CN117993275A - Training method and device of data generation model, electronic equipment and storage medium - Google Patents
Training method and device of data generation model, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117993275A CN117993275A CN202211332201.2A CN202211332201A CN117993275A CN 117993275 A CN117993275 A CN 117993275A CN 202211332201 A CN202211332201 A CN 202211332201A CN 117993275 A CN117993275 A CN 117993275A
- Authority
- CN
- China
- Prior art keywords
- data
- model
- training
- data generation
- generation model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012549 training Methods 0.000 title claims abstract description 126
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000004590 computer program Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 abstract description 11
- 238000005516 engineering process Methods 0.000 description 9
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 238000013473 artificial intelligence Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013506 data mapping Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005057 refrigeration Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Feedback Control In General (AREA)
Abstract
The disclosure discloses a training method and device of a data generation model, electronic equipment and a storage medium, and relates to the field of data processing, comprising the following steps: the training data is input into a data generation model, prediction data is obtained, the prediction data is input into a judgment model, a first judgment result of the data judgment model on the prediction data is obtained, model parameters of the data generation model are adjusted according to the first judgment result, the prediction data is generated according to the adjusted data generation model and the training data, the data generation model and the data judgment model are subjected to countermeasure training, the data generation model is constrained by the data judgment model obtained by training the real data under different driving conditions, the data generation model is updated according to the first judgment result, the similarity between the generated data and the real data is improved, the generation capacity of the data generation model is improved, and the accuracy of the model obtained based on the generated data training is further guaranteed.
Description
Technical Field
The disclosure relates to the technical field of data processing, and in particular relates to a training method and device of a data generation model, electronic equipment and a storage medium.
Background
The building of the model generally requires a large amount of data in a database as support, and the data source of the database generally obtains real vehicle data by a mode, but the data amount of the real vehicle data obtained by the mode is smaller.
In order to solve the problem that the data volume of real vehicle data in the database is smaller, data similar to the real vehicle data can be used as a data source of the database, and more data can be provided for the database by acquiring the similar data.
Although the method increases the data volume of the database, the accuracy of the model obtained by training based on the similar data is poor due to the fact that errors of the similar data and the real data of the vehicle are large.
Disclosure of Invention
The disclosure provides a training method and device of a data generation model, electronic equipment and a storage medium. The method mainly aims to realize the expansion of data in a database on the premise of ensuring the model training effect.
According to a first aspect of the present disclosure, there is provided a training method of a data generation model, including:
Model countermeasure training step:
inputting training data into a latest data generation model to obtain prediction data; the training data are data of different driving conditions obtained from related vehicles, wherein the related vehicles are vehicles with the degree of correlation with the target vehicle higher than a preset degree of correlation threshold;
Inputting the predicted data into a data discrimination model, and obtaining a first discrimination result of the data discrimination model on the predicted data; the data discrimination model is obtained by training real data of different driving conditions;
Updating the data generation model according to the first judging result;
If the latest data generation model meets the preset model requirement, then
Ending the model countermeasure training step, and taking the latest data generation model as a trained data generation model; otherwise
The model challenge training step continues to be performed.
Optionally, the updating the data generation model according to the first discrimination result further includes:
Determining a first error gradient of the data generation model according to the first discrimination result;
And adjusting model parameters of the data generation model according to the first error gradient, and updating the data generation model.
Optionally, the preset model requirement is specifically: the first error gradient is less than a preset error gradient threshold.
Optionally, before inputting the predicted data into a data discrimination model and obtaining a first discrimination result of the data discrimination model on the predicted data, the method further includes:
After the real data of different driving conditions of the target vehicle are input into the data discrimination model, the real data of the different driving conditions are discriminated to obtain a second discrimination result of the real data;
determining a second error gradient of the data discrimination model according to the second discrimination result;
and adjusting model parameters of the data discrimination model according to the second error gradient, and updating the data discrimination model.
Optionally, the method further comprises:
Receiving a predicted data acquisition request corresponding to a target working condition of a target vehicle;
And responding to the acquisition request, and outputting predicted data corresponding to the target working condition of the target vehicle based on the trained data generation model.
Optionally, the target condition includes a driving condition related to a driving range.
