CN109409533A - A kind of generation method of machine learning model, device, equipment and storage medium - Google Patents
A kind of generation method of machine learning model, device, equipment and storage medium Download PDFInfo
- Publication number
- CN109409533A CN109409533A CN201811143220.4A CN201811143220A CN109409533A CN 109409533 A CN109409533 A CN 109409533A CN 201811143220 A CN201811143220 A CN 201811143220A CN 109409533 A CN109409533 A CN 109409533A
- Authority
- CN
- China
- Prior art keywords
- model
- data
- machine learning
- generation
- learning 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.)
- Granted
Links
Abstract
The embodiment of the invention discloses a kind of generation method of machine learning model, device, equipment and storage mediums, the described method includes: obtaining the model interaction parameter that user is inputted by human-computer interaction interface, the model interaction parameter includes: that model generates data and shows information type；Data and the displaying information type are generated according to the model, are generated and at least one matched machine learning model of at least one model generation algorithm；According to the machine learning model of generation, generate and the matched intended display information of displaying information type；The intended display information is supplied to the user.The technical solution of the embodiment of the present invention can make the more generalization and automation of machine learning modeling procedure.
Description
Technical field
The present embodiments relate to machine learning techniques field more particularly to a kind of generation method of machine learning model,
Device, equipment and storage medium.
Background technique
Machine learning is a very important research field in computer science and artificial intelligence, and development machines learn mould
Type be a time-consuming, expert driving workflow, this process include data preparation, feature selecting, model or the choice of technology,
Trained and tuning etc..
The automatic machinery learning model building tool increased income in existing machine learning field, such as Tpot, Auto-sklearn
Deng, modeling procedure can simplify by the built-in function for calling its packed, it is automatic to search for super ginseng, reach automation or semi-automatic
Change the purpose of modeling.
In the implementation of the present invention, the discovery prior art has following defects that inventor
Existing machine learning modeling tool is not mature enough, and is not to be applicable in per capita any.Use existing machine
Learning model building tool needs certain machine learning using basis and correlation experience, this limits machine learning to a certain extent
The use scope of tool.In addition, the algorithm that existing machine learning modeling tool is supported is based on traditional algorithm, such as decision tree or
GBDT (Gradient Boosting Decision Tree, the classification based on decision tree return) scheduling algorithm.But current industry is normal
With and the good algorithm of effect do not support that, such as xgboost, lgbm or catboost scheduling algorithm, this is also limited to a certain extent
Being widely used of machine learning modeling tool.
Summary of the invention
The embodiment of the present invention provides generation method, device, equipment and the storage medium of a kind of machine learning model, so that machine
The more generalization and automation of device learning model building process.
In a first aspect, the embodiment of the invention provides a kind of generation methods of machine learning model, comprising:
The model interaction parameter that user is inputted by human-computer interaction interface is obtained, the model interaction parameter includes: model
It generates data and shows information type；
Data and the displaying information type are generated according to the model, are generated and at least one model generation algorithm
At least one machine learning model matched；
According to the machine learning model of generation, generate and the matched intended display information of displaying information type；
The intended display information is supplied to the user.
Second aspect, the embodiment of the invention also provides a kind of generating means of machine learning model, comprising:
Model interaction parameter acquisition module, the model interaction parameter inputted for obtaining user by human-computer interaction interface,
The model interaction parameter includes: that model generates data and shows information type；
Machine learning model generation module, it is raw for generating data and the displaying information type according to the model
At at least one matched machine learning model of at least one model generation algorithm；
Intended display information generating module generates and believes with the displaying for the machine learning model according to generation
Cease the intended display information of type matching；
Intended display information providing module, for the intended display information to be supplied to the user.
The third aspect, the embodiment of the invention also provides a kind of computer equipment, the computer equipment includes:
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 processing
Device realizes the generation method of machine learning model provided by any embodiment of the invention.
Fourth aspect, the embodiment of the invention also provides a kind of computer storage mediums, are stored thereon with computer program,
The program realizes the generation method of machine learning model provided by any embodiment of the invention when being executed by processor.
The model interaction parameter that the embodiment of the present invention is inputted by obtaining user by human-computer interaction interface, is closed according to model
The model for including in connection parameter generates data and shows information type, generates matched extremely at least one model generation algorithm
A few machine learning model, and generated according to the machine learning model of generation and believed with the matched intended display of information type is shown
It ceases to be supplied to user, realizes and automatically generate required machine learning model in the case where user does not contact code, and according to
User demand provides visual information for it and shows function, and it is lower to solve versatility existing for existing machine learning modeling tool
The problem of, improve the generalization and automation performance of machine learning modeling tool.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the generation method for machine learning model that the embodiment of the present invention one provides；
Fig. 2 a is a kind of flow chart of the generation method of machine learning model provided by Embodiment 2 of the present invention；
Fig. 2 b is a kind of flow chart of the generation method of machine learning model provided by Embodiment 2 of the present invention；
Fig. 2 c is a kind of flow chart of the generation method of machine learning model provided by Embodiment 2 of the present invention；
Fig. 2 d is a kind of human-computer interaction interface signal of machine learning model Core Generator provided by Embodiment 2 of the present invention
Figure；
Fig. 2 e is a kind of schematic diagram of model report file provided by Embodiment 2 of the present invention；
Fig. 3 is a kind of schematic diagram of the generating means for machine learning model that the embodiment of the present invention three provides；
Fig. 4 is a kind of structural schematic diagram for computer equipment that the embodiment of the present invention four provides.
Specific embodiment
The present invention 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 the present invention rather than limiting the invention.
It also should be noted that only the parts related to the present invention are shown for ease of description, in attached drawing rather than
Full content.It should be mentioned that some exemplary embodiments are described before exemplary embodiment is discussed in greater detail
At the processing or method described as flow chart.Although operations (or step) are described as the processing of sequence by flow chart,
It is that many of these operations can be implemented concurrently, concomitantly or simultaneously.In addition, the sequence of operations can be by again
It arranges.The processing can be terminated when its operations are completed, it is also possible to have the additional step being not included in attached drawing.
The processing can correspond to method, function, regulation, subroutine, subprogram etc..
