CN110990870A - Operation and maintenance, processing method, device, equipment and medium using model library - Google Patents
Operation and maintenance, processing method, device, equipment and medium using model library Download PDFInfo
- Publication number
- CN110990870A CN110990870A CN201911202049.4A CN201911202049A CN110990870A CN 110990870 A CN110990870 A CN 110990870A CN 201911202049 A CN201911202049 A CN 201911202049A CN 110990870 A CN110990870 A CN 110990870A
- Authority
- CN
- China
- Prior art keywords
- model
- vehicle
- mounted terminal
- target
- models
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
- G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
- G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Abstract
The invention provides a processing method, a device, equipment and a medium for an operation and maintenance model library, wherein the processing method for the operation and maintenance model library is applied to a server and comprises the following steps: determining a current model base, wherein the model base is provided with N models, the models can be updated in a federal learning mode, and model selection information sent by any one first vehicle-mounted terminal is received; the model selection information is used for representing a target model selected by the first vehicle-mounted terminal in the N models; sending the target model to the first vehicle-mounted terminal so that the first vehicle-mounted terminal can use the target model; determining that the first vehicle-mounted terminal is one of the federally learned nodes of the target model. According to the invention, data does not need to be uploaded to the server, so that the privacy of the user and the safety of the data are guaranteed, and the vehicle-mounted intelligence can conveniently learn various skills and become more intelligent.
Description
Technical Field
The invention relates to the field of vehicle-mounted terminals, in particular to a method, a device, equipment and a medium for processing an operation and maintenance model library and a use model library.
Background
In order to meet the requirements of users such as driving, entertainment and interaction in a vehicle equipped with the vehicle-mounted terminal, corresponding software can be configured in the vehicle-mounted terminal, and part of the software needs to use a machine learning model.
In the prior art, machine learning needs to integrate multi-party data to train a model, the model is maintained by a server, the server can be requested to process when a vehicle-mounted terminal runs corresponding software, and at the moment, the server can input data received from the vehicle-mounted terminal into the model and further send a result fed back by the model to the vehicle-mounted terminal.
However, in the process, data of the vehicle-mounted terminal needs to be uploaded to the server, and whether the data is related to the user or the vehicle, privacy of the user can be involved, and uploading the data to the server easily causes potential safety hazards of the data.
Disclosure of Invention
The invention provides a processing method, a processing device, a processing equipment and a processing medium for operation and maintenance and use of a model library, and aims to solve the problem of potential safety hazard of data.
According to a first aspect of the present invention, there is provided a method for processing an operation and maintenance model library, applied to a server, including:
determining a current model library, wherein the model library has N models, the models can be updated in a federal learning mode, and N is an integer greater than or equal to 1;
receiving model selection information sent by any one first vehicle-mounted terminal; the model selection information is used for representing a target model selected by the first vehicle-mounted terminal in the N models;
sending the target model to the first vehicle-mounted terminal so that the first vehicle-mounted terminal can use the target model;
determining that the first vehicle-mounted terminal is one of the federally learned nodes of the target model.
Optionally, the models in the model library include at least one of the following models:
the system comprises an entertainment system model, a chassis adjustment calculation model, a driving strategy calculation model, an engine fuel-saving strategy calculation model, a motor power-saving strategy calculation model, a fatigue driving evaluation model, a driver physical condition evaluation model and a driving comfort evaluation model.
Optionally, after determining that the first vehicle-mounted terminal is one of the nodes learned by the federation of the target model, the method further includes:
receiving a node-trained model or node-trained parameters obtained by training the target model by the first vehicle-mounted terminal or another second vehicle-mounted terminal; the second on-board terminal is also one of the federal learning nodes of the target model;
and updating the target model in the model base according to the model after the node training or the parameter after the node training.
Optionally, after updating the target model in the model library according to the post-node-training model or the post-node-training parameter, the method further includes:
and updating the target model in each vehicle-mounted terminal as the federal learning node of the target model according to the updated target model in the model base.
Optionally, before receiving the model selection information sent by any one of the target vehicle-mounted terminals, the method further includes:
and sending model description information of at least one model in the N models to each vehicle-mounted terminal so that each vehicle-mounted terminal can determine the model selection information according to the model description information.
Optionally, before determining that the first vehicle-mounted terminal is one of the nodes learned by the federation of the target model, the method further includes:
and receiving join confirmation information which is sent by the first vehicle-mounted terminal and used for representing that the first vehicle-mounted terminal joins the federal study of the target model.
According to a second aspect of the present invention, there is provided a processing method using a model library, applied to a first vehicle-mounted terminal, including:
sending model selection information to a server of an operation and maintenance model library, wherein the model library comprises N models, and the model selection information is used for representing a target model selected by the first vehicle-mounted terminal in the N models;
receiving the target model sent by the server and enabling the target model to be used by the first vehicle-mounted terminal;
determining that the first vehicle-mounted terminal is one of the federally learned nodes of the target model.
