generate the first information
With the second information.
3. method described in one of -2 according to claim 1, wherein the method also includes:
In response to determining that initial model not complete by training, adjusts the relevant parameter in initial model, and from the sample set
Again sample is chosen, uses initial model adjusted as initial model, continues to execute the training step.
4. a kind of for generating the device of information prediction model, comprising:
Acquiring unit is configured to obtain sample set, wherein sample include the first data of sample and the second data of sample and with
The first data of sample and the corresponding sample first information of the second data of sample and the second information of sample;
Training unit is configured to choose sample from the sample set, and executes following training step: by the sample of selection
The first data of sample and the second data of sample input initial model, obtain the first information of sample and the second information of sample;
The first information is analyzed with the corresponding sample first information, determines the penalty values of the first information;By the second information with it is corresponding
The second information of sample analyzed, determine the penalty values of the second information;By the relationship between the first information and the second information into
Row analysis, determines the penalty values between the first information and the second information;According to preset first information weight, the second information weight
Relationship weight between the first information and the second information, by the penalty values of the first information of sample, the second information of sample
Total losses of the weighted results of penalty values between second information of the first information and sample of penalty values and sample as sample
Value, and the total losses value of sample is compared with target value;Determine whether initial model trains completion according to comparison result;
In response to determining that initial model training is completed, using initial model as information prediction model.
5. device according to claim 4, wherein the initial model includes that fisrt feature extract layer, fisrt feature mention
Network, second feature is taken to extract network, second feature extract layer and output layer;And
The training unit is further configured to:
The first data of sample of the sample of selection are inputted into the fisrt feature extract layer, generate fisrt feature and second feature;
Fisrt feature generated and second feature are inputted into the fisrt feature extraction network respectively and the second feature mentions
Network is taken, third feature and fourth feature are obtained;
The second data of sample of the sample of selection are inputted into the second feature extract layer, generate fifth feature;
Obtained third feature, fourth feature and fifth feature generated are inputted into the output layer, generate the first information
With the second information.
6. the device according to one of claim 4-5, wherein described device further include:
Adjustment unit is configured in response to determine that initial model not complete by training, adjusts the relevant parameter in initial model, with
And sample is chosen again from the sample set, use initial model adjusted as initial model, continues to execute the instruction
Practice step.
7. a kind of method for generating information, comprising:
Obtain the track data of target object, wherein the track data is obtained and recording the track of the target object
It arrives;
Based on the track data, the first data and the second data are determined;
The information that first data and second data input are generated using the method as described in one of claim 1-3
In prediction model, the first information and the second information are generated.
8. a kind of for generating the device of information, comprising:
Acquiring unit is configured to obtain the track data of target object, wherein the track data is by recording the mesh
Obtained from the track for marking object;
Determination unit is configured to determine the first data and the second data based on the track data;
Generation unit is configured to first data and second data input using such as one of claim 1-3 institute
In the information prediction model that the method stated generates, the first information and the second information are generated.
9. a kind of electronic equipment, comprising:
One or more processors;
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
The now method as described in any in claim 1-3,7.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
The method as described in any in claim 1-3,7 is realized when execution.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811145079.1A CN109242043A (en) | 2018-09-29 | 2018-09-29 | Method and apparatus for generating information prediction model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811145079.1A CN109242043A (en) | 2018-09-29 | 2018-09-29 | Method and apparatus for generating information prediction model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109242043A true CN109242043A (en) | 2019-01-18 |
Family
ID=65054532
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811145079.1A Pending CN109242043A (en) | 2018-09-29 | 2018-09-29 | Method and apparatus for generating information prediction model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109242043A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109902872A (en) * | 2019-02-28 | 2019-06-18 | 北京首钢股份有限公司 | A kind of method and apparatus that performance prediction model is combined with research and development of products |
CN110175544A (en) * | 2019-05-14 | 2019-08-27 | 广州虎牙信息科技有限公司 | Construction method, device, electronic equipment and the storage medium of object module |
CN111709248A (en) * | 2020-05-28 | 2020-09-25 | 北京百度网讯科技有限公司 | Training method and device of text generation model and electronic equipment |
CN112396445A (en) * | 2019-08-16 | 2021-02-23 | 京东数字科技控股有限公司 | Method and device for identifying user identity information |
CN112784168A (en) * | 2021-02-25 | 2021-05-11 | 平安健康保险股份有限公司 | Information push model training method and device, and information push method and device |
CN114781766A (en) * | 2022-06-22 | 2022-07-22 | 长江水利委员会长江科学院 | Hydrological information prediction method, device, equipment and storage medium for hydrological site |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104346607A (en) * | 2014-11-06 | 2015-02-11 | 上海电机学院 | Face recognition method based on convolutional neural network |
CN106022317A (en) * | 2016-06-27 | 2016-10-12 | 北京小米移动软件有限公司 | Face identification method and apparatus |
WO2017031856A1 (en) * | 2015-08-25 | 2017-03-02 | 百度在线网络技术(北京)有限公司 | Information prediction method and device |
CN108197618A (en) * | 2018-04-08 | 2018-06-22 | 百度在线网络技术(北京)有限公司 | For generating the method and apparatus of Face datection model |
CN108520324A (en) * | 2018-04-13 | 2018-09-11 | 北京京东金融科技控股有限公司 | Method and apparatus for generating information |
-
2018
- 2018-09-29 CN CN201811145079.1A patent/CN109242043A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104346607A (en) * | 2014-11-06 | 2015-02-11 | 上海电机学院 | Face recognition method based on convolutional neural network |
WO2017031856A1 (en) * | 2015-08-25 | 2017-03-02 | 百度在线网络技术(北京)有限公司 | Information prediction method and device |
CN106022317A (en) * | 2016-06-27 | 2016-10-12 | 北京小米移动软件有限公司 | Face identification method and apparatus |
CN108197618A (en) * | 2018-04-08 | 2018-06-22 | 百度在线网络技术(北京)有限公司 | For generating the method and apparatus of Face datection model |
CN108520324A (en) * | 2018-04-13 | 2018-09-11 | 北京京东金融科技控股有限公司 | Method and apparatus for generating information |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109902872A (en) * | 2019-02-28 | 2019-06-18 | 北京首钢股份有限公司 | A kind of method and apparatus that performance prediction model is combined with research and development of products |
CN109902872B (en) * | 2019-02-28 | 2021-10-01 | 北京首钢股份有限公司 | Method and device for combining performance prediction model with product research and development |
CN110175544A (en) * | 2019-05-14 | 2019-08-27 | 广州虎牙信息科技有限公司 | Construction method, device, electronic equipment and the storage medium of object module |
CN112396445A (en) * | 2019-08-16 | 2021-02-23 | 京东数字科技控股有限公司 | Method and device for identifying user identity information |
CN111709248A (en) * | 2020-05-28 | 2020-09-25 | 北京百度网讯科技有限公司 | Training method and device of text generation model and electronic equipment |
CN112784168A (en) * | 2021-02-25 | 2021-05-11 | 平安健康保险股份有限公司 | Information push model training method and device, and information push method and device |
CN112784168B (en) * | 2021-02-25 | 2023-08-22 | 平安健康保险股份有限公司 | Information push model training method and device, information push method and device |
CN114781766A (en) * | 2022-06-22 | 2022-07-22 | 长江水利委员会长江科学院 | Hydrological information prediction method, device, equipment and storage medium for hydrological site |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109242043A (en) | Method and apparatus for generating information prediction model | |
CN109460514A (en) | Method and apparatus for pushed information | |
CN104737523B (en) | The situational model in mobile device is managed by assigning for the situation label of data clustering | |
CN107908789A (en) | Method and apparatus for generating information | |
CN109460513A (en) | Method and apparatus for generating clicking rate prediction model | |
CN107633218A (en) | Method and apparatus for generating image | |
CN107644209A (en) | Method for detecting human face and device | |
CN108898185A (en) | Method and apparatus for generating image recognition model | |
CN107633317A (en) | Establish stroke planning model and cook up the method, apparatus of parade journey | |
CN107590482A (en) | information generating method and device | |
CN107908740A (en) | Information output method and device | |
CN109145784A (en) | Method and apparatus for handling video | |
CN109740018A (en) | Method and apparatus for generating video tab model | |
CN109165573A (en) | Method and apparatus for extracting video feature vector | |
CN108932220A (en) | article generation method and device | |
CN108197618A (en) | For generating the method and apparatus of Face datection model | |
CN107944481A (en) | Method and apparatus for generating information | |
CN109545192A (en) | Method and apparatus for generating model | |
CN108989882A (en) | Method and apparatus for exporting the snatch of music in video | |
CN107315824A (en) | Method and apparatus for generating thermodynamic chart | |
CN108960316A (en) | Method and apparatus for generating model | |
CN109993150A (en) | The method and apparatus at age for identification | |
CN110225366A (en) | Video data processing and advertisement position determine method, apparatus, medium and electronic equipment | |
CN109299477A (en) | Method and apparatus for generating text header | |
CN107845016A (en) | information output 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 | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: Room 221, 2nd floor, Block C, 18 Kechuang 11th Street, Daxing Economic and Technological Development Zone, Beijing, 100176 Applicant after: Jingdong Digital Technology Holding Co., Ltd. Address before: Room 221, 2nd floor, Block C, 18 Kechuang 11th Street, Daxing Economic and Technological Development Zone, Beijing, 100176 Applicant before: Beijing Jingdong Financial Technology Holding Co., Ltd. |
|
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190118 |