Embodiment one
Fig. 1 is a kind of flow chart of the generation method for machine learning model that the embodiment of the present invention one provides, the present embodiment
It is applicable to the case where automatically generating machine learning model and matched intended display information, this method can be by machine learning mould
The generating means of type execute, which can be realized by the mode of software and/or hardware, and can generally be integrated in computer
In equipment.Correspondingly, as shown in Figure 1, this method includes following operation:
S110, the model interaction parameter that user is inputted by human-computer interaction interface is obtained, the model interaction parameter includes:
Model generates data and shows information type.
Wherein, model interaction parameter can be the relevant parameter for generating machine learning model, such as model generates number
According to model generation algorithm etc..Model, which generates data, can be what user specified, for generating the data source of machine learning model,
Including training data and test data.Show that information type can be for user's selection, for showing several types of correlation function
Type.For example, showing that information type may include important feature, index parameter and model report file etc..
In embodiments of the present invention, user can pass through engineering when modeling using machine learning model Core Generator
It practises the visualization human-computer interaction interface input that model Core Generator provides and generates data for generating the model of machine learning model,
And specific displaying information type is selected in human-computer interaction interface, so that machine learning model Core Generator can be according to user institute
The model of offer generates data and shows that information type automatically generates corresponding machine learning model.
In an alternate embodiment of the present invention where, it obtains user and number is generated by the model that human-computer interaction interface inputs
According to may include:
It obtains the user and data is generated by the model that the human-computer interaction interface uploads；Or
It obtains the user and the model selected in list generation data is stored in data by the human-computer interaction interface.
In embodiments of the present invention, model, which generates data, can be the corresponding document that user is uploaded by human-computer interaction interface
Data are also possible to user by human-computer interaction interface and store the dependency number prestored in the server specified in list in data
According to the acquisition modes that the embodiment of the present invention does not generate data to model are defined.
S120, data and the displaying information type are generated according to the model, generates and is generated at least one model
At least one machine learning model of algorithmic match.
Wherein, model generation algorithm can be the related algorithm for generating machine learning model, as catboost,
Xgboost or lgbm etc..
Correspondingly, after user generates data and show information type by human-computer interaction interface designated model, engineering
Model Core Generator is practised to produce and at least one matched machine learning model of at least one model generation algorithm.Wherein,
Model generation algorithm can be specified by user, can also be randomly assigned by machine learning model Core Generator, the embodiment of the present invention
To this and it is not limited.
S130, the machine learning model according to generation are generated and the matched intended display of the displaying information type
Information.
Wherein, intended display information can be the displaying information type that machine learning model Core Generator is selected according to user
Provided display data, such as the significant data feature of screening model generating algorithm, the algorithm effect of display model generating algorithm
Fruit and the corresponding model report file of acquisition machine learning model etc..Intended display information can be raw according to displaying information type
At any to can be used as intended display information, this hair by the displaying information that machine learning model Core Generator extends
Bright embodiment is not defined the content of intended display information.
It in embodiments of the present invention, can be according to user after machine learning model Core Generator generates machine learning model
Matched intended display information is generated by the displaying information type that human-computer interaction interface selects.
In an alternate embodiment of the present invention where, data and the displaying info class are being generated according to the model
Type can also include: to described before generation and at least one matched machine learning model of at least one model generation algorithm
Model generates data and carries out pretreatment operation so that the model generates data and meets model generation demand, wherein the pre- place
Managing operation includes at least one in processing shortage of data value, processing character type-word section and processing unbalanced data.
In embodiments of the present invention, optionally, user is obtained in machine learning model Core Generator pass through human-computer interaction circle
After the model of face input generates data, pretreatment operation can be carried out to the data received first.Wherein, pretreatment operation packet
Include but be not limited to processing shortage of data value, processing character type-word section and processing unbalanced data etc..To model generate data into
Row pretreatment operation can be used family specified data and be more in line with modeling demand.
S140, the intended display information is supplied to the user.
Correspondingly, machine learning model Core Generator, which is generated, generates matched intended display information with information type is shown
Afterwards, intended display information can be supplied to user to use or refer to.
In embodiments of the present invention, user need to only need corresponding data source in human-computer interaction interface and specify corresponding
Model generation algorithm automatically generates corresponding machine learning model using machine learning model Core Generator, and whole process is without using
Family inputs any code, and any machine learning model Core Generator that can use per capita carrys out automatic modeling, it can be seen that, the present invention
The generation method versatility and automation performance of machine learning model provided by embodiment are stronger.
The model interaction parameter that the embodiment of the present invention is inputted by obtaining user by human-computer interaction interface, is closed according to model
The model for including in connection parameter generates data and shows information type, generates matched extremely at least one model generation algorithm
A few machine learning model, and generated according to the machine learning model of generation and believed with the matched intended display of information type is shown
It ceases to be supplied to user, realizes and automatically generate required machine learning model in the case where user does not contact code, and according to
User demand provides visual information for it and shows function, and it is lower to solve versatility existing for existing machine learning modeling tool
The problem of, improve the generalization and automation performance of machine learning modeling tool.
Embodiment two
Fig. 2 a is a kind of flow chart of the generation method of machine learning model provided by Embodiment 2 of the present invention, and Fig. 2 b is this
Inventive embodiments two provide a kind of machine learning model generation method flow chart, Fig. 2 c be second embodiment of the present invention provides
A kind of machine learning model generation method flow chart, the present embodiment embodied based on above-described embodiment,
In the present embodiment, give matched extremely according to the generation of other kinds of model interaction parameter and at least one model generation algorithm
A few machine learning model, and according to the machine learning model of generation, it generates matched with the displaying information type
The specific implementation of intended display information.
Correspondingly, as shown in Figure 2 a, when model interaction parameter include model generate data, model generation algorithm and with institute
When stating the matched machine learning model file of model generation algorithm and model report file, the method for the present embodiment may include:
S210a, the model interaction parameter that user is inputted by human-computer interaction interface, the model interaction parameter packet are obtained
Include: model generate data, model generation algorithm and with the matched machine learning model file of the model generation algorithm and mould
Type report file.
In embodiments of the present invention, optionally, model interaction parameter may include that model generates data, model generation algorithm
And with the matched machine learning model file of model generation algorithm and model report file.That is, user can be by man-machine
Interactive interface designated model generates data, and selection generates the algorithm of machine learning model, and appointment display information type is and mould
The matched machine learning model file of type generating algorithm and model report file.
S220a, model generation data are divided into model training subdata and model measurement subdata.
Wherein, model training subdata can be the training data (i.e. training set) for generating machine mould, and model is surveyed
Swab data can be the test data (i.e. test set) for generating machine mould.
Correspondingly, user generates data and model generation algorithm, and appointment display by human-computer interaction interface designated model
Information type is with after the matched machine learning model file of model generation algorithm and model report file, and machine learning model is raw
It needs to generate data to model at tool and be handled to generate corresponding model training subdata and model measurement subdata.
Wherein, the label that model measurement subdata is used to generate machine learning model is verified.
S230a, according to the model training subdata and the model measurement subdata, using Bayesian Optimization Algorithm
Parameter adjustment is carried out at least one hyper parameter for including in the model generation algorithm under limited times number of attempt.
It in embodiments of the present invention, optionally, can be with to determining model training subdata and model measurement subdata
It carries out adjusting ginseng in detail using the model generation algorithm that Bayesian Optimization Algorithm and user select.Bayesian Optimization Algorithm can
In limited number of attempt, the probability distribution of super ginseng is searched for, reasonable parameter is finally obtained to model, is compared with the traditional method, energy
Enough save a large amount of time and computing resource.
S240a, data and the hyper parameter adjusted are generated according to the model, generates to generate with the model and calculates
The matched machine learning model of method.
Correspondingly, machine learning model Core Generator can be with operational model generating algorithm, to generate after the completion of adjusting ginseng operation
With the matched machine learning model of model generation algorithm.
S250a, the machine learning model according to generation are generated and the matched machine learning of the model generation algorithm
Model file and model report file are supplied to user.
In embodiments of the present invention, optionally, machine learning model Core Generator can also generate and model generation algorithm
Matched detailed model report file, and machine learning model file and model report file are supplied to user.
Correspondingly, as shown in Figure 2 b, when model interaction parameter include model generate data, model generation algorithm and with institute
When stating the matched preset data feature of model generation algorithm, the method for the present embodiment may include:
S210b, the model interaction parameter that user is inputted by human-computer interaction interface, the model interaction parameter packet are obtained
Include: model generate data, model generation algorithm and with the matched preset data feature of the model generation algorithm.
Wherein, preset data feature can be the significant data feature of model generation algorithm.
In embodiments of the present invention, optionally, model interaction parameter can also include that model generates data, model is generated and calculated
Method and with the matched preset data feature of model generation algorithm.That is, user can pass through human-computer interaction interface designated model
Data are generated, selection generates the algorithm of machine learning model, and appointment display information type is matched with model generation algorithm
Preset data feature.
S220b, model generation data are divided into model training subdata and model measurement subdata.
Similarly, user generates data and model generation algorithm by human-computer interaction interface designated model, and appointment display is believed
Ceasing type is with after the matched preset data feature of model generation algorithm, and machine learning model Core Generator needs to generate model
Data are handled to generate corresponding model training subdata and model measurement subdata.
S230b, data are generated according to default hyper parameter corresponding with the model generation algorithm and the model, it is raw
At machine learning model corresponding with the model generation algorithm.
Wherein, default hyper parameter can be the hyper parameter of model generation algorithm default.
In embodiments of the present invention, if user is only intended to check that correlation model is generated by machine learning model Core Generator
The significant data feature in the process of running of algorithm then in human-computer interaction interface appointment display information type can be and model
The matched preset data feature of generating algorithm.In the data characteristics checked in model generation algorithm operational process, without
Optimization algorithm carries out tune ginseng, and the mould of user's selection can be used to determining model training subdata and model measurement subdata
Type generating algorithm is run according to the hyper parameter of default, to generate machine learning model corresponding with model generation algorithm.
S240b, model generation data are ranked up according to setting ordering rule.
Wherein, setting ordering rule can be preset, for generating the rule of data sorting to model.For example, setting
Ordering rule can be the significance level sequence by field in data, and the embodiment of the present invention is not to the specific of setting ordering rule
Form is defined.
Correspondingly, model can be generated to data according to setting ordering rule in order to provide significant data feature to user
It is ranked up.
S250b, the model generation algorithm matching is generated according to the machine learning model and ranking results of generation
At least one preset data characteristic value.
It in embodiments of the present invention, can be most heavy according to setting ordering rule output in model generation algorithm operational process
The several preset data features and corresponding eigenvalue (the significance level score value of such as data characteristics) wanted, with screening for reference
Feature.
Correspondingly, as shown in Figure 2 c, when model interaction parameter includes that model generates data, at least two model generation algorithms
And each model generation algorithm performance comparison when, the method for the present embodiment may include:
S210c, the model interaction parameter that user is inputted by human-computer interaction interface, the model interaction parameter packet are obtained
Include: model generates the performance comparison of data, at least two model generation algorithms and each model generation algorithm.
In embodiments of the present invention, optionally, model interaction parameter can also include that model generates data, at least two moulds
The performance comparison of type generating algorithm and each model generation algorithm.That is, user can pass through human-computer interaction interface designated model
Data are generated, select at least two algorithms for generating machine learning model, and appointment display information type is that the generation of each model is calculated
The performance comparison of method.
S220c, model generation data are divided into model training subdata and model measurement subdata.
Similarly, user generates data and at least two model generation algorithms by human-computer interaction interface designated model, and refers to
Surely after showing the performance comparison that information type is each model generation algorithm, machine learning model Core Generator needs to generate model
Data are handled to generate corresponding model training subdata and model measurement subdata.
S230c, number is generated according to each corresponding default hyper parameter of model generation algorithm and the model
According to generating machine learning model corresponding with each model generation algorithm.
In embodiments of the present invention, if user is only intended to compare correlation model generation by machine learning model Core Generator
The performance of algorithm, then can be in the performance comparison that human-computer interaction interface appointment display information type is each model generation algorithm.Together
Reason, when machine learning model Core Generator quickly runs each model generation algorithm, carries out tune ginseng without optimization algorithm, can
To use the model generation algorithm of user's selection according to default determining model training subdata and model measurement subdata
Hyper parameter operation, to generate corresponding with each model generation algorithm machine learning model.
S240c, the machine learning model according to generation generate model corresponding with each model generation algorithm and comment
Valence index.