According to a third aspect of the present invention, there is provided a processing apparatus for an operation and maintenance model library, including:
the model library determining module is used for determining a current model library, wherein the model library is provided with N models which can be updated in a federal learning mode, and N is an integer greater than or equal to 1;
the selected information receiving module is used for receiving model selected information sent by any one of the first vehicle-mounted terminals; the model selection information is used for representing a target model selected by the first vehicle-mounted terminal in the N models;
a model sending module, configured to send the target model to the first vehicle-mounted terminal, so that the first vehicle-mounted terminal can use the target model;
and the server side node determining module is used for determining that the first vehicle-mounted terminal is one of the nodes for federal learning of the target model.
According to a fourth aspect of the present invention, there is provided a processing apparatus using a model library, comprising:
the selected information sending module is used for sending model selected information to a server of an operation and maintenance model library, the model library is provided with N models, and the model selected information is used for representing a target model selected by a first vehicle-mounted terminal in the N models;
the model receiving module is used for receiving the target model sent by the server and enabling the target model to be used by the first vehicle-mounted terminal;
and the terminal side node determining module is used for determining that the first vehicle-mounted terminal is one of the nodes of the federal learning of the target model.
According to a fifth aspect of the present invention, there is provided an electronic device, comprising a memory and a processor,
the memory is used for storing codes;
the processor is configured to execute the codes in the memory to implement the processing method of the operation and maintenance model library related to the first aspect and the optional aspects thereof, or the processing method of the operation and maintenance model library related to the second aspect and the optional aspects thereof.
According to a sixth aspect of the present invention, there is provided a storage medium having stored thereon a program which, when executed by a processor, implements the method of processing an operation and maintenance model library according to the first aspect and alternatives thereof, or the method of processing using a model library according to the second aspect and alternatives thereof.
According to the processing method, device, equipment and medium for the operation and maintenance and the use of the model base, the server can operate and maintain the model base with various functions, meanwhile, the model in the model base can be sent to the corresponding vehicle-mounted terminal according to the requirements of the vehicle-mounted terminal, further, the vehicle-mounted terminal can complete the corresponding function by using the local model, data do not need to be uploaded to the server side, and the privacy of a user and the safety of the data are guaranteed.
Meanwhile, due to the establishment of the model base, the server can integrate various models, and based on the model base, the vehicle-mounted terminal can correspondingly download the required models, so that the functions of the vehicle-mounted terminal can be expanded through the downloading and the use of the models.
In addition, in the invention, the vehicle-mounted terminal can be used as a node for federal learning of the model, and further, the vehicle-mounted terminal which becomes the node can train the model by using the data of the vehicle-mounted terminal, thereby providing a basis for updating the model in the server. On the basis, each model of the server can be updated iteratively based on the training result of the model in the vehicle-mounted terminal, and the server still cannot directly receive related data serving as a training material while the iterative updating is realized, so that the privacy of a user and the safety of the data are further guaranteed.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a first flowchart illustrating a method for processing an operation and maintenance model library according to an embodiment of the present invention;
FIG. 2 is a second flowchart illustrating a method for processing the operation and maintenance model library according to an embodiment of the present invention;
FIG. 3 is a third schematic flow chart illustrating a processing method of the operation and maintenance model library according to an embodiment of the present invention;
FIG. 4 is a first flowchart illustrating a processing method using a model library according to an embodiment of the present invention;
FIG. 5 is a second flowchart illustrating a processing method using a model library according to an embodiment of the present invention;
FIG. 6 is a third flowchart illustrating a processing method using a model library according to an embodiment of the present invention;
FIG. 7 is a first flowchart illustrating a first exemplary embodiment of a processing apparatus for an operation and maintenance model library;
FIG. 8 is a second schematic diagram of the program modules of the apparatus for processing the operation and maintenance model library according to an embodiment of the present invention;
FIG. 9 is a first block diagram illustrating program modules of a processing device using a model library according to an embodiment of the present invention;
FIG. 10 is a second block diagram of a processing device using a model library according to an embodiment of the present invention;
fig. 11 is a schematic structural diagram of an electronic device in an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The technical solution of the present invention will be described in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.
Fig. 1 is a first flowchart illustrating a method for processing an operation and maintenance model library according to an embodiment of the present invention.
The processing method of the operation and maintenance model library according to the embodiment can be applied to a server, and correspondingly, the processing method using the model library can be applied to a vehicle-mounted terminal.
The server can be understood as any device or collection of devices with certain data storage and data processing capabilities, and is further equipped with any communication circuit capable of communicating with the outside.
The vehicle-mounted terminal can be a vehicle machine of the vehicle, and can also be any other intelligent terminal connected to the vehicle machine.
Referring to fig. 1, the method for processing the operation and maintenance model library includes:
s101: a current model library is determined.
The model library can be understood as a database with N models, wherein N is an integer greater than or equal to 1, model management can be performed in the model library according to functions of the model library, relevance of data processing and the like, and the models can be updated in a federal learning mode.
Further, the database may have one model or a plurality of different models for the same function. For example, different brands, models of vehicles may correspond to different models for the same function.
In one embodiment, the models in the model library include at least one of the following models:
the system comprises an entertainment system model, a chassis adjustment calculation model, a driving strategy calculation model, an engine fuel-saving strategy calculation model, a motor power-saving strategy calculation model, a fatigue driving evaluation model, a driver physical condition evaluation model and a driving comfort evaluation model.
Federal learning, which may also be described as league learning, joint learning, federal machine learning, etc., may be specifically understood as: a fed machine Learning or a fed Learning.
In one example, the working principle of federal learning applied to the field of vehicle-mounted technology may be, for example: the terminal as the node can download the current models from the server end respectively; wherein, part or all terminals can use respective data to train the model; each terminal transmits the trained model or the relevant trained parameters thereof to the server; the server aggregates the received models of the terminals or the parameters thereof into a final model.
The manner in which the update model is trained, transferred, and aggregated may be any means known or developed in the art.
After step S101, the method may further include:
s102: and receiving the model selection information sent by any one of the first vehicle-mounted terminals.
The model selection information can be understood as a target model selected by the first vehicle-mounted terminal in the N models; in a specific implementation process, the model selection information may include, for example, an identifier of the corresponding model, where the identifier may be a name, a customized number, or the like of the model, and may also include other information describing the target model, such as a storage location, a writer, a training time, a version number, or the like of the model. In any way, the description of the present embodiment is not deviated as long as the server can be made to know that the model required by the first in-vehicle terminal is the target model.
S103: sending the target model to the first vehicle-mounted terminal so that the first vehicle-mounted terminal can use the target model.
In the vehicle-mounted terminal, the use of the model may be independent of other software and programs, or may be performed in cooperation with (for example, called by) other software and programs to perform corresponding functions. Meanwhile, the model can be transmitted together with the transmission of the corresponding program and software.
In one embodiment, in order to guide the first vehicle-mounted terminal to use the target model, a corresponding matching algorithm may be sent to the first vehicle-mounted terminal, and the matching algorithm may function as: when the first vehicle-mounted terminal needs to use the function of a certain module, the vehicle-mounted terminal can directly utilize the target model in the vehicle-mounted terminal to process according to the guidance of the matching algorithm without sending the target model to a server for processing. Meanwhile, in some examples, the matching algorithm can also enable the vehicle-mounted terminal to train the model in the vehicle-mounted terminal under the guidance of the matching algorithm and feed back the training result.
In the specific implementation process, the target model can be called by software and programs (which can be software and programs running in the foreground or software and programs running in the background), and one target model can be used by one or more of the software and the programs.
In addition, if vehicles of different brands and models can correspond to different models for the same function, before step S103, at least two models of the same function corresponding to the function can be determined according to the model selection information, and then the target model can be determined according to the brand and/or model of the vehicle to which the first vehicle-mounted terminal belongs. Correspondingly, before step S103, the first vehicle-mounted terminal may report the relevant information of the brand and the model to the server, so as to provide a basis for the server to process.
Therefore, in the above embodiment, the server can operate and maintain the model libraries with various functions, and meanwhile, the models in the model libraries can be sent to the corresponding vehicle-mounted terminals according to the requirements of the vehicle-mounted terminals, so that the vehicle-mounted terminals can complete corresponding functions by using local models without uploading data to the server, and the privacy of users and the safety of the data are guaranteed.
Meanwhile, due to the establishment of the model base, the server can integrate various models, and based on the model base, the vehicle-mounted terminal can correspondingly download the required models, so that the functions of the vehicle-mounted terminal can be expanded through the downloading and the use of the models.
In this embodiment, after step S103, the method may further include:
s104: determining that the first vehicle-mounted terminal is one of the federally learned nodes of the target model.
Furthermore, the vehicle-mounted terminal as the federal learning node can participate in the training of the model.
Therefore, in the above embodiment, the vehicle-mounted terminal can also be used as a node for federal learning of the model, and further, the vehicle-mounted terminal serving as the node can train the model by using its own data, thereby providing a basis for updating the model in the server. On the basis, each model of the server can be updated iteratively based on the training result of the model in the vehicle-mounted terminal, and the server still cannot directly receive related data serving as a training material while the iterative updating is realized, so that the privacy of a user and the safety of the data are further guaranteed.
Furthermore, the present embodiment can be understood as being capable of implementing a model base SaaS service based on league learning, where SaaS specifically includes: Software-as-a-Service, abbreviated name.
Fig. 2 is a flowchart illustrating a processing method of the operation and maintenance model library according to an embodiment of the present invention.
Referring to fig. 2, in an embodiment, before step S102 and after step S101, the method may further include:
s105: and sending model description information of at least one model in the N models to each vehicle-mounted terminal so that each vehicle-mounted terminal can determine the model selection information according to the model description information.
The model description information may be any information capable of describing a corresponding model, such as information describing a function of the model, and may further include description information of software and a program adapted to be used by the model.
In addition, if vehicles of different brands and models can correspond to different models for the same function, the model description information may also specifically include information of the brands and vehicles corresponding to the models.
Further, the model described by the model description information may be specifically a model that the corresponding vehicle-mounted terminal does not have.
In a specific implementation process, step S105 may further include a process of sending ranking information to the vehicle-mounted terminal, and further, in the vehicle-mounted terminal, the model description information of each model may be arranged in an order indicated by the ranking information.
At any time before step S104, for example before or after step S103, or simultaneously with step S102, the method may further include:
s106: and receiving join confirmation information which is sent by the first vehicle-mounted terminal and used for representing that the first vehicle-mounted terminal joins the federal study of the target model.
Through the implementation mode, the vehicle-mounted terminal can independently select whether the node becomes the node for federal learning or not, if the node is not selected, the model of the vehicle-mounted terminal does not need to be trained, and the training result does not need to be fed back to the server; if the node is selected, the model of the vehicle-mounted terminal can be trained independently, and then the training result (such as the trained model or the trained parameter) is fed back to the server, so that the model in the model base is updated.
Meanwhile, the server can also screen terminals capable of entering the federal study according to the configuration condition of the corresponding vehicle-mounted terminal, and then the server can also receive software and hardware configuration information of the corresponding vehicle-mounted terminal when receiving the entering confirmation information, and further, whether the vehicle-mounted terminal is suitable to be used as a node of the federal study of the target model can be determined according to the software and hardware configuration information. For example, the software and hardware configuration information may be compared with a preset software and hardware standard, and if matching, it may be used as a node, otherwise, it may not be used as a node.
Fig. 3 is a third schematic flow chart illustrating a processing method of the operation and maintenance model library according to an embodiment of the present invention.
Referring to fig. 3, after step S104, the method may further include:
s107: receiving a node-trained model or node-trained parameters obtained by training the target model by the first vehicle-mounted terminal or another second vehicle-mounted terminal; the second on-board terminal is also one of the federal learning nodes of the target model;
s108: updating the target model in the model base according to the model after the node training or the parameter after the node training;
s109: and updating the target model in each vehicle-mounted terminal as the federal learning node of the target model according to the updated target model in the model base.
In addition, if vehicles of different brands and models can correspond to different models for the same function, the target model of the vehicle-mounted terminal of the vehicle of the same brand and model is updated in step S109 specifically according to the target model.
Through the above embodiment, the model in the server can be iteratively updated based on the training result of the model in the vehicle-mounted terminal.
In summary, in the processing method for the operation and maintenance model base provided in this embodiment, the server may operate and maintain the model bases with various functions, and meanwhile, the models in the model bases may be sent to the corresponding vehicle-mounted terminals according to the requirements of the vehicle-mounted terminals, and then the vehicle-mounted terminals may use the local models to complete the corresponding functions, without uploading data to the server, thereby ensuring the privacy of the user and the security of the data.
Meanwhile, due to the establishment of the model base, the server can integrate various models, and based on the model base, the vehicle-mounted terminal can correspondingly download the required models, so that the functions of the vehicle-mounted terminal can be expanded through the downloading and the use of the models.
In addition, in this embodiment, the vehicle-mounted terminal may also be used as a node for federal learning of the model, and further, the vehicle-mounted terminal serving as the node may train the model by using its own data, thereby providing a basis for updating the model in the server. On the basis, each model of the server can be updated iteratively based on the training result of the model in the vehicle-mounted terminal, and the server still cannot directly receive related data serving as a training material while the iterative updating is realized, so that the privacy of a user and the safety of the data are further guaranteed.
FIG. 4 is a first flowchart illustrating a processing method using a model library according to an embodiment of the present invention.
Referring to fig. 4, a processing method using a model library is applied to a first vehicle-mounted terminal, and includes:
s201: sending model selection information to a server of an operation and maintenance model library; further, the server may perform step S103 and step S104;
s202: receiving the target model sent by the server, and enabling the target model to be used by the first vehicle-mounted terminal.
The contents of the model library, the models, and the manner of using the models in the above process can be understood with reference to the related descriptions of the embodiments shown in fig. 1 to 3, and the repeated contents will not be described again.
Further, in step S202, the received target model may be a server.
In a specific implementation process, the model selection information may be generated by a user performing a selection operation in the interactive interface.
FIG. 5 is a flowchart illustrating a processing method using a model library according to an embodiment of the present invention.
Referring to fig. 5, before step S201, the method may further include:
s204: and receiving the model description information of at least one model in the N models sent by the server.
In a specific implementation process, after the model description information is received, the model description information can be output externally through a display interface or a voice form, and further, after the user learns the model description information, the corresponding model selection information can be determined through selection operation.
In an example, when each piece of model description information is output to the outside, each piece of model description information may be further sorted according to the above-mentioned sorting information determined by the server, or: and the vehicle-mounted terminal directly sorts the description information of each model according to the preference of the user, the configuration of the vehicle and other conditions, and then outputs the description information of each model according to the sorting result.
Referring to fig. 5, the processing method using the model library may further include: s203: determining that the first vehicle-mounted terminal is one of the federally learned nodes of the target model. It may be implemented after step S202 as shown in fig. 5, or before step S202.
In a specific implementation process, the process of step S203 may be confirmed by the user through an interactive interface, and further, the join confirmation information referred to later may be generated. Also, embodiments in which automatic confirmation without involving user intervention is not excluded, in which the user may not have the option to actively select whether to join.
After step S203, may include: s205: sending adding confirmation information to a server of the operation and maintenance model library; the corresponding server may implement steps S106 and S104.
FIG. 6 is a third flowchart illustrating a processing method using a model library according to an embodiment of the present invention;
referring to fig. 6, after step S205, the method may further include:
s206: training a target model in the first vehicle-mounted terminal to obtain a model or parameters after node training;
s207: sending the model after the node training or the parameter after the node training to the server; further, the server may implement step S108;
s208: and updating the target model in the first vehicle-mounted terminal according to the updated target model in the model library.
The specific implementation process of step S208 may, for example, directly receive the updated target model in the model library, and further replace the original target model in the first vehicle-mounted terminal; step S208 may also, for example, receive update information that may be used to implement model update, so as to update the target model in the first vehicle-mounted terminal to the target model in the model library according to the update information.
In summary, in the processing method using the model library provided in this embodiment, the server may operate and maintain the model libraries with various functions, and meanwhile, the models in the model library may be sent to the corresponding vehicle-mounted terminal according to the requirements of the vehicle-mounted terminal, and then the vehicle-mounted terminal may use the local models to complete the corresponding functions, without uploading data to the server, thereby ensuring the privacy of the user and the security of the data.
Meanwhile, due to the establishment of the model base, the server can integrate various models, and based on the model base, the vehicle-mounted terminal can correspondingly download the required models, so that the functions of the vehicle-mounted terminal can be expanded through the downloading and the use of the models.
In addition, in this embodiment, the vehicle-mounted terminal may also be used as a node for federal learning of the model, and further, the vehicle-mounted terminal serving as the node may train the model by using its own data, thereby providing a basis for updating the model in the server. On the basis, each model of the server can be updated iteratively based on the training result of the model in the vehicle-mounted terminal, and the server still cannot directly receive related data serving as a training material while the iterative updating is realized, so that the privacy of a user and the safety of the data are further guaranteed.
FIG. 7 is a first flowchart illustrating a first exemplary embodiment of a processing apparatus for an operation and maintenance model library; fig. 8 is a schematic diagram of a second program module of the processing apparatus of the operation and maintenance model library according to an embodiment of the present invention.
Referring to fig. 7, the operation and maintenance model library processing apparatus 300 includes:
a model library determining module 301, configured to determine a current model library, where the model library has N models, and the models can be updated in a federal learning manner, where N is an integer greater than or equal to 1;
a selected information receiving module 302, configured to receive model selected information sent by any one of the first vehicle-mounted terminals; the model selection information is used for representing a target model selected by the first vehicle-mounted terminal in the N models;
a model sending module 303, configured to send the target model to the first vehicle-mounted terminal, so that the first vehicle-mounted terminal can use the target model;
a server-side node determining module 304, configured to determine that the first vehicle-mounted terminal is one of the nodes learned by the federation of the target model.
Optionally, the models in the model library include at least one of the following models:
the system comprises an entertainment system model, a chassis adjustment calculation model, a driving strategy calculation model, an engine fuel-saving strategy calculation model, a motor power-saving strategy calculation model, a fatigue driving evaluation model, a driver physical condition evaluation model and a driving comfort evaluation model.
Optionally, referring to fig. 8, the operation and maintenance model library processing apparatus 300 may further include:
a node training receiving module 307, configured to receive a node-trained model or a node-trained parameter obtained by training the target model by the first vehicle-mounted terminal or another second vehicle-mounted terminal; the second on-board terminal is also one of the federal learning nodes of the target model;
and the model base updating module 308 is configured to update the target model in the model base according to the model after the node training or the parameter after the node training.
Optionally, referring to fig. 8, the operation and maintenance model library processing apparatus 300 may further include:
and the server side node updating module is used for updating the target model in each vehicle-mounted terminal of the federal learning node as the target model according to the updated target model in the model base.
Optionally, referring to fig. 8, the operation and maintenance model library processing apparatus 300 may further include:
a description information sending module 305, configured to send model description information of at least one of the N models to each vehicle-mounted terminal, so that each vehicle-mounted terminal can determine the model selection information according to the model description information.
Optionally, referring to fig. 8, the operation and maintenance model library processing apparatus 300 may further include:
a joining confirmation receiving module 306, configured to receive joining confirmation information sent by the first vehicle-mounted terminal and used for representing that the first vehicle-mounted terminal joins the federal learning of the target model.
In summary, in the processing apparatus for an operation and maintenance model library provided in this embodiment, a server may operate and maintain the model libraries with various functions, and meanwhile, the models in the model libraries may be sent to the corresponding vehicle-mounted terminals according to the requirements of the vehicle-mounted terminals, and then the vehicle-mounted terminals may use local models to complete corresponding functions, without uploading data to a server, thereby ensuring privacy of users and security of data.
Meanwhile, due to the establishment of the model base, the server can integrate various models, and based on the model base, the vehicle-mounted terminal can correspondingly download the required models, so that the functions of the vehicle-mounted terminal can be expanded through the downloading and the use of the models.
In addition, in this embodiment, the vehicle-mounted terminal may also be used as a node for federal learning of the model, and further, the vehicle-mounted terminal serving as the node may train the model by using its own data, thereby providing a basis for updating the model in the server. On the basis, each model of the server can be updated iteratively based on the training result of the model in the vehicle-mounted terminal, and the server still cannot directly receive related data serving as a training material while the iterative updating is realized, so that the privacy of a user and the safety of the data are further guaranteed.
FIG. 9 is a first block diagram illustrating program modules of a processing device using a model library according to an embodiment of the present invention; FIG. 10 is a second flowchart of a processing apparatus using a model library according to an embodiment of the present invention.
Referring to fig. 9, a processing apparatus 400 using a model library includes:
a selected information sending module 401, configured to send model selected information to a server of an operation and maintenance model library, where the model library has N models, and the model selected information is used to represent a target model selected by a first vehicle-mounted terminal in the N models;
a model receiving module 402, configured to receive the target model sent by the server and enable the target model to be used by the first vehicle-mounted terminal;
a terminal side node determining module 403, configured to determine that the first vehicle-mounted terminal is one of the nodes learned by the federation of the target model.
Optionally, referring to fig. 10, the processing apparatus 400 using the model library may further include:
a description information receiving module 404, configured to receive model description information of at least one model of the N models sent by the server.
Optionally, referring to fig. 10, the processing apparatus 400 using the model library may further include:
and a join confirmation sending module 405, configured to send join confirmation information to the server of the operation and maintenance model library.
Optionally, referring to fig. 10, the processing apparatus 400 using the model library may further include:
a node training module 406, configured to train a target model in the first vehicle-mounted terminal to obtain a model after node training or a parameter after node training;
a node training sending module 407, configured to send the node-trained model or the node-trained parameter to the server;
and the terminal side node updating module 408 is configured to update the target model in the first vehicle-mounted terminal according to the updated target model in the model library.
In summary, in the processing apparatus using the model library provided in this embodiment, the server may operate and maintain the model libraries with various functions, and meanwhile, the models in the model libraries may be sent to the corresponding vehicle-mounted terminals according to the requirements of the vehicle-mounted terminals, and then the vehicle-mounted terminals may use the local models to complete the corresponding functions, without uploading data to the server, thereby ensuring the privacy of the user and the security of the data.
Meanwhile, due to the establishment of the model base, the server can integrate various models, and based on the model base, the vehicle-mounted terminal can correspondingly download the required models, so that the functions of the vehicle-mounted terminal can be expanded through the downloading and the use of the models.
In addition, in this embodiment, the vehicle-mounted terminal may also be used as a node for federal learning of the model, and further, the vehicle-mounted terminal serving as the node may train the model by using its own data, thereby providing a basis for updating the model in the server. On the basis, each model of the server can be updated iteratively based on the training result of the model in the vehicle-mounted terminal, and the server still cannot directly receive related data serving as a training material while the iterative updating is realized, so that the privacy of a user and the safety of the data are further guaranteed.
Fig. 11 is a schematic structural diagram of an electronic device according to an embodiment of the invention.
Referring to fig. 11, an electronic device 50 is provided, including:
a processor 51; and the number of the first and second groups,
a memory 52 for storing executable instructions of the processor;
wherein the processor 51 is configured to perform the above-mentioned method via execution of the executable instructions.
The processor 51 is capable of communicating with the memory 52 via a bus 53.
The present embodiments also provide a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the above-mentioned method.
Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (11)
1. A processing method of an operation and maintenance model library is applied to a server and is characterized by comprising the following steps:
determining a current model library, wherein the model library has N models, the models can be updated in a federal learning mode, and N is an integer greater than or equal to 1;
receiving model selection information sent by any one first vehicle-mounted terminal; the model selection information is used for representing a target model selected by the first vehicle-mounted terminal in the N models;
sending the target model to the first vehicle-mounted terminal so that the first vehicle-mounted terminal can use the target model;
determining that the first vehicle-mounted terminal is one of the federally learned nodes of the target model.
2. The method of claim 1, wherein the models in the model library comprise at least one of:
the system comprises an entertainment system model, a chassis adjustment calculation model, a driving strategy calculation model, an engine fuel-saving strategy calculation model, a motor power-saving strategy calculation model, a fatigue driving evaluation model, a driver physical condition evaluation model and a driving comfort evaluation model.
3. The method of claim 1, after determining that the first on-board terminal is one of the federately learned nodes of the target model, further comprising:
receiving a node-trained model or node-trained parameters obtained by training the target model by the first vehicle-mounted terminal or another second vehicle-mounted terminal; the second on-board terminal is also one of the federal learning nodes of the target model;
and updating the target model in the model base according to the model after the node training or the parameter after the node training.
4. The method of claim 3, further comprising, after updating the target model in the model library according to the node-trained model or the node-trained parameter:
and updating the target model in each vehicle-mounted terminal as the federal learning node of the target model according to the updated target model in the model base.
5. The method according to any one of claims 1 to 4, wherein before receiving the model selection information sent by any one of the target vehicle-mounted terminals, the method further comprises:
and sending model description information of at least one model in the N models to each vehicle-mounted terminal so that each vehicle-mounted terminal can determine the model selection information according to the model description information.
6. The method of any one of claims 1 to 4, wherein prior to determining that the first on-board terminal is one of the federately learned nodes of the target model, further comprising:
and receiving join confirmation information which is sent by the first vehicle-mounted terminal and used for representing that the first vehicle-mounted terminal joins the federal study of the target model.
7. A processing method using a model library is applied to a first vehicle-mounted terminal, and is characterized by comprising the following steps:
sending model selection information to a server of an operation and maintenance model library, wherein the model library comprises N models, and the model selection information is used for representing a target model selected by the first vehicle-mounted terminal in the N models;
receiving the target model sent by the server and enabling the target model to be used by the first vehicle-mounted terminal;
determining that the first vehicle-mounted terminal is one of the federally learned nodes of the target model.
8. An operation and maintenance model library processing device, comprising:
the model library determining module is used for determining a current model library, wherein the model library is provided with N models which can be updated in a federal learning mode, and N is an integer greater than or equal to 1;
the selected information receiving module is used for receiving model selected information sent by any one of the first vehicle-mounted terminals; the model selection information is used for representing a target model selected by the first vehicle-mounted terminal in the N models;
a model sending module, configured to send the target model to the first vehicle-mounted terminal, so that the first vehicle-mounted terminal can use the target model;
and the server side node determining module is used for determining that the first vehicle-mounted terminal is one of the nodes for federal learning of the target model.
9. A processing apparatus that uses a model library, comprising:
the selected information sending module is used for sending model selected information to a server of an operation and maintenance model library, the model library is provided with N models, and the model selected information is used for representing a target model selected by a first vehicle-mounted terminal in the N models;
the model receiving module is used for receiving the target model sent by the server and enabling the target model to be used by the first vehicle-mounted terminal;
and the terminal side node determining module is used for determining that the first vehicle-mounted terminal is one of the nodes of the federal learning of the target model.
10. An electronic device, comprising a memory and a processor,
the memory is used for storing codes;
the processor is configured to execute the codes in the memory to implement the processing method of the operation and maintenance model library according to any one of claims 1 to 6, or the processing method using the model library according to claim 7.
11. A storage medium having a program stored thereon, wherein the program, when executed by a processor, implements the method for processing an operation and maintenance model library according to any one of claims 1 to 6, or the method for processing using a model library according to claim 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911202049.4A CN110990870A (en) | 2019-11-29 | 2019-11-29 | Operation and maintenance, processing method, device, equipment and medium using model library |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911202049.4A CN110990870A (en) | 2019-11-29 | 2019-11-29 | Operation and maintenance, processing method, device, equipment and medium using model library |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110990870A true CN110990870A (en) | 2020-04-10 |
Family
ID=70088333
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911202049.4A Pending CN110990870A (en) | 2019-11-29 | 2019-11-29 | Operation and maintenance, processing method, device, equipment and medium using model library |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110990870A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112863175A (en) * | 2020-12-31 | 2021-05-28 | 平安科技（深圳）有限公司 | Automobile road monitoring data processing method, device, equipment and storage medium |
| WO2022077232A1 (en) * | 2020-10-13 | 2022-04-21 | 北京小米移动软件有限公司 | Wireless communication method and apparatus, communication device, and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766940A (en) * | 2017-11-20 | 2018-03-06 | 北京百度网讯科技有限公司 | Method and apparatus for generation model |
| CN107944566A (en) * | 2017-11-28 | 2018-04-20 | 杭州云脑科技有限公司 | A kind of machine learning method, host node, working node and system |
| CN110262819A (en) * | 2019-06-04 | 2019-09-20 | 深圳前海微众银行股份有限公司 | A kind of the model parameter update method and device of federal study |
| CN110363305A (en) * | 2019-07-17 | 2019-10-22 | 深圳前海微众银行股份有限公司 | Federal learning method, system, terminal device and storage medium |
| US20190340534A1 (en) * | 2016-09-26 | 2019-11-07 | Google Llc | Communication Efficient Federated Learning |
| CN110442457A (en) * | 2019-08-12 | 2019-11-12 | 北京大学深圳研究生院 | Model training method, device and server based on federation's study |
-
2019
- 2019-11-29 CN CN201911202049.4A patent/CN110990870A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190340534A1 (en) * | 2016-09-26 | 2019-11-07 | Google Llc | Communication Efficient Federated Learning |
| CN107766940A (en) * | 2017-11-20 | 2018-03-06 | 北京百度网讯科技有限公司 | Method and apparatus for generation model |
| CN107944566A (en) * | 2017-11-28 | 2018-04-20 | 杭州云脑科技有限公司 | A kind of machine learning method, host node, working node and system |
| CN110262819A (en) * | 2019-06-04 | 2019-09-20 | 深圳前海微众银行股份有限公司 | A kind of the model parameter update method and device of federal study |
| CN110363305A (en) * | 2019-07-17 | 2019-10-22 | 深圳前海微众银行股份有限公司 | Federal learning method, system, terminal device and storage medium |
| CN110442457A (en) * | 2019-08-12 | 2019-11-12 | 北京大学深圳研究生院 | Model training method, device and server based on federation's study |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022077232A1 (en) * | 2020-10-13 | 2022-04-21 | 北京小米移动软件有限公司 | Wireless communication method and apparatus, communication device, and storage medium |
| CN112863175A (en) * | 2020-12-31 | 2021-05-28 | 平安科技（深圳）有限公司 | Automobile road monitoring data processing method, device, equipment and storage medium |
| CN112863175B (en) * | 2020-12-31 | 2022-11-22 | 平安科技（深圳）有限公司 | Automobile road monitoring data processing method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111091200B (en) | Updating method and system of training model, intelligent device, server and storage medium | |
| DE102015208750A1 (en) | OVER-THE-AIR VEHICLE PROBLEM SOLVING | |
| CN102901511B (en) | Navigation method and system, terminal and mobile terminal | |
| DE102014202306A1 (en) | System and method for a man-machine interface | |
| CN111008704B (en) | Processing method, device, equipment and storage medium for federal learning of electric automobile | |
| CN110990870A (en) | Operation and maintenance, processing method, device, equipment and medium using model library | |
| DE102014219540A1 (en) | Method and apparatus for on-demand wireless module update | |
| CN103701914B (en) | Off-line downloading method and device for serial resources | |
| CN104581640A (en) | Cloud server, system and method for realizing Internet of Vehicles pickup function | |
| CN110262413A (en) | Intelligent home furnishing control method, control device, car-mounted terminal and readable storage medium storing program for executing | |
| CN111527389A (en) | Vehicle diagnosis method, vehicle diagnosis device and storage medium | |
| CN103198156A (en) | Automobile maintenance information inquiry and feedback methods and server | |
| KR20200075626A (en) | Wireless update system and method capable of ecu update of vehicle | |
| CN115912526A (en) | Terminal charging method and device, electronic equipment and storage medium | |
| CN110430553B (en) | Interaction method and device between vehicles, storage medium and control terminal | |
| CN104485011A (en) | Cloud server, system and method for realizing Internet of vehicles (IoV) pickup function | |
| CN108062361A (en) | Early education equipment, resource supplying system and the method for updating early education device resource | |
| CN105847350A (en) | Vehicle device parameter setting method, terminal, vehicle and vehicle control system | |
| CN111240351A (en) | Method and device for controlling unmanned aerial vehicle based on vehicle, storage medium and vehicle | |
| CN104205783A (en) | Mobile terminal control apparatus | |
| CN108415983B (en) | Intelligent problem solving method and device based on interaction | |
| CN111123727B (en) | Unmanned vehicle simulation building method, device, equipment and computer readable medium | |
| CN112182431A (en) | Reference point recommendation method, device, equipment and storage medium | |
| CN110674275A (en) | Knowledge question answering method and device | |
| CN109511087A (en) | Unmanned vehicle website broadcasting method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200410 |
|
| RJ01 | Rejection of invention patent application after publication |