According to a second aspect of the present disclosure, there is provided a training apparatus of a data generation model, comprising:
Model countermeasure training device:
The input unit is used for inputting training data into the latest data generation model to obtain prediction data; the training data are data of different driving conditions obtained from related vehicles, wherein the related vehicles are vehicles with the degree of correlation with the target vehicle higher than a preset degree of correlation threshold;
The first judging unit is used for inputting the predicted data into a data judging model and obtaining a first judging result of the data judging model on the predicted data; the data discrimination model is obtained by training real data of different driving conditions;
the updating unit is used for updating the data generation model according to the first judging result;
the judging unit is used for ending the model countermeasure training step when the latest data generating model meets the preset model requirement, and taking the latest data generating model as a trained data generating model; otherwise, continuing to adopt the model countermeasure training device to execute the model countermeasure training step.
Optionally, the updating unit further includes:
the determining module is used for determining a first error gradient of the data generation model according to the first judging result;
And the updating module is used for adjusting the model parameters of the data generation model according to the first error gradient and updating the data generation model.
Optionally, the preset model requirement is specifically: the first error gradient is less than a preset error gradient threshold.
Optionally, before the first discriminating unit, the apparatus further includes:
the second judging unit is used for inputting the real data of different driving conditions of the target vehicle into the data judging model, and judging the real data of different driving conditions to obtain a second judging result of the real data;
The determining unit is used for determining a second error gradient of the data discrimination model according to the second discrimination result;
And the adjusting unit is used for adjusting the model parameters of the data discrimination model according to the second error gradient and updating the data discrimination model.
Optionally, the apparatus further includes:
the receiving unit is used for receiving a predicted data acquisition request corresponding to the target working condition of the target vehicle;
And the output unit is used for responding to the acquisition request, outputting the predicted data corresponding to the target working condition of the target vehicle based on the trained data generation model.
Optionally, the target condition includes a driving condition related to a driving range.
According to a third aspect of the present disclosure, there is provided an electronic device comprising:
at least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of the first aspect.
According to a fourth aspect of the present disclosure, there is provided a non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of the preceding first aspect.
According to a fifth aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the method of the first aspect described above.
The training method and device for the data generation model, the electronic equipment and the storage medium provided by the disclosure have the main technical scheme that: model countermeasure training step: inputting training data into a latest data generation model to obtain prediction data; the training data are data of different driving conditions obtained from related vehicles, wherein the related vehicles are vehicles with the degree of correlation with the target vehicle higher than a preset degree of correlation threshold; inputting the predicted data into a data discrimination model, and obtaining a first discrimination result of the data discrimination model on the predicted data; the data discrimination model is obtained by training real data of different driving conditions; updating the data generation model according to the first judging result; if the latest data generation model meets the preset model requirement, ending the model countermeasure training step, and taking the latest data generation model as a trained data generation model; otherwise, continuing to execute the model countermeasure training step. Compared with the related art, the embodiment of the application has the advantages that through the countermeasure training of the data generation model and the data discrimination model, the data generation model is constrained by the data discrimination model obtained through the real data training based on different driving conditions, the data generation model is updated according to the first discrimination result, the similarity between the generated data and the real data is improved, the generation capacity of the data generation model is improved, the error between the generated data and the real data is eliminated, and the accuracy of the model obtained through the generated data training is further ensured.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.
Drawings
The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
fig. 1 is a flow chart of a training method of a data generation model according to an embodiment of the disclosure;
FIG. 2 is a flow chart of a training method for a data discrimination model according to an embodiment of the disclosure;
FIG. 3 is a schematic flow chart of training a data generation model and a data discrimination model according to an embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of a training device of a data generation model according to an embodiment of the present disclosure;
FIG. 5 is a schematic structural diagram of another training device for a data generation model according to an embodiment of the present disclosure;
Fig. 6 is a schematic block diagram of an example electronic device provided by an embodiment of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
Training methods and apparatuses for a data generation model, electronic devices, and storage media of embodiments of the present disclosure are described below with reference to the accompanying drawings.
Fig. 1 is a flowchart of a training method of a data generation model according to an embodiment of the present disclosure.
As shown in fig. 1, the method comprises the following model challenge training steps:
Step 101, inputting training data into a latest data generation model to obtain prediction data; the training data are data of different driving conditions obtained from related vehicles, and the related vehicles are vehicles with the correlation degree with the target vehicle higher than a preset correlation degree threshold value.