Correspondingly, each mould can be generated in order to which at least two model generation algorithms selected user carry out performance comparison
The corresponding model-evaluation index of type generating algorithm, with the corresponding model generation algorithm of selection for reference.
In an alternate embodiment of the present invention where, model generation data are divided into model measurement subdata, it can
To include:
The user is generated into the target data specified in data as the model measurement subdata in the model, or
The target data that the model generates setting ratio in data is generated as the model measurement subdata.
Wherein, setting ratio can be the preset ratio of machine learning model Core Generator, such as 20% or 30%
Deng can specifically be set according to actual demand, the embodiment of the present invention is not defined the specific value of setting ratio.
In embodiments of the present invention, optionally, when determining test set, if user is specified by human-computer interaction interface
Test set data, then the target data that can directly specify user is as model measurement subdata.If user does not specify
Test set data, then machine learning model Core Generator can be surveyed the partial target data of model training subdata as model
Swab data.For example, using in model training subdata 20% data as model measurement subdata.
It should be noted that in embodiments of the present invention, model report file is a detailed report, can wrap simultaneously
Include the contents such as the significant data feature of algorithm and the model-evaluation index of algorithm.
It is further to note that Fig. 2 a, Fig. 2 b and Fig. 2 c are only a kind of schematic diagrames of implementation, S210a-S250a,
There is no sequencing relationship between S210b-S250b and S210c-S250c, three respectively corresponds user and selects different displayings
Performed operation when information type, three can select an implementation.
Fig. 2 d is a kind of human-computer interaction interface signal of machine learning model Core Generator provided by Embodiment 2 of the present invention
Figure, Fig. 2 e is a kind of schematic diagram of model report file provided by Embodiment 2 of the present invention.In a specific example, such as scheme
Shown in 2d, optional operational mode can be provided in the human-computer interaction interface of machine learning model Core Generator for user
(part STEP1:CHOOSE RUN MODE), namely show information type, user can pass through each displaying in operational mode
Information type is selected, so that machine learning model Core Generator provides corresponding service.Wherein, Important
Feature, which can be, shows preset data feature, and Fast Compare can be the performance comparison for showing each model generation algorithm,
Detail Report (Slow) can be displaying and the matched machine learning model file of model generation algorithm and model report text
Part.The part STEP2:INPUT Y NAME is the tag names of user's input in Fig. 2 d, i.e., expected machine learning model is corresponding
The label of output.Human-computer interaction interface can also provide optional model generation algorithm for user, such as the STEP3 in Fig. 2 d:
The part CHOOSE ALGORITHM (JUST POR DETALL MODE), including Random Forest, Catboost, Xgboost
Or Logistic Regression etc., can also other model generation algorithm options be provided for user according to actual needs.
The part STEP4:CHOOSE DATA is the designated model generation data function that user provides.User can pass through From
Database selects data storage list to generate data for the data prestored in the server as model, can also pass through From
Upload uploads local file as model and generates data.After user completes all steps, " Click Me " can be clicked and pressed
Button, machine learning model Core Generator can be automatically operated at this time, and is generated and corresponded to according to the displaying information type of user's selection
Machine learning model and intended display information.Specifically, if user needs complete model report file, it can be in people
In machine interactive interface select Detail Report, generate model report file in can include simultaneously preset data feature,
The model-evaluation index of algorithm and other related contents.It, can be if user only needs to check the important feature of algorithm
Important Feature is selected in human-computer interaction interface, the intended display information generated at this time only includes important according to field
The characteristic information of degree ranking.If user needs to compare the performance indicator of several algorithms, Fast Compare can choose,
The intended display information generated at this time only includes the corresponding model evaluation index of several model generation algorithms.It should be noted that
Fig. 2 d is only a kind of schematic diagram, can carry out further function to human-computer interaction interface according to actual needs on the basis of Fig. 2 d
It can extend, the embodiment of the present invention is to this and is not limited.
Illustratively, the partial data content of model report file can refer to Fig. 2 e, wherein Index is the mould of algorithm
Type evaluation index, Sample Info indicate sample information, are some statistical contents to sample size, and Dev indicates model training
The corresponding each model-evaluation index of subdata and the corresponding numerical value of sample statistics, Val and Off indicate that model measurement subdata is corresponding
Each model-evaluation index and the corresponding numerical value of sample statistics.All Variables can be model generation algorithm operational process
In the total data feature that is related to, Importance Top 50 then can be to be shown according to the significance level of each data characteristics
Preceding 50 data characteristicses and corresponding data feature values namely feature significance level score value.It should be noted that Fig. 2 e
It is only a kind of schematic diagram, the typesetting format of model report file and the content information of display can be set according to actual needs
Fixed, the embodiment of the present invention is to this and is not limited.
By adopting the above technical scheme, it may be implemented to automatically generate required engineering in the case where user does not contact code
Model is practised, and provides visual information according to user demand for it and shows function, existing machine learning modeling tool is solved and deposits
The lower problem of versatility, improve the generalization and automation performance of machine learning modeling tool.
Embodiment three
Fig. 3 is a kind of schematic diagram of the generating means for machine learning model that the embodiment of the present invention three provides, such as Fig. 3 institute
Show, described device includes: model interaction parameter acquisition module 310, machine learning model generation module 320, intended display information
Generation module 330 and intended display information providing module 340, in which:
Model interaction parameter acquisition module 310 is joined for obtaining user by the model interaction that human-computer interaction interface inputs
Number, the model interaction parameter include: that model generates data and shows information type；
Machine learning model generation module 320, for generating data and the displaying information type according to the model,
It generates and at least one matched machine learning model of at least one model generation algorithm；
Intended display information generating module 330 generates and the displaying for the machine learning model according to generation
The matched intended display information of information type；
Intended display information providing module 340, for the intended display information to be supplied to the user.
The model interaction parameter that the embodiment of the present invention is inputted by obtaining user by human-computer interaction interface, is closed according to model
The model for including in connection parameter generates data and shows information type, generates matched extremely at least one model generation algorithm
A few machine learning model, and generated according to the machine learning model of generation and believed with the matched intended display of information type is shown
It ceases to be supplied to user, realizes and automatically generate required machine learning model in the case where user does not contact code, and according to
User demand provides visual information for it and shows function, and it is lower to solve versatility existing for existing machine learning modeling tool
The problem of, improve the generalization and automation performance of machine learning modeling tool.
Optionally, the model interaction parameter further include: model generation algorithm；The displaying information type includes: and institute
State the matched machine learning model file of model generation algorithm and model report file；Machine learning model generation module 320, tool
Body is used to model generation data being divided into model training subdata and model measurement subdata；It is instructed according to the model
Practice subdata and the model measurement subdata, using Bayesian Optimization Algorithm to the model under limited times number of attempt
At least one hyper parameter for including in generating algorithm carries out parameter adjustment；Data and institute adjusted are generated according to the model
Hyper parameter is stated, is generated and the matched machine learning model of the model generation algorithm.
Optionally, the model interaction parameter further include: model generation algorithm；The displaying information type includes: and institute
State the matched preset data feature of model generation algorithm；Machine learning model generation module 320 is specifically used for the model is raw
Model training subdata and model measurement subdata are divided at data；According to corresponding with the model generation algorithm default
Hyper parameter and the model generate data, generate machine learning model corresponding with the model generation algorithm；Intended display
Information generating module 330 is ranked up specifically for the model is generated data according to setting ordering rule；According to generation
The machine learning model and ranking results generate described at least one matched preset data characteristic value of model generation algorithm.
Optionally, the model interaction parameter further include: at least two model generation algorithms；The displaying information type packet
It includes: the performance comparison of each model generation algorithm；Machine learning model generation module 320 is specifically used for the model is raw
Model training subdata and model measurement subdata are divided at data；It is respectively corresponded according to each model generation algorithm
Default hyper parameter and the model generate data, generate machine learning corresponding with each model generation algorithm
Model；Intended display information generating module 330, specifically for the machine learning model according to generation, generate with it is each described
The corresponding model-evaluation index of model generation algorithm.
Optionally, model interaction parameter acquisition module 310 passes through the human-computer interaction interface for obtaining the user
The model of biography generates data；Or
It obtains the user and the model selected in list generation data is stored in data by the human-computer interaction interface.
Optionally, machine learning model generation module 320 is also used to refer to the user in model generation data
Fixed target data as the model measurement subdata, or
The target data that the model generates setting ratio in data is generated as the model measurement subdata.
Optionally, described device further include: data preprocessing module is pre-processed for generating data to the model
It operates so that the model generates data and meets model generation demand, wherein the pretreatment operation includes processing shortage of data
At least one of in value, processing character type-word section and processing unbalanced data.
Machine learning model provided by any embodiment of the invention can be performed in the generating means of above-mentioned machine learning model
Generation method, have the corresponding functional module of execution method and beneficial effect.The not technology of detailed description in the present embodiment
Details, reference can be made to the generation method for the machine learning model that any embodiment of that present invention provides.
Example IV
Fig. 4 is a kind of structural schematic diagram for computer equipment that the embodiment of the present invention four provides.Fig. 4, which is shown, to be suitable for being used to
Realize the block diagram of the computer equipment 412 of embodiment of the present invention.The computer equipment 412 that Fig. 4 is shown is only an example,
Should not function to the embodiment of the present invention and use scope bring any restrictions.
As shown in figure 4, computer equipment 412 is showed in the form of universal computing device.The component of computer equipment 412 can
To include but is not limited to: one or more processor 416, storage device 428 connect different system components (including storage dress
Set 428 and processor 416) bus 418.
Bus 418 indicates one of a few class bus structures or a variety of, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using any bus structures in a variety of bus structures.It lifts
For example, these architectures include but is not limited to industry standard architecture (Industry Standard
Architecture, ISA) bus, microchannel architecture (Micro Channel Architecture, MCA) bus, enhancing
Type isa bus, Video Electronics Standards Association (Video Electronics Standards Association, VESA) local
Bus and peripheral component interconnection (Peripheral Component Interconnect, PCI) bus.
Computer equipment 412 typically comprises a variety of computer system readable media.These media can be it is any can
The usable medium accessed by computer equipment 412, including volatile and non-volatile media, moveable and immovable Jie
Matter.
Storage device 428 may include the computer system readable media of form of volatile memory, such as arbitrary access
Memory (Random Access Memory, RAM) 430 and/or cache memory 432.Computer equipment 412 can be into
One step includes other removable/nonremovable, volatile/non-volatile computer system storage mediums.Only as an example, it deposits
Storage system 434 can be used for reading and writing immovable, non-volatile magnetic media, and (Fig. 4 do not show, commonly referred to as " hard drive
Device ").Although not shown in fig 4, the disk for reading and writing removable non-volatile magnetic disk (such as " floppy disk ") can be provided to drive
Dynamic device, and to removable anonvolatile optical disk (such as CD-ROM (Compact Disc-Read Only Memory, CD-
ROM), digital video disk (Digital Video Disc-Read Only Memory, DVD-ROM) or other optical mediums) read-write
CD drive.In these cases, each driver can pass through one or more data media interfaces and bus 418
It is connected.Storage device 428 may include at least one program product, which has one group of (for example, at least one) program
Module, these program modules are configured to perform the function of various embodiments of the present invention.
Program 436 with one group of (at least one) program module 426, can store in such as storage device 428, this
The program module 426 of sample includes but is not limited to operating system, one or more application program, other program modules and program
It may include the realization of network environment in data, each of these examples or certain combination.Program module 426 usually executes
Function and/or method in embodiment described in the invention.
Computer equipment 412 can also with one or more external equipments 414 (such as keyboard, sensing equipment, camera,
Display 424 etc.) communication, the equipment interacted with the computer equipment 412 communication can be also enabled a user to one or more,
And/or with any equipment (such as net that the computer equipment 412 is communicated with one or more of the other calculating equipment
Card, modem etc.) communication.This communication can be carried out by input/output (I/O) interface 422.Also, computer
Equipment 412 can also pass through network adapter 420 and one or more network (such as local area network (Local Area
Network, LAN), wide area network Wide Area Network, WAN) and/or public network, such as internet) communication.As schemed
Show, network adapter 420 is communicated by bus 418 with other modules of computer equipment 412.Although should be understood that in figure not
It shows, other hardware and/or software module can be used in conjunction with computer equipment 412, including but not limited to: microcode, equipment
Driver, redundant processing unit, external disk drive array, disk array (Redundant Arrays of Independent
Disks, RAID) system, tape drive and data backup storage system etc..
The program that processor 416 is stored in storage device 428 by operation, thereby executing various function application and number
According to processing, such as realize the generation method of machine learning model provided by the above embodiment of the present invention.
That is, the processing unit is realized when executing described program: obtaining the mould that user is inputted by human-computer interaction interface
Type relevant parameter, the model interaction parameter include: that model generates data and shows information type；It is generated according to the model
Data and the displaying information type generate and at least one matched machine learning mould of at least one model generation algorithm
Type；According to the machine learning model of generation, generate and the matched intended display information of displaying information type；It will be described
Intended display information is supplied to the user.
Embodiment five
The embodiment of the present invention five also provides a kind of computer storage medium for storing computer program, the computer program
When being executed by computer processor for executing the generation side of any machine learning model of the above embodiment of the present invention
Method: obtaining the model interaction parameter that user is inputted by human-computer interaction interface, and the model interaction parameter includes: that model generates number
Accordingly and show information type；Data and the displaying information type are generated according to the model, are generated and at least one mould
At least one matched machine learning model of type generating algorithm；According to the machine learning model of generation, generate and the exhibition
Show the matched intended display information of information type；The intended display information is supplied to the user.
The computer storage medium of the embodiment of the present invention, can be using any of one or more computer-readable media
Combination.Computer-readable medium can be computer-readable signal media or computer readable storage medium.It is computer-readable
Storage medium for example may be-but not limited to-the system of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, device or
Device, or any above combination.The more specific example (non exhaustive list) of computer readable storage medium includes: tool
There are electrical connection, the portable computer diskette, hard disk, random access memory (RAM), read-only memory of one or more conducting wires
(Read Only Memory, ROM), erasable programmable read only memory ((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 above-mentioned any appropriate combination.In this document, computer readable storage medium, which can be, any includes
Or the tangible medium of storage program, which can be commanded execution system, device or device use or in connection make
With.
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 unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited
In wireless, electric wire, optical cable, radio frequency (Radio Frequency, RF) etc. or above-mentioned any appropriate combination.
The computer for executing operation of the present invention can be write with one or more programming languages or combinations thereof
Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
Further include conventional procedural programming language --- such as " C " language or similar programming language.Program code can
Fully to execute, partly execute on the user computer on the user computer, be executed as an independent software package,
Part executes on the remote computer or executes on a remote computer or server completely on the user computer for part.
In situations involving remote computers, remote computer can pass through the network of any kind --- including local area network (LAN)
Or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as utilize Internet service
Provider is connected by internet).
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (10)
1. a kind of generation method of machine learning model characterized by comprising
The model interaction parameter that user is inputted by human-computer interaction interface is obtained, the model interaction parameter includes: that model generates
Data and displaying information type；
Data and the displaying information type are generated according to the model, are generated matched at least one model generation algorithm
At least one machine learning model；
According to the machine learning model of generation, generate and the matched intended display information of displaying information type；
The intended display information is supplied to the user.
2. the method according to claim 1, wherein the model interaction parameter further include: model generation algorithm；
The displaying information type includes: and the matched machine learning model file of the model generation algorithm and model report file；
Data and the displaying information type are generated according to the model, are generated matched at least one model generation algorithm
At least one machine learning model, comprising:
Model generation data are divided into model training subdata and model measurement subdata；
According to the model training subdata and the model measurement subdata, tasted using Bayesian Optimization Algorithm in limited times
Parameter adjustment is carried out at least one hyper parameter for including in the model generation algorithm under examination number；
Data and the hyper parameter adjusted are generated according to the model, are generated and the matched machine of the model generation algorithm
Device learning model.
3. the method according to claim 1, wherein the model interaction parameter further include: model generation algorithm；
The displaying information type includes: and the matched preset data feature of the model generation algorithm；
Data and the displaying information type are generated according to the model, are generated matched at least one model generation algorithm
At least one machine learning model, comprising:
Model generation data are divided into model training subdata and model measurement subdata；
Data are generated according to default hyper parameter corresponding with the model generation algorithm and the model, are generated and the mould
The corresponding machine learning model of type generating algorithm；
According to the machine learning model of generation, generate and the matched intended display information of displaying information type, comprising:
The model is generated data to be ranked up according to setting ordering rule；
According to the machine learning model and ranking results of generation generate the model generation algorithm it is matched at least one
Preset data characteristic value.
4. the method according to claim 1, wherein the model interaction parameter further include: at least two models
Generating algorithm；It is described to show that information type includes: the performance comparison of each model generation algorithm；
Data and the displaying information type are generated according to the model, are generated matched at least one model generation algorithm
At least one machine learning model, comprising:
Model generation data are divided into model training subdata and model measurement subdata；
Generate data according to each corresponding default hyper parameter of model generation algorithm and the model, generate with
The corresponding machine learning model of each model generation algorithm；
According to the machine learning model of generation, generate and the matched intended display information of displaying information type, comprising:
According to the machine learning model of generation, model-evaluation index corresponding with each model generation algorithm is generated.
5. method according to claim 1-4, which is characterized in that obtain user and inputted by human-computer interaction interface
Model generate data, comprising:
It obtains the user and data is generated by the model that the human-computer interaction interface uploads；Or
It obtains the user and the model selected in list generation data is stored in data by the human-computer interaction interface.
6. according to any method of claim 2-4, which is characterized in that model generation data are divided into model and are surveyed
Swab data, comprising:
The user is generated into the target data specified in data as the model measurement subdata in the model, or
The target data that the model generates setting ratio in data is generated as the model measurement subdata.
7. the method according to claim 1, wherein generating data and displaying letter according to the model
Type is ceased, before generation and at least one matched machine learning model of at least one model generation algorithm, further includes:
To model generation data progress pretreatment operation so that the model generates data and meets model generation demand,
In, the pretreatment operation includes processing shortage of data value, processing character type-word section and handles in unbalanced data at least
One.
8. a kind of generating means of machine learning model characterized by comprising
Model interaction parameter acquisition module, the model interaction parameter inputted for obtaining user by human-computer interaction interface are described
Model interaction parameter includes: that model generates data and shows information type；
Machine learning model generation module, for generating data and the displaying information type according to the model, generate with
At least one matched machine learning model of at least one model generation algorithm；
Intended display information generating module generates and the displaying info class for the machine learning model according to generation
The matched intended display information of type；
Intended display information providing module, for the intended display information to be supplied to the user.
9. a kind of computer equipment, which is characterized in that the equipment includes:
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 generation method of the machine learning model as described in any in claim 1-7.
10. a kind of computer storage medium, is stored thereon with computer program, which is characterized in that the program is executed by processor
The generation method of machine learning model of the Shi Shixian as described in any in claim 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811143220.4A CN109409533B (en) | 2018-09-28 | 2018-09-28 | Method, device, equipment and storage medium for generating machine learning model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811143220.4A CN109409533B (en) | 2018-09-28 | 2018-09-28 | Method, device, equipment and storage medium for generating machine learning model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109409533A true CN109409533A (en) | 2019-03-01 |
| CN109409533B CN109409533B (en) | 2021-07-27 |
Family
ID=65466418
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811143220.4A Active CN109409533B (en) | 2018-09-28 | 2018-09-28 | Method, device, equipment and storage medium for generating machine learning model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109409533B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110008121A (en) * | 2019-03-19 | 2019-07-12 | 合肥中科类脑智能技术有限公司 | A kind of personalization test macro and its test method |
| CN110569984A (en) * | 2019-09-10 | 2019-12-13 | Oppo广东移动通信有限公司 | configuration information generation method, device, equipment and storage medium |
| CN110991649A (en) * | 2019-10-28 | 2020-04-10 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Deep learning model building method, device, equipment and storage medium |
| CN111221517A (en) * | 2019-10-12 | 2020-06-02 | 中国平安财产保险股份有限公司 | Model creating method and device, computer equipment and readable storage medium |
| CN112685958A (en) * | 2020-12-30 | 2021-04-20 | 西南交通大学 | SiC MOSFET blocking voltage determination method based on neural network |
| CN112784181A (en) * | 2019-11-08 | 2021-05-11 | 阿里巴巴集团控股有限公司 | Information display method, image processing method, information display device, image processing equipment and information display device |
| CN112966439A (en) * | 2021-03-05 | 2021-06-15 | 北京金山云网络技术有限公司 | Machine learning model training method and device and virtual experiment box |
| CN113570062A (en) * | 2020-04-28 | 2021-10-29 | 大唐移动通信设备有限公司 | Machine learning model parameter transmission method and device |
| CN113706285A (en) * | 2021-07-08 | 2021-11-26 | 长江大学 | Credit card fraud detection method |
| CN114579822A (en) * | 2021-12-13 | 2022-06-03 | 北京市建筑设计研究院有限公司 | Method and device for pushing modeling tool, electronic equipment and storage medium |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030059844A1 (en) * | 2001-08-15 | 2003-03-27 | Proteom Limited | Apparatus and method for predicting rules of protein sequence interactions |
| CN101782976A (en) * | 2010-01-15 | 2010-07-21 | 南京邮电大学 | Automatic selection method for machine learning in cloud computing environment |
| CN102608285A (en) * | 2012-02-21 | 2012-07-25 | 南京工业大学 | Prediction method for explosion characteristics of organic mixture based on support vector machine |
| CN103678580A (en) * | 2013-12-07 | 2014-03-26 | 浙江大学 | Multitask machine learning method for text classification and device thereof |
| US20140211991A1 (en) * | 2013-01-30 | 2014-07-31 | Imimtek, Inc. | Systems and methods for initializing motion tracking of human hands |
| CN104200087A (en) * | 2014-06-05 | 2014-12-10 | 清华大学 | Parameter optimization and feature tuning method and system for machine learning |
| CN105068661A (en) * | 2015-09-07 | 2015-11-18 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method and system based on artificial intelligence |
| CN106067040A (en) * | 2016-06-01 | 2016-11-02 | 深圳市寒武纪智能科技有限公司 | A kind of method by fragment interactive training machine learning image recognition algorithm model |
| CN106169096A (en) * | 2016-06-24 | 2016-11-30 | 山西大学 | A kind of appraisal procedure of machine learning system learning performance |
| CN106446230A (en) * | 2016-10-08 | 2017-02-22 | 国云科技股份有限公司 | Method for optimizing word classification in machine learning text |
| CN106897570A (en) * | 2017-03-02 | 2017-06-27 | 山东师范大学 | A kind of COPD test system based on machine learning |
| US20170249547A1 (en) * | 2016-02-26 | 2017-08-31 | The Board Of Trustees Of The Leland Stanford Junior University | Systems and Methods for Holistic Extraction of Features from Neural Networks |
| CN107391603A (en) * | 2017-06-30 | 2017-11-24 | 北京奇虎科技有限公司 | User's portrait method for building up and device for mobile terminal |
| CN107480435A (en) * | 2017-07-31 | 2017-12-15 | 广东精点数据科技股份有限公司 | A kind of automatic searching machine learning system and method applied to clinical data |
| US10032111B1 (en) * | 2017-02-16 | 2018-07-24 | Rockwell Collins, Inc. | Systems and methods for machine learning of pilot behavior |
| CN108416363A (en) * | 2018-01-30 | 2018-08-17 | 平安科技（深圳）有限公司 | Generation method, device, computer equipment and the storage medium of machine learning model |
-
2018
- 2018-09-28 CN CN201811143220.4A patent/CN109409533B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030059844A1 (en) * | 2001-08-15 | 2003-03-27 | Proteom Limited | Apparatus and method for predicting rules of protein sequence interactions |
| CN101782976A (en) * | 2010-01-15 | 2010-07-21 | 南京邮电大学 | Automatic selection method for machine learning in cloud computing environment |
| CN102608285A (en) * | 2012-02-21 | 2012-07-25 | 南京工业大学 | Prediction method for explosion characteristics of organic mixture based on support vector machine |
| US20140211991A1 (en) * | 2013-01-30 | 2014-07-31 | Imimtek, Inc. | Systems and methods for initializing motion tracking of human hands |
| CN103678580A (en) * | 2013-12-07 | 2014-03-26 | 浙江大学 | Multitask machine learning method for text classification and device thereof |
| CN104200087A (en) * | 2014-06-05 | 2014-12-10 | 清华大学 | Parameter optimization and feature tuning method and system for machine learning |
| CN105068661A (en) * | 2015-09-07 | 2015-11-18 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method and system based on artificial intelligence |
| US20170249547A1 (en) * | 2016-02-26 | 2017-08-31 | The Board Of Trustees Of The Leland Stanford Junior University | Systems and Methods for Holistic Extraction of Features from Neural Networks |
| CN106067040A (en) * | 2016-06-01 | 2016-11-02 | 深圳市寒武纪智能科技有限公司 | A kind of method by fragment interactive training machine learning image recognition algorithm model |
| CN106169096A (en) * | 2016-06-24 | 2016-11-30 | 山西大学 | A kind of appraisal procedure of machine learning system learning performance |
| CN106446230A (en) * | 2016-10-08 | 2017-02-22 | 国云科技股份有限公司 | Method for optimizing word classification in machine learning text |
| US10032111B1 (en) * | 2017-02-16 | 2018-07-24 | Rockwell Collins, Inc. | Systems and methods for machine learning of pilot behavior |
| CN106897570A (en) * | 2017-03-02 | 2017-06-27 | 山东师范大学 | A kind of COPD test system based on machine learning |
| CN107391603A (en) * | 2017-06-30 | 2017-11-24 | 北京奇虎科技有限公司 | User's portrait method for building up and device for mobile terminal |
| CN107480435A (en) * | 2017-07-31 | 2017-12-15 | 广东精点数据科技股份有限公司 | A kind of automatic searching machine learning system and method applied to clinical data |
| CN108416363A (en) * | 2018-01-30 | 2018-08-17 | 平安科技（深圳）有限公司 | Generation method, device, computer equipment and the storage medium of machine learning model |
Non-Patent Citations (2)
| Title |
|---|
| NASRAOUI O等: "Human-Algorithm Interaction Biases in the Big Data Cycle: A Markov Chain Iterated Learning Framework", 《ARXIV》 * |
| 刘淇: "基于用户兴趣建模的推荐方法及应用研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110008121B (en) * | 2019-03-19 | 2022-07-12 | 合肥中科类脑智能技术有限公司 | Personalized test system and test method thereof |
| CN110008121A (en) * | 2019-03-19 | 2019-07-12 | 合肥中科类脑智能技术有限公司 | A kind of personalization test macro and its test method |
| CN110569984A (en) * | 2019-09-10 | 2019-12-13 | Oppo广东移动通信有限公司 | configuration information generation method, device, equipment and storage medium |
| CN111221517A (en) * | 2019-10-12 | 2020-06-02 | 中国平安财产保险股份有限公司 | Model creating method and device, computer equipment and readable storage medium |
| CN110991649A (en) * | 2019-10-28 | 2020-04-10 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Deep learning model building method, device, equipment and storage medium |
| CN112784181A (en) * | 2019-11-08 | 2021-05-11 | 阿里巴巴集团控股有限公司 | Information display method, image processing method, information display device, image processing equipment and information display device |
| CN113570062A (en) * | 2020-04-28 | 2021-10-29 | 大唐移动通信设备有限公司 | Machine learning model parameter transmission method and device |
| CN113570062B (en) * | 2020-04-28 | 2023-10-10 | 大唐移动通信设备有限公司 | Machine learning model parameter transmission method and device |
| CN112685958A (en) * | 2020-12-30 | 2021-04-20 | 西南交通大学 | SiC MOSFET blocking voltage determination method based on neural network |
| CN112685958B (en) * | 2020-12-30 | 2022-11-01 | 西南交通大学 | SiC MOSFET blocking voltage determination method based on neural network |
| CN112966439A (en) * | 2021-03-05 | 2021-06-15 | 北京金山云网络技术有限公司 | Machine learning model training method and device and virtual experiment box |
| CN113706285A (en) * | 2021-07-08 | 2021-11-26 | 长江大学 | Credit card fraud detection method |
| CN114579822A (en) * | 2021-12-13 | 2022-06-03 | 北京市建筑设计研究院有限公司 | Method and device for pushing modeling tool, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109409533B (en) | 2021-07-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109409533A (en) | A kind of generation method of machine learning model, device, equipment and storage medium | |
| US10719301B1 (en) | Development environment for machine learning media models | |
| CN109165249B (en) | Data processing model construction method and device, server and user side | |
| CN110580501B (en) | Zero sample image classification method based on variational self-coding countermeasure network | |
| US11556746B1 (en) | Fast annotation of samples for machine learning model development | |
| Dabbagh et al. | An approach for integrating the prioritization of functional and nonfunctional requirements | |
| CN111143226B (en) | Automatic test method and device, computer readable storage medium and electronic equipment | |
| US11960858B2 (en) | Performance based system configuration as preprocessing for system peformance simulation | |
| US20180300333A1 (en) | Feature subset selection and ranking | |
| US11861469B2 (en) | Code generation for Auto-AI | |
| US20230168639A1 (en) | Failure mode analytics | |
| EP3701403B1 (en) | Accelerated simulation setup process using prior knowledge extraction for problem matching | |
| US20210192376A1 (en) | Automated, progressive explanations of machine learning results | |
| US20220366244A1 (en) | Modeling Human Behavior in Work Environment Using Neural Networks | |
| CN112256537A (en) | Model running state display method and device, computer equipment and storage medium | |
| CN114341795A (en) | Conversation design robot for system design | |
| Yang et al. | Foundation models meet visualizations: Challenges and opportunities | |
| EP3842940A1 (en) | Method and system for uniformly performing feature extraction | |
| KR20230039488A (en) | Method for automatically generating artificial intelligence that trains training dataset including image data, and apparatus therefor | |
| JP2021500639A (en) | Prediction engine for multi-step pattern discovery and visual analysis recommendations | |
| US11822896B2 (en) | Contextual diagram-text alignment through machine learning | |
| CN104077669B (en) | The universal method that a kind of autonomous customization of computer operation stream is performed with driving | |
| CN113190154B (en) | Model training and entry classification methods, apparatuses, devices, storage medium and program | |
| US20220383150A1 (en) | Instantiating machine-learning models at on-demand cloud-based systems with user-defined datasets | |
| US11620550B2 (en) | Automated data table discovery for automated machine learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |