CN108664998A - A kind of picture training method and system that paraphrase is reinforced - Google Patents

Abstract

The embodiment of the present application discloses a kind of picture training method and system that paraphrase is reinforced, and is related to intelligent terminal technical field.The method includes：Obtain the first label of the first data set；It is multiple two level labels to disassemble first label according to paraphrase；According to the second data set of the two level label, training first label；Generation can extensive model.The picture training method and system that the paraphrase of the application is reinforced, the first label obtained by paraphrase dismantling is multiple two level labels, and the first label is trained according to the data set of two level label, and produce the first label can extensive model, reinforce picture training effect, improves user experience.

Description

A kind of picture training method and system that paraphrase is reinforced

Technical field

This application involves picture training methods and system that intelligent terminal technical field more particularly to paraphrase are reinforced.

Background technology

With the rapid development of the communication technology, people’s lives, work etc. are ceased with intelligent terminal (for example, communication terminal etc.) Manner of breathing closes.Currently, the function of intelligent terminal is gradually diversified, for example, the automatic classification of album picture may be implemented in intelligent terminal The functions such as arrangement.Wherein, picture classification is mainly realized by neural network training model, and specifically including input has the training of label Collection exports final mask, and classify to other pictures using the model that training obtains by the training of neural network.So And the training set of many labels is considerably less or is difficult to obtain, for example, since the training set of abnormal scene, special tag is difficult to obtain , terminal is difficult to be trained for the labels of these collection that lack in training.

Accordingly, it is desired to provide picture training method and system that a kind of paraphrase is reinforced, first obtained by paraphrase dismantling Label be multiple two level labels, according to the data set of two level label train the first label, and produce the first label can extensive mould Type reinforces picture training effect, improves user experience.

Invention content

According to some embodiments of the present application in a first aspect, provide the picture training method that a kind of paraphrase reinforces, answer For in terminal (for example, electronic equipment etc.), the method may include：Obtain the first label of the first data set；According to releasing It is multiple two level labels that justice, which disassembles first label,；According to the second data set of the two level label, training first mark Label；Generation can extensive model.

In some embodiments, the method may further include：Determine the weight of the two level label.

In some embodiments, the method may further include：Rewrite the training module of neural network model.

In some embodiments, the neural network model includes the model of increasing income realized by code, the model packet Include picture classification model.

In some embodiments, the picture classification model includes VGGNet models, ResNet models, InceptionV3 Model.

In some embodiments, the neural network model include data input layer, convolutional calculation layer, ReLU excitation layers, Pond layer, full articulamentum, data output layer.

In some embodiments, the rewriting training module is included in data output layer increase weight layer.

In some embodiments, the training first label includes training first label according to training algorithm, The training algorithm includes gradient decline, stochastic gradient descent.

In some embodiments, the method may further include：According to data set of the category of model without label.

According to the second aspect of some embodiments of the present application, a system is provided, including：One memory, by with It is set to storage data and instruction；One is established the processor communicated with memory, wherein when executing the instruction in memory, The processor is configured as：Obtain the first label of the first data set；It is multiple two to disassemble first label according to paraphrase Grade label；According to the second data set of the two level label, training first label；Generation can extensive model.

Therefore, the picture training method and system reinforced according to the paraphrase of some embodiments of the present application, are torn open by paraphrase The first label that solution obtains is multiple two level labels, trains the first label according to the data set of two level label, and produce the first mark Label can extensive model, reinforce picture training effect, improve user experience.

Description of the drawings

To more fully understand and illustrating some embodiments of the present application, below with reference to the description of attached drawing reference implementation example, In the drawings, same digital number indicates corresponding part in the accompanying drawings.

Fig. 1 is the illustrative diagram of the Environment System provided according to some embodiments of the present application.

Fig. 2 is the exemplary cell schematic diagram of the electronic functionalities configuration provided according to some embodiments of the present application.

Fig. 3 is the exemplary flow for the picture training method reinforced according to the paraphrase that some embodiments of the present application provide Figure.

Specific implementation mode

Below with reference to being described as convenient for Integrated Understanding the application defined in claim and its equivalent for attached drawing Various embodiments.These embodiments include various specific details in order to understand, but these are considered only as illustratively.Cause This, it will be appreciated by those skilled in the art that carrying out variations and modifications without departing from this to various embodiments described here The scope and spirit of application.In addition, briefly and to be explicitly described the application, the application will be omitted to known function and structure Description.

The term and phrase used in following description and claims is not limited to literal meaning, and being merely can Understand and consistently understands the application.Therefore, for those skilled in the art, it will be understood that provide to the various implementations of the application The description of example is only the purpose to illustrate, rather than limits appended claims and its application of Equivalent definitions.

Below in conjunction with the attached drawing in the application some embodiments, technical solutions in the embodiments of the present application carries out clear Chu is fully described by, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments. Based on the embodiment in the application, obtained by those of ordinary skill in the art without making creative efforts all Other embodiment shall fall in the protection scope of this application.

It should be noted that the term used in the embodiment of the present application is the mesh only merely for description specific embodiment , it is not intended to be limiting the application." one " of the embodiment of the present application and singulative used in the attached claims, "one", "an", " described " and "the" be also intended to including most forms, unless context clearly shows that other meanings.Also It should be appreciated that term "and/or" used herein refers to and includes that one or more list items purposes mutually bound are any Or all possible combinations.Expression " first ", " second ", " first " and " second " be for modify respective element without Consideration sequence or importance are used only for distinguishing a kind of element and another element, without limiting respective element.

Terminal according to the application some embodiments can be electronic equipment, the electronic equipment may include smart mobile phone, PC (PC, such as tablet computer, desktop computer, notebook, net book, palm PC PDA), mobile phone, e-book Reader, portable media player (PMP), audio/video player (MP3/MP4), video camera, virtual reality device (VR) and the combination of one or more of wearable device etc..According to some embodiments of the present application, the wearable device May include type of attachment (such as wrist-watch, ring, bracelet, glasses or wear-type device (HMD)), integrated type (such as electronics Clothes), decorated type (such as pad skin, tatoo or built in electronic device) etc. or several combinations.In some realities of the application It applies in example, the electronic equipment can be flexible, be not limited to above equipment, or can be one kind in above-mentioned various equipment Or several combination.In this application, term " user " can be indicated using the people of electronic equipment or setting using electronic equipment Standby (such as artificial intelligence electronic equipment).

The embodiment of the present application provides a kind of picture training method that paraphrase is reinforced.For the ease of understanding that the application is implemented Example, is described in detail the embodiment of the present application below with reference to attached drawing.

Fig. 1 is the illustrative diagram of the Environment System 100 provided according to some embodiments of the present application.Such as Fig. 1 Shown, Environment System 100 may include electronic equipment 110, network 120 and server 130 etc..Electronic equipment 110 can be with Including bus 111, processor 112, memory 113, input/output module 114, display 115, communication module 116 and physics Key 117 etc..In some embodiments of the present application, electronic equipment 110 can be omitted one or more elements, or can be into one Step includes one or more of the other element.

Bus 111 may include circuit.The circuit can be with one or more element (examples in interconnection electronics 110 Such as, bus 111, processor 112, memory 113, input/output module 114, display 115, communication module 116 and secondary or physical bond 117).The circuit can also be realized between one or more elements in electronic equipment 110 communication (for example, obtain and/or Send information).

Processor 112 may include one or more coprocessors (Co-processor), application processor (AP, Application Processor) and communication processor (Communication Processor).As an example, processor 112 can execute with the control of one or more elements of electronic equipment 110 and/or data processing (for example, starting picture training Deng operation).

Memory 113 can store data.The data may include other with one or more of electronic equipment 110 The relevant instruction of element or data.For example, the data may include the initial data before the processing of processor 112, intermediate data And/or treated data.Memory 113 may include impermanent memory memory and/or permanent memory memory.Make For example, memory 113 can store neural network model, storage label etc..

According to some embodiments of the present application, memory 113 can store software and/or program.Described program can wrap It includes kernel, middleware, Application Programming Interface (API, Application Programming Interface) and/or applies journey Sequence (or " application ").

At least part of the kernel, the middleware or the Application Programming Interface may include operating system (OS, Operating System).As an example, the kernel can be controlled or be managed for executing other programs (for example, intermediate Part, Application Programming Interface and application program) in realize operation or function system resource (for example, bus 111, processor 112, memory 113 etc.).In addition, the kernel can provide interface.The interface can by the middleware, described answer One or more elements of electronic equipment 110 are accessed to control or manage system resource with programming interface or the application program.

The middleware can be as the middle layer of data transmission.The data transmission can allow Application Programming Interface or Application program exchanges data with the kernel communication.As an example, the middleware can be handled from the application program One or more task requests of acquisition.For example, the middleware can distribute electronic equipment to one or more application program The priority of 110 system resource (for example, bus 111, processor 112, memory 113 etc.), and processing it is one or Multiple tasks are asked.The Application Programming Interface can be the application program for controlling from the kernel or the middleware The interface of function is provided.The Application Programming Interface can also include one or more interfaces or function (for example, instruction).It is described Function can be used for starting control, data channel control, security control, communication control, document control, window control, text control System, image procossing, information processing etc..

What input/output module 114 can be inputted to the transmission of the other elements of electronic equipment 110 from user or external equipment Instruction or data.Input/output module 114 can also be defeated by the instruction or data that are obtained from the other elements of electronic equipment 110 Go out to user or external equipment.In some embodiments, input/output module 114 may include input unit, and user can lead to Cross the input unit input information or instruction.

Display 115 can show content.The content can to user show various types (for example, text, image, Video, icon and/or symbol etc. or several combinations).Display 115 may include liquid crystal display (LCD, Liquid Crystal Display), light emitting diode (LED, Light-Emitting Diode) display, Organic Light Emitting Diode (OLED, Organic Light Emitting Diode) display, Micro Electro Mechanical System (MEMS, Micro Electro Mechanical Systems) display or electric paper display etc. or several combinations.Display 115 may include display Screen, touch screen etc..The display screen can show the label etc. of picture classification.In some embodiments, display 115 can be shown Show virtual key.The touch screen can obtain the input of the virtual key.Display 115 can be obtained defeated by the touch screen Enter.The input may include touch input, gesture input, action input, close input, electronic pen or user body part It inputs (for example, hovering input).

Communication module 116 can configure the communication between equipment.In some embodiments, Environment System 100 can be with Further comprise electronic equipment 140.As an example, the communication between the equipment may include electronic equipment 110 and other set Communication between standby (for example, server 130 or electronic equipment 140).For example, communication module 116 can by radio communication or Wire communication is connected to network 120, is communicated with other equipment (for example, server 130 or electronic equipment 140) realization.

The wireless communication may include microwave communication and/or satellite communication etc..The wireless communication may include honeycomb Communication is (for example, global mobile communication (GSM, Global System for Mobile Communications), CDMA (CDMA, Code Division MultipleAccess), 3G (Third Generation) Moblie (3G, The 3rd Generation Telecommunication), forth generation mobile communication (4G), the 5th third-generation mobile communication (5G), Long Term Evolution (LTE, Long Term Evolution), Long Term Evolution upgrade version (LTE-A, LTE-Advanced), wideband code division multiple access (WCDMA, Wideband Code Division MultipleAccess), Universal Mobile Communication System (UMTS, Universal Mobile Telecommunications System), WiMAX (WiBro, Wireless Broadband) etc. or several Combination.According to some embodiments of the present application, the wireless communication may include WLAN (WiFi, Wireless Fidelity), bluetooth, low-power consumption bluetooth (BLE, Bluetooth Low Energy), ZigBee protocol (ZigBee), near-field communication (NFC, Near Field Communication), magnetic safe transmission, radio frequency and body area network (BAN, BodyAreaNetwork) Deng or several combinations.According to some embodiments of the present application, the wire communication may include Global Navigation Satellite System (Glonass/GNSS, Global Navigation Satellite System), global positioning system (GPS, Global Position System), Beidou navigation satellite system or Galileo (European Global Satellite Navigation System) etc..The cable modem Letter may include universal serial bus (USB, Universal Serial Bus), high-definition media interface (HDMI, High- Definition Multimedia Interface), proposed standard 232 (RS-232, Recommend Standard 232), And/or plain old telephone service (POTS, Plain Old Telephone Service) etc. or several combinations.

Secondary or physical bond 117 can be used for user's interaction.Secondary or physical bond 117 may include one or more entity keys.In some realities It applies in example, user can be with the function of self-defined secondary or physical bond 117.As an example, secondary or physical bond 117 can send instruction.Described instruction May include starting picture classification etc..

In some embodiments, electronic equipment 110 may further include sensor.The sensor may include but not It is limited to photosensitive sensor, acoustic sensor, gas sensor, chemical sensor, voltage sensitive sensor, temp-sensitive sensor, fluid to pass Sensor, biosensor, laser sensor, Hall sensor, intelligence sensor etc. or several combinations.

In some embodiments, electronic equipment 110 may further include infrared equipment, image capture device etc..As Example, the infrared equipment can identify by infrared ray mode of delivery, and blink, watch the technical limit spacings eyes such as identification attentively Information.For example, the infrared equipment is acted by acquiring the blink of user come certification user information.As an example, described image Collecting device may include camera, iris device etc..The functions such as eyeball tracking may be implemented in the camera.The iris dress Authentication (for example, certification user information) can be carried out using iris recognition technology by setting.The iris device may include rainbow Film camera, the iris camera can obtain iris information, and the iris information can be stored in memory 113.

Network 120 may include communication network.The communication network may include computer network (for example, LAN (LAN, Local Area Network) or wide area network (WAN, Wide Area Network)), internet and/or telephone network Deng or several combinations.Network 120 can connect in Environment System 100 other equipment (for example, electronic equipment 110, Server 130, electronic equipment 140 etc.) and realize that information is transmitted.

Server 130 can connect the other equipment in Environment System 100 (for example, electronic equipment by network 120 110, electronic equipment 140 etc.).In some embodiments, when electronic equipment 110 is lost, server 130 can pass through network 120 send positioning instruction to electronic equipment；Electronic equipment 110 can send location information by network 120 to server 130 Deng.In some embodiments, server 130, which can be sent by network 120 to electronic equipment 110, starts picture training instruction Deng.

Electronic equipment 140 can be identical or different with electronic equipment 110 type.According to some embodiments of the present application, Some or all of execution operation can be in another equipment or multiple equipment (for example, electronic equipment 140 in electronic equipment 110 And/or server 130) in execute.In some embodiments, when electronic equipment 110 be automatically or in response to request execute it is a kind of or When multiple functions and/or service, electronic equipment 110 can ask other equipment (for example, electronic equipment 140 and/or server 130) it substitutes and executes function and/or service.In some embodiments, electronic equipment 110 is in addition to executing function or service, further Execute relative one or more functions.In some embodiments, other equipment is (for example, electronic equipment 140 and/or clothes Business device 130) requested function or other relevant one or more functions can be executed, implementing result can be sent to electricity Sub- equipment 110.Electronic equipment 110 can repeat result or be further processed implementing result, to provide requested function Or service.As an example, electronic equipment 110 can use cloud computing, distributed computing technology and/or client-server end to calculate meter Calculation etc. or several combinations.In some embodiments, according to the difference of cloud computing service property, the cloud computing may include Public cloud, private clound and mixed cloud etc..In some embodiments, when electronic equipment 110 is lost, electronic equipment 140 can be to Electronic equipment 110 sends positioning instruction, according to the location information that electronic equipment 110 is fed back, into line trace.In some embodiments In, electronic equipment 110 can send picture classification model etc. to electronic equipment 140.

It should be noted that the description for Environment System 100 above only for convenience of description can not be this Shen It please be limited within the scope of illustrated embodiment.It is appreciated that for those skilled in the art, the principle based on this system can Arbitrary combination can be carried out to each element, or constitute subsystem and connect with other elements under the premise of without departing substantially from the principle, To implementing the various modifications and variations of the above method and the progress of systematic difference field in form and details.For example, network environment System 100 may further include database.In another example electronic equipment 110 can not include secondary or physical bond 117 etc..It is all such The deformation of class, within the protection domain of the application.

Fig. 2 is the exemplary cell block diagram of the electronic functionalities configuration provided according to some embodiments of the present application.Such as Shown in Fig. 2, processor 112 may include processing module 200, and the processing module 200 may include acquiring unit 210, processing Unit 220, determination unit 230, generation unit 240, control unit 250.

According to some embodiments of the present application, acquiring unit 210 can obtain information.In some embodiments, the letter Breath can include but is not limited to text, picture, audio, video, action, gesture, sound, eyes (for example, iris information etc.), gas Breath, light etc. or several combinations.In some embodiments, described information can include but is not limited to input information, system information And/or communication information etc..As an example, acquiring unit 210 can pass through the touch of input/output module 114, display 115 Screen, secondary or physical bond 117 and/or sensor obtain the input information of electronic equipment 110.The input information may include other equipment The input of (for example, electronic equipment 140) and/or user, for example, the input of key-press input, touch-control, gesture input, action input, remote Journey input, transmission input, eyes input, voice input, breath input, light input etc. or several combinations.The input information Obtaining widget can include but is not limited to infrared equipment, image capture device, sensor etc. or several combinations.As showing Example, acquiring unit 210 can obtain the image data etc. of terminal by sensor image collecting device.

In some embodiments, acquiring unit 210 can obtain the communication information by network 120.The communication information can To include application software information, communication signal (for example, voice signal, vision signal etc.), short message etc..In some embodiments In, acquiring unit 210 can obtain system information by network 120, memory 113 and/or sensor.The system information can To include but not limited to the information that stores of the system mode of electronic equipment 110, presupposed information, memory 113 (for example, iris is recognized Demonstrate,prove information etc.) etc. or several combinations.As an example, acquiring unit 210 can obtain label or the mark by network 120 The data set etc. of label.

In some embodiments, described information may include instruction.Described instruction includes user instruction and/or system command Deng or several combinations.Described instruction may include triggering command, certification instruction, fill in instruction etc. or several combinations.Institute It may include certification user information, startup picture training instruction, startup picture classification instruction etc. to state certification instruction.As an example, When user presses secondary or physical bond (for example, shortcut key etc.), electronic equipment 110 can start picture classification etc..

According to some embodiments of the present application, processing unit 220 can handle data.In some embodiments, processing is single Member 220 can handle the input data of nerve network system.As an example, processing unit 220 can handle the label of picture.Example Such as, processing unit 220 can disassemble described label etc. according to the paraphrase of the label.In another example processing unit 220 can basis It is multiple two level labels that the label is disassembled in paraphrase.The paraphrase of the label may include the combination of the multiple two level label. It is that the paraphrase words such as near synonym, conjunctive word obtain the two level label that the paraphrase dismantling, which may include by disassembling the label,.

According to some embodiments of the present application, determination unit 230 can determine information.In some embodiments, it determines single Member 230 can determine the weight of two level label.As an example, determination unit 230 can be according to the two level label and the mark The relevance of label determines weight.For example, determination unit 230 can determine that the sum of weight of multiple two level labels is equal to 1.

According to some embodiments of the present application, generation unit 240 can generate data.In some embodiments, it generates single Member 240 can generate can extensive model.As an example, generation unit 240 can pass through the data set of the training two level label Generate picture classification model.It is described can extensive model can be the model that is obtained by neural metwork training data set, the mould Type can be adapted for the data other than the data set.

According to some embodiments of the present application, control unit 250 can be with control electronics.In some embodiments, it controls Unit 250 processed can utilize data set training picture classification model of the two level label etc..In some embodiments, control is single Member 250 can utilize data set etc. of the picture classification category of model without label.

It should be noted that the unit in processing module 200 is described above, it only for convenience of description, can not be this Application is limited within the scope of illustrated embodiment.It is appreciated that for those skilled in the art, the principle based on this system, Arbitrary combination may be carried out to each unit under the premise of without departing substantially from the principle, or constitute submodule and connect with other units It connects, the function to implementing above-mentioned module and unit carries out various modifications and variations in form and details.For example, processing module 200 can not include determination unit 230 and/or generation unit 240, can realize phase by analytic unit and/or other units The function of answering.In another example processing module 200 may further include storage unit, the storage unit can store label Data set etc..Suchlike deformation, within the protection domain of the application.

Fig. 3 is the exemplary flow for the picture training method reinforced according to the paraphrase that some embodiments of the present application provide Figure.As shown in figure 3, flow 300 can be realized by processing module 200.In some embodiments, the picture that the paraphrase is reinforced Training method can be started by instructing.Described instruction may include user instruction, system command, action command etc. or several Combination.As an example, the information that the system command can be obtained by sensor generates.The user instruction may include Voice, gesture, action, secondary or physical bond 117 and/or virtual key etc. or several combinations.

301, the first label of the first data set is obtained.Operation 301 can pass through the acquiring unit of processing module 200 210 realize.In some embodiments, acquiring unit 210 can obtain the first label of the first data set by network 120.Make Picture number for example, first data set is less than predetermined threshold value.In some embodiments, if label lacks data set, The training effect of model is poor, and accuracy of the model for picture classification when is caused to decline；By flow 300, can improve Lack the training precision of the label of data set.

302, it is multiple two level labels to disassemble first label according to paraphrase.Operation 302 can pass through processing module 200 processing unit 220 is realized.In some embodiments, it is multiple that processing unit 220 can disassemble the label according to paraphrase Two level label.The paraphrase of the label may include the combination of the multiple two level label.The paraphrase dismantling may include leading to It is that the paraphrase words such as near synonym, conjunctive word obtain the two level label to cross and disassemble the label.

303, the weight of the two level label is determined.Operation 303 can pass through the determination unit 230 of processing module 200 It realizes.In some embodiments, determination unit 230 can determine the weight of two level label.As an example, determination unit 230 can To determine weight according to the relevance of the two level label and the label.For example, when the first label is disassembled as two two levels Label, when described two two level labels are that near synonym are disassembled, determination unit 230 can determine the weight of described two two level labels It is respectively 0.5.In some embodiments, the sum of weight of the multiple two level label is equal to 1.According to some implementations of the application Example, flow 300 can not execute operation 303.

304, according to the second data set of the two level label, training first label.Operation 304 can pass through place The control unit 250 for managing module 200 is realized.In some embodiments, control unit 250 can utilize the number of the two level label According to collection training picture classification model.As an example, control unit 250 can train first label according to training algorithm.Institute It may include gradient decline, stochastic gradient descent etc. or several combinations to state training algorithm.

305, generation can extensive model.Operation 305 can be realized by the generation unit 240 of processing module 200.One In a little embodiments, data set generation picture classification model that generation unit 240 passes through the training two level label.It is described can be extensive Model can be the model obtained by neural metwork training data set, and the model can be adapted for other than the data set Data.

306, according to data set of the category of model without label.Operation 306 can pass through the control of processing module 200 Unit 250 is realized.In some embodiments, control unit 250 can utilize the number of the picture classification category of model without label According to collection etc..The picture classification model can include but is not limited to VGGNet models, ResNet models, Inception V3 models Deng or several combinations.According to some embodiments of the present application, flow 300 can not execute operation 306.

According to some embodiments of the present application, flow 300 may further include the training mould for rewriting neural network model Block.The training module for rewriting neural network model can be realized by the processing unit 220 of processing module 200.The god May include the model of increasing income realized by code through network model, the model may include picture classification model.The figure Piece disaggregated model can include but is not limited to VGGNet models, ResNet models, Inception V3 models etc. or several groups It closes.In some embodiments, the neural network model may include data input layer, convolutional calculation layer, ReLU excitation layers, pond Change layer, full articulamentum, data output layer.In some embodiments, the training module for rewriting neural network model can be with base In the weight that operation 303 determines.As an example, the rewriting training module may include increasing weight layer in data output layer. The data output layer may include softmax layers etc..Described softmax layers can be according to softmax functions by multiple scalars It is mapped as a probability distribution.The described softmax layers the model calculation as the output of data output layer can pass through probability Form performance.Open interval of the range of each value of the softmax layers of output in (0,1).For example, processing unit 220 It can increase weight logic at the softmax layers of model and judge.In another example processing unit 220 can connect in the complete of model end Layer is connect, weight layer is increased.Weight function etc. may be implemented in the weight layer.

It should be noted that the description for flow 300 above can not only for convenience of description be limited in the application Within the scope of illustrated embodiment.It is appreciated that for those skilled in the art, the principle based on this system may not carry on the back Under the premise of from the principle, arbitrary combination is carried out to each operation, or constitute sub-process and other operative combinations, in implementation The function of stating flow and operation carries out various modifications and variations in form and details.For example, flow 300 can not execute operation 303, and/or operation 306 etc.；In another example flow 300 may further include the behaviour such as the training module for rewriting neural network model Make.Suchlike deformation, within the protection domain of the application.

In conclusion according to picture training method and system that the paraphrase of the embodiment of the present application is reinforced, disassembled by paraphrase The first label obtained is multiple two level labels, trains the first label according to the data set of two level label, and produce the first label Can extensive model, reinforce picture training effect, improve user experience.

It should be noted that the above embodiments are intended merely as example, the application is not limited to such example, but can To carry out various change.

It should be noted that in the present specification, the terms "include", "comprise" or its any other variant are intended to Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those Element, but also include other elements that are not explicitly listed, or further include for this process, method, article or equipment Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that There is also other identical elements in process, method, article or equipment including the element.

Finally, it is to be noted that, it is above-mentioned it is a series of processing include not only with sequence described here in temporal sequence The processing of execution, and include the processing executed parallel or respectively rather than in chronological order.

One of ordinary skill in the art will appreciate that realizing all or part of flow in above-described embodiment method, being can be with It is completed by the relevant hardware of computer program instructions, the program can be stored in a computer readable storage medium, The program is when being executed, it may include such as the flow of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic disc, CD, read-only memory (Read-Only Memory, ROM) or random access memory (RandomAccess Memory, RAM) etc..

Above disclosed is only some preferred embodiments of the application, and the right model of the application cannot be limited with this It encloses, those skilled in the art can understand all or part of the processes for realizing the above embodiment, and is wanted according to the application right Equivalent variations made by asking, still belong to the scope covered by the invention.

Claims (10)

1. the picture training method that a kind of paraphrase is reinforced, which is characterized in that including：

Obtain the first label of the first data set；

It is multiple two level labels to disassemble first label according to paraphrase；

According to the second data set of the two level label, training first label；

Generation can extensive model.

2. the picture training method that paraphrase according to claim 1 is reinforced, which is characterized in that further comprise：

Determine the weight of the two level label.

3. the picture training method that paraphrase according to claim 2 is reinforced, which is characterized in that further comprise：

Rewrite the training module of neural network model.

4. the picture training method that paraphrase according to claim 3 is reinforced, which is characterized in that the neural network model packet The model of increasing income realized by code is included, the model includes picture classification model.

5. the picture training method that paraphrase according to claim 4 is reinforced, which is characterized in that the picture classification model packet Include VGGNet models, ResNet models, Inception V3 models.

6. the picture training method that paraphrase according to claim 4 is reinforced, which is characterized in that the neural network model packet Include data input layer, convolutional calculation layer, ReLU excitation layers, pond layer, full articulamentum, data output layer.

7. the picture training method that paraphrase according to claim 6 is reinforced, which is characterized in that the rewriting training module packet It includes and increases weight layer in data output layer.

8. the picture training method that paraphrase according to claim 1 is reinforced, which is characterized in that training first mark Label include according to training algorithm training first label, and the training algorithm includes gradient decline, stochastic gradient descent.

9. the picture training method that paraphrase according to claim 1 is reinforced, which is characterized in that further comprise：

According to data set of the category of model without label.

10. a system, which is characterized in that including：

One memory is configured as storage data and instruction；

One is established the processor communicated with memory, wherein when executing the instruction in memory, the processor is configured For：

Obtain the first label of the first data set；

It is multiple two level labels to disassemble first label according to paraphrase；

According to the second data set of the two level label, training first label；

Generation can extensive model.