The related vehicles can be vehicles similar to the types, weights, battery capacities and battery types of the target vehicles, the more similar points are, the more similar the predicted data of the related vehicles are to the predicted data of the target vehicles, in actual use, the related parameters of the vehicles such as the types, weights and battery capacities can be given scores, the similarity between each parameter of the target vehicles and each parameter of the related vehicles is assessed, and finally the correlation between the target vehicles and the related vehicles is obtained; the correlation preset correlation threshold is a tested value, and can be set according to requirements, for example, the correlation preset correlation threshold can be set to 90% or 95%, however, it is to be noted that the higher the similarity threshold is, the smaller the error value between the predicted data output by the trained model and the real data of the target vehicle is.
The type of the prediction data can be set according to the requirements of the user, for example, the trained data generation model is used for predicting the driving range of the vehicle under different working conditions, and the type of the prediction data can be the driving range prediction data of the related vehicle under different working conditions; the type of the prediction data can be the prediction data of the power consumption of the refrigeration system of the related vehicle under different working conditions, the embodiment of the application uses the type of the prediction data as the driving range prediction data of the related vehicle under different working conditions as an example for explanation, but the description mode is not a limitation of application scenes, and the embodiment of the application does not limit the limitation.
102, Inputting the predicted data into a data discrimination model, and obtaining a first discrimination result of the data discrimination model on the predicted data; the data discrimination model is obtained by training the real data of different driving conditions.
The method comprises the steps that driving range prediction data of a target vehicle under different working conditions generated by a data generation model are input into a data judgment model, the data judgment model judges according to working condition information in the data and the driving range prediction data, and whether the current data is real data or data generated by the data generation model is judged; the data discrimination model is trained in advance by real data under different working conditions, and has the recognition capability for the driving mileage data under the trained working conditions.
And step 103, updating the data generation model according to the first judging result.
When the data generation model is updated according to the first judging result, the parameters of the data generation model can be updated according to methods such as a loss function calculation method or an error gradient calculation method and the like according to the first judging result, and the generation capacity of the data generation model is enhanced.
104, Ending the model countermeasure training step if the latest data generation model meets the preset model requirement, and taking the latest data generation model as a trained data generation model; otherwise, continuing to execute the model countermeasure training step.
The preset model requirement is a preset condition, and can be set according to actual requirements, for example, the generating capability of the data generating model can be set to meet the preset condition, or the error value of the predicted data and the real data generated by the data generating model is smaller than a threshold value, etc., which is not limited in the embodiment of the present application.
And judging whether the data generation model after the updating parameters are judged meets the condition of finishing training or not.
The training method of the data generation model provided by the disclosure mainly comprises the following steps: model countermeasure training step: inputting training data into a latest data generation model to obtain prediction data; the training data are data of different driving conditions obtained from related vehicles, wherein the related vehicles are vehicles with the degree of correlation with the target vehicle higher than a preset degree of correlation threshold; inputting the predicted data into a data discrimination model, and obtaining a first discrimination result of the data discrimination model on the predicted data; the data discrimination model is obtained by training real data of different driving conditions; updating the data generation model according to the first judging result; if the latest data generation model meets the preset model requirement, ending the model countermeasure training step, and taking the latest data generation model as a trained data generation model; otherwise, continuing to execute the model countermeasure training step. Compared with the related art, the embodiment of the application has the advantages that through the countermeasure training of the data generation model and the data discrimination model, the data generation model is constrained by the data discrimination model obtained through the real data training based on different driving conditions, the data generation model is updated according to the first discrimination result, the similarity between the generated data and the real data is improved, the generation capacity of the data generation model is improved, the error between the generated data and the real data is eliminated, and the accuracy of the model obtained through the generated data training is further ensured.
In one implementation manner of the embodiment of the present application, when updating the data generation model, the following method may be adopted: determining a first error gradient of the data generation model according to the first discrimination result; and adjusting model parameters of the data generation model according to the first error gradient, and updating the data generation model.
As an implementation manner of the embodiment of the present application, when step 104 determines whether to stop training, it may be determined whether the first error gradient is smaller than a preset error gradient threshold value; the higher the preset error gradient threshold is a tested value, the higher the accuracy of the predicted data generated by the training completion data generation model is, and in practical application, the preset error gradient threshold can be set to be 95% or 97%, which is not limited by the embodiment of the present application.
As an implementation manner of the embodiment of the present application, after the model parameters of the data generation model are adjusted according to the first error gradient, the prediction data is continuously determined based on the data determination model after the model parameters are adjusted until the first error gradient is smaller than a preset error gradient threshold, and the countermeasure training for the data generation model and the data determination model is stopped.
The data generation model and the data discrimination model complement each other, and the generation/discrimination capability of the data generation model and the data discrimination model is continuously improved in the countermeasure training, and if one party has the generation/discrimination capability far higher than the discrimination/generation capability of the other party, the training effect is poor, even the training is carried out in the wrong direction, so that only one party cannot be independently trained when the data generation model and the data discrimination model are trained; as shown in fig. 2, fig. 2 is a flow chart of a training method for a data discrimination model according to an embodiment of the present application, including:
step 201, after inputting the real data of different driving conditions of the target vehicle into the data discrimination model, discriminating the real data of different driving conditions, and obtaining a second discrimination result of the real data.
Step 202, determining a second error gradient of the data discrimination model according to the second discrimination result.
And step 203, adjusting model parameters of the data discrimination model according to the second error gradient.
Please refer to step 102 to step 104 for the specific implementation of step 201 to step 203, which are the same, and detailed descriptions of the embodiments of the present application are omitted here.
In the embodiment of the present application, the mode of alternating training is described as an example, but the mode of description is not limited to a specific training mode, and the embodiment of the present application is not limited to this.
In order to facilitate understanding of training of the data discrimination model and the data generation model in the embodiment of the present application, an alternate training manner will be described below as an example, referring to fig. 3, fig. 3 is a schematic flow chart of training the data generation model and the data discrimination model provided in the present application, including:
step 301, obtaining real data of different driving conditions of a target vehicle through experiments;
Step 302, training data of corresponding driving conditions are obtained from related vehicles;
step 303, obtaining real data and training data under the same working condition;
It should be noted that, multiple sets of real data and training data may exist under the same working condition, and the data quantity under the same working condition is not limited in the embodiment of the present application.
Step 304, training the data discrimination model and the data generation model by alternating real data and training data under the same working condition;
Step 3041, inputting training data into a data generation model;
Step 3042, performing data mapping through a data generation model to generate prediction data of the target vehicle;
step 3043, judging the predicted data through a data judging model to generate a first judging result;
step 3044, obtaining a first error gradient through the first discrimination result, and transmitting the first error gradient back to the data generation model to adjust model parameters of the data generation model;
Step 3045, inputting the real data into a data discrimination model;
step 3046, judging the real data through a data judging model to generate a second judging result;
step 3047, obtaining a second error gradient through a second discrimination result, and transmitting the second error gradient back to the data discrimination model to adjust model parameters of the data discrimination model;
In the embodiment of the present application, a description mode of training the data generation model and then training the data discrimination model is adopted, but the description mode is not a limitation of the training sequence, and the embodiment of the present application does not limit the training sequence of the data generation model and the data discrimination model. Step 305, determining whether the first error gradient of the data discrimination model is smaller than a preset error gradient threshold, if so, executing step 306, and if not, executing step 304;
step 306, the model challenge training step is ended.
When determining whether training is completed, the method may determine whether the training is completed by using the error gradient of the data discrimination model and the preset error gradient threshold, or determine whether the data discrimination model can discriminate the real data from the predicted data according to whether the data discrimination model can discriminate the real data from the predicted data.
For the description of steps 301 to 306, reference may be made to the above embodiments, and the embodiments of the present application are not described herein.
As an implementation manner of the embodiment of the present application, after training of the data generation model and the data discrimination model is completed, the data generation model may be applied to data generation, and when generating data, the following method may be adopted: receiving a predicted data acquisition request corresponding to a target working condition of a target vehicle; and responding to the acquisition request, and outputting predicted data corresponding to the target working condition of the target vehicle based on the trained data generation model.
As an extension to the embodiments of the above application, the target conditions include driving conditions related to the driving range, such as a running form of the vehicle: starting, accelerating, uniform speed, decelerating, ascending and descending slopes and the like, and loading conditions of the vehicle such as no load, full load and overload; and data such as the temperature and the remaining power of the power battery.
Corresponding to the training method of the data generation model, the invention also provides a training device of the data generation model. Since the device embodiment of the present invention corresponds to the above-mentioned method embodiment, details not disclosed in the device embodiment may refer to the above-mentioned method embodiment, and details are not described in detail in the present invention.
Fig. 4 is a schematic structural diagram of a training device of a data generation model according to an embodiment of the present disclosure, where, as shown in fig. 4, the training device includes: model countermeasure training device:
An input unit 41 for inputting training data into the latest data generation model to obtain predicted data; the training data are data of different driving conditions obtained from related vehicles, wherein the related vehicles are vehicles with the degree of correlation with the target vehicle higher than a preset degree of correlation threshold;
A first discriminating unit 42, configured to input the predicted data into a data discriminating model, and obtain a first discriminating result of the data discriminating model on the predicted data; the data discrimination model is obtained by training real data of different driving conditions;
an updating unit 43, configured to update the data generation model according to the first discrimination result;
a judging unit 44, configured to end the model countermeasure training step when the latest data generation model meets the preset model requirement, and take the latest data generation model as a trained data generation model; otherwise, continuing to adopt the model countermeasure training device to execute the model countermeasure training step.
The training device of the data generation model provided by the disclosure comprises the following main technical scheme: model countermeasure training step: inputting training data into a latest data generation model to obtain prediction data; the training data are data of different driving conditions obtained from related vehicles, wherein the related vehicles are vehicles with the degree of correlation with the target vehicle higher than a preset degree of correlation threshold; inputting the predicted data into a data discrimination model, and obtaining a first discrimination result of the data discrimination model on the predicted data; the data discrimination model is obtained by training real data of different driving conditions; updating the data generation model according to the first judging result; if the latest data generation model meets the preset model requirement, ending the model countermeasure training step, and taking the latest data generation model as a trained data generation model; otherwise, continuing to execute the model countermeasure training step. Compared with the related art, the embodiment of the application has the advantages that through the countermeasure training of the data generation model and the data discrimination model, the data generation model is constrained by the data discrimination model obtained through the real data training based on different driving conditions, the data generation model is updated according to the first discrimination result, the similarity between the generated data and the real data is improved, the generation capacity of the data generation model is improved, the error between the generated data and the real data is eliminated, and the accuracy of the model obtained through the generated data training is further ensured.
Further, in a possible implementation manner of this embodiment, as shown in fig. 5, the updating unit 43 further includes:
a determining module 431, configured to determine a first error gradient of the data generating model according to the first discrimination result;
And the updating module 432 is configured to adjust model parameters of the data generation model according to the first error gradient, and update the data generation model.
Further, in one possible implementation manner of this embodiment, the preset model requirement is specifically: the first error gradient is less than a preset error gradient threshold.
Further, in one possible implementation manner of this embodiment, as shown in fig. 5, before the first determining unit 42, the apparatus further includes:
A second judging unit 45, configured to input real data of different driving conditions of the target vehicle into the data judging model, and then judge the real data of the different driving conditions to obtain a second judging result of the real data;
a determining unit 46, configured to determine a second error gradient of the data discrimination model according to the second discrimination result;
and an adjusting unit 47, configured to adjust the model parameters of the data discrimination model according to the second error gradient, and update the data discrimination model.
Further, in a possible implementation manner of this embodiment, as shown in fig. 5, the apparatus further includes:
A receiving unit 48, configured to receive a predicted data acquisition request corresponding to a target condition of a target vehicle;
And an output unit 49, configured to output, based on the trained data generation model, prediction data corresponding to a target condition of the target vehicle in response to the acquisition request.
Further, in a possible implementation manner of this embodiment, the target operating condition includes a driving condition related to a driving range.
The foregoing explanation of the method embodiment is also applicable to the apparatus of this embodiment, and the principle is the same, and this embodiment is not limited thereto.
According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.
Fig. 6 shows a schematic block diagram of an example electronic device 500 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.
As shown in fig. 6, the apparatus 500 includes a computing unit 501 that can perform various appropriate actions and processes according to a computer program stored in a ROM (Read-Only Memory) 502 or a computer program loaded from a storage unit 508 into a RAM (Random Access Memory ) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, ROM 502, and RAM 503 are connected to each other by a bus 504. An I/O (Input/Output) interface 505 is also connected to bus 504.
Various components in the device 500 are connected to the I/O interface 505, including: an input unit 506 such as a keyboard, a mouse, etc.; an output unit 507 such as various types of displays, speakers, and the like; a storage unit 508 such as a magnetic disk, an optical disk, or the like; and a communication unit 509 such as a network card, modem, wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.
The computing unit 501 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 501 include, but are not limited to, a CPU (Central Processing Unit ), a GPU (Graphic Processing Units, graphics processing unit), various specialized AI (ARTIFICIAL INTELLIGENCE ) computing chips, various computing units running machine learning model algorithms, a DSP (DIGITAL SIGNAL Processor ), and any suitable Processor, controller, microcontroller, etc. The computing unit 501 performs the respective methods and processes described above, for example, a training method of the data generation model. For example, in some embodiments, the training method of the data generation model may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program may be loaded and/or installed onto the device 500 via the ROM 502 and/or the communication unit 509. When the computer program is loaded into RAM 503 and executed by computing unit 501, one or more steps of the method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the foregoing training method of the data generation model in any other suitable way (e.g. by means of firmware).
Various implementations of the systems and techniques described here above can be implemented in digital electronic circuitry, integrated Circuit systems, FPGAs (Field Programmable GATE ARRAY, field programmable gate arrays), ASICs (Application-SPECIFIC INTEGRATED circuits), ASSPs (Appl ication SPECIFIC STANDARD products, application-specific standard products), SOCs (systems On Chip systems), CPLDs (Complex Programmable Logic Device, complex programmable logic devices), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, RAM, ROM, EPROM (ELECTRICALLY PROGRAMMABLE READ-Only-Memory, erasable programmable read-Only Memory) or flash Memory, an optical fiber, a CD-ROM (Compact Disc Read-Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., CRT (Cathode-Ray Tube) or LCD (Liquid CRYSTAL DISPLAY) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: LAN (Local Area Network ), WAN (Wide Area Network, wide area network), internet and blockchain networks.
The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service ("Virtual PRIVATE SERVER" or simply "VPS") are overcome. The server may also be a server of a distributed system or a server that incorporates a blockchain.
It should be noted that, artificial intelligence is a subject of studying a certain thought process and intelligent behavior (such as learning, reasoning, thinking, planning, etc.) of a computer to simulate a person, and has a technology at both hardware and software level. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing, and the like; the artificial intelligence software technology mainly comprises a computer vision technology, a voice recognition technology, a natural language processing technology, a machine learning/deep learning technology, a big data processing technology, a knowledge graph technology and the like.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel or sequentially or in a different order, provided that the desired results of the technical solutions of the present disclosure are achieved, and are not limited herein.
The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.
Claims (10)
1. A method of training a data generation model, comprising:
Model countermeasure training step:
inputting training data into a latest data generation model to obtain prediction data; the training data are data of different driving conditions obtained from related vehicles, wherein the related vehicles are vehicles with the degree of correlation with the target vehicle higher than a preset degree of correlation threshold;
Inputting the predicted data into a data discrimination model, and obtaining a first discrimination result of the data discrimination model on the predicted data; the data discrimination model is obtained by training real data of different driving conditions;
Updating the data generation model according to the first judging result;
if the latest data generation model meets the preset model requirement, ending the model countermeasure training step, and taking the latest data generation model as a trained data generation model; otherwise, continuing to execute the model countermeasure training step.
2. The method of claim 1, wherein updating the data generation model based on the first discrimination result further comprises:
Determining a first error gradient of the data generation model according to the first discrimination result;
And adjusting model parameters of the data generation model according to the first error gradient, and updating the data generation model.
3. The method according to claim 2, wherein the predetermined model requirements are specifically: the first error gradient is less than a preset error gradient threshold.
4. The method of claim 1, wherein prior to inputting the predicted data into a data discrimination model to obtain a first discrimination result of the data discrimination model for the predicted data, the method further comprises:
After the real data of different driving conditions of the target vehicle are input into the data discrimination model, the real data of the different driving conditions are discriminated to obtain a second discrimination result of the real data;
determining a second error gradient of the data discrimination model according to the second discrimination result;
and adjusting model parameters of the data discrimination model according to the second error gradient, and updating the data discrimination model.
5. The method according to any one of claims 1-4, further comprising:
Receiving a predicted data acquisition request corresponding to a target working condition of a target vehicle;
And responding to the acquisition request, and outputting predicted data corresponding to the target working condition of the target vehicle based on the trained data generation model.
6. The method of claim 5, wherein the target operating condition comprises a driving condition related to range.
7. A training device for a data generation model, comprising:
Model countermeasure training device:
The input unit is used for inputting training data into the latest data generation model to obtain prediction data; the training data are data of different driving conditions obtained from related vehicles, wherein the related vehicles are vehicles with the degree of correlation with the target vehicle higher than a preset degree of correlation threshold;
The first judging unit is used for inputting the predicted data into a data judging model and obtaining a first judging result of the data judging model on the predicted data; the data discrimination model is obtained by training real data of different driving conditions;
the updating unit is used for updating the data generation model according to the first judging result;
the judging unit is used for ending the model countermeasure training step when the latest data generation model meets the preset model requirement, and taking the latest data generation model as a trained data generation model; otherwise, continuing to adopt the model countermeasure training device to execute the model countermeasure training step.
8. An electronic device, comprising:
at least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.
9. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-6.
10. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211332201.2A CN117993275A (en) | 2022-10-28 | 2022-10-28 | Training method and device of data generation model, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211332201.2A CN117993275A (en) | 2022-10-28 | 2022-10-28 | Training method and device of data generation model, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117993275A true CN117993275A (en) | 2024-05-07 |
Family
ID=90893932
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211332201.2A Pending CN117993275A (en) | 2022-10-28 | 2022-10-28 | Training method and device of data generation model, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117993275A (en) |
-
2022
- 2022-10-28 CN CN202211332201.2A patent/CN117993275A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115221795A (en) | Training method, prediction method, device, equipment and medium of capacity prediction model | |
| CN113627536A (en) | Model training method, video classification method, device, equipment and storage medium | |
| CN113904943B (en) | Account detection method and device, electronic equipment and storage medium | |
| CN112528995B (en) | Method for training target detection model, target detection method and device | |
| CN112784102B (en) | Video retrieval method and device and electronic equipment | |
| CN114417780A (en) | State synchronization method and device, electronic equipment and storage medium | |
| CN112632251B (en) | Reply content generation method, device, equipment and storage medium | |
| CN117474091A (en) | Knowledge graph construction method, device, equipment and storage medium | |
| CN113657468A (en) | Pre-training model generation method and device, electronic equipment and storage medium | |
| EP4092544A1 (en) | Method, apparatus and storage medium for deduplicating entity nodes in graph database | |
| CN117993275A (en) | Training method and device of data generation model, electronic equipment and storage medium | |
| CN113408304B (en) | Text translation method and device, electronic equipment and storage medium | |
| CN114048863A (en) | Data processing method, data processing device, electronic equipment and storage medium | |
| CN112966607A (en) | Model training method, face video generation method, device, equipment and medium | |
| CN113963433B (en) | Motion search method, motion search device, electronic equipment and storage medium | |
| CN116416500B (en) | Image recognition model training method, image recognition device and electronic equipment | |
| CN114926447B (en) | Method for training a model, method and device for detecting a target | |
| EP4036861A2 (en) | Method and apparatus for processing point cloud data, electronic device, storage medium, computer program product | |
| CN113361575B (en) | Model training method and device and electronic equipment | |
| CN115860055B (en) | Performance determination method, performance optimization method, device, electronic equipment and medium | |
| CN115860077B (en) | Method, device, equipment and storage medium for processing state data | |
| CN112579842A (en) | Model searching method, model searching apparatus, electronic device, storage medium, and program product | |
| CN117933353A (en) | Reinforced learning model training method and device, electronic equipment and storage medium | |
| CN117950306A (en) | PID control parameter determining method, device, equipment and medium | |
| CN117539602A (en) | Method and device for task speculation behind, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |