CN114819095A - Method, device and electronic device for generating business data processing model - Google Patents

Abstract

The utility model provides a method and a device for generating a business data processing model and an electronic device, which relate to the technical field of artificial intelligence, in particular to the technical fields of deep learning, computer vision and the like, and comprise the following steps: the method comprises the steps of determining model evaluation information of an initial business data processing model, determining a target operator according to the model evaluation information, determining target description information of the target operator according to the model evaluation information, and generating a target business data processing model according to the target description information and the target operator.

Description

Method, device and electronic device for generating business data processing model

technical field

The present disclosure relates to the technical field of artificial intelligence, in particular to the technical fields of deep learning and computer vision, and in particular, to a method, device and electronic device for generating a business data processing model.

Background technique

Artificial intelligence is the study of making computers to simulate certain thinking processes and intelligent behaviors of people (such as learning, reasoning, thinking, planning, etc.), both hardware-level technology and software-level technology. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, and big data processing; artificial intelligence software technologies mainly include computer vision technology, speech recognition technology, natural language processing technology, and machine learning/depth Learning, big data processing technology, knowledge graph technology and other major directions.

In the related art, when generating a business data processing model, the determination efficiency of the operators required by the model and the operator description information (the operator description information such as the hyperparameters of the operator) is not high, resulting in insufficient generation process of the target business data processing model. It is flexible and affects the generation efficiency of the target business data processing model.

SUMMARY OF THE INVENTION

The present disclosure provides a method, apparatus, electronic device, storage medium and computer program product for generating a business data processing model.

According to a first aspect of the present disclosure, a method for generating a business data processing model is provided, including: determining model evaluation information of an initial business data processing model; determining a target operator according to the model evaluation information; evaluating according to the model information, determine the target description information of the target operator; and generate a target service data processing model according to the target description information and the target operator.

According to a second aspect of the present disclosure, there is provided an apparatus for generating a business data processing model, comprising: a first determination module for determining model evaluation information of an initial business data processing model; a second determination module for model evaluation information to determine a target operator; a third determination module for determining target description information of the target operator according to the model evaluation information; and a generation module for determining the target description information and the target according to the target description information The operator generates the target business data processing model.

According to a third aspect of the present disclosure, there is provided an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor. The at least one processor executes to enable the at least one processor to execute the method for generating a business data processing model according to the first aspect of the present disclosure.

According to a fourth aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the method for generating a business data processing model according to the first aspect of the present disclosure.

According to a fifth aspect of the present disclosure, there is provided a computer program product, comprising a computer program that, when executed by a processor, implements the steps of the method for generating a business data processing model according to the first aspect of the present disclosure.

It should be understood that what is described in this section is not intended to identify key or critical features of embodiments of the disclosure, nor is it intended to limit the scope of the disclosure. Other features of the present disclosure will become readily understood from the following description.

Description of drawings

The accompanying drawings are used for better understanding of the present solution, and do not constitute a limitation to the present disclosure. in:

1 is a schematic diagram according to a first embodiment of the present disclosure;

2 is a schematic diagram according to a second embodiment of the present disclosure;

3 is a schematic diagram according to a third embodiment of the present disclosure;

4 is a schematic diagram according to a fourth embodiment of the present disclosure;

5 is a schematic diagram according to a fifth embodiment of the present disclosure;

6 is a schematic diagram according to a sixth embodiment of the present disclosure;

FIG. 7 is a schematic diagram according to a seventh embodiment of the present disclosure;

8 is a block diagram of an apparatus for generating an exemplary business data processing model in an embodiment of the present disclosure;

FIG. 9 shows a schematic block diagram of an example electronic device that can be used to implement the business data processing model generation method of the present disclosure.

Detailed ways

Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding and should be considered as exemplary only. Accordingly, those 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 from the following description for clarity and conciseness.

FIG. 1 is a schematic diagram according to a first embodiment of the present disclosure.

It should be noted that the execution body of the method for generating a business data processing model in this embodiment is a device for generating a business data processing model, the device may be implemented by software and/or hardware, and the device may be configured in an electronic device , the electronic device may include, but is not limited to, a terminal, a server, and the like.

The embodiments of the present disclosure relate to the technical field of artificial intelligence, and specifically to the technical fields of deep learning, computer vision, and the like.

Among them, artificial intelligence (Artificial Intelligence), the English abbreviation is AI. It is a new technical science that studies and develops theories, methods, techniques and application systems for simulating, extending and expanding human intelligence.

Deep learning is to learn the inherent laws and representation levels of sample data, and the information obtained during these learning processes is of great help to the interpretation of data such as text, images and sounds. The ultimate goal of deep learning is to enable machines to have the ability to analyze and learn like humans, and to recognize data such as words, images, and sounds.

Computer vision refers to the use of cameras and computers instead of human eyes to identify, track and measure targets, and further perform graphics processing to make computer processing images more suitable for human eyes to observe or transmit to instruments for detection.

In the technical solutions of the present disclosure, the collection, storage, use, processing, transmission, provision, and disclosure of the user's personal information involved are all in compliance with relevant laws and regulations, and do not violate public order and good customs.

As shown in Figure 1, the generation method of the business data processing model includes:

S101: Determine model evaluation information of an initial business data processing model.

The model may be, for example, an artificial intelligence model, such as a neural network model, a machine learning model, or the like.

The business data processing model may refer to a model used to process various business data.

The initial business data processing model may be the business data processing model obtained in the initial stage of training the artificial intelligence model, or, when multiple rounds of iterative training are performed on the artificial intelligence model, the previous The business data processing model obtained by the rounds of iterative training is used as the initial business data processing model, or the initial business data processing model may also be a preconfigured initial business data processing model.

In the embodiment of the present disclosure, the model performance of the initial business data processing model can be analyzed and evaluated to obtain model evaluation information, and then, based on the model evaluation information, the selection of operators and the setting of hyperparameters in the next round of iterative training process are guided. This does not limit.

In this embodiment of the present disclosure, the initial business data processing model may be constructed by multiple operators, and obtained by training the business data processing models constructed by the multiple operators. When constructing the business data processing model based on multiple operators, it is also possible to determine The description information of each operator (the description information is, for example, hyperparameters), and then configure each operator based on the description information to obtain a business data processing model, and start the training process of the subsequent business data processing model to obtain the initial business data processing model.

In the embodiment of the present disclosure, in the process of obtaining the initial business data processing model through training, the performance of the initial business data processing model can also be dynamically analyzed to obtain its model evaluation information, which is used for operator selection in the next round of iterations. and hyperparameter settings, there are no restrictions on this.

In this embodiment of the present disclosure, an operator search space may be pre-configured, and the operator search space may include multiple types of operators, and a parameter search space may be set, and the parameter search space may include operators corresponding to various types of operators. , in each iteration process, the target operator required for the current iteration can be selected from the operator search space, and the hyperparameters required for the current iteration can be selected from the parameter search space (which can be used as the target). description), there is no restriction on this.

In the embodiment of the present disclosure, the number of the above-mentioned initial business data processing models may be one or more, and the business data that the business data model can process may be, for example, financial risk control data, video data, image data, text data, audio data, etc. , there is no restriction on this.

Wherein, the model evaluation information may refer to evaluating the attribute features of one or more dimensions of the above-mentioned business data processing model, and the obtained relevant information, for example, may be the work efficiency evaluation information, stability and stability of the initial business data processing model. Evaluation information, output accuracy evaluation information, etc., are not limited.

In the embodiment of the present disclosure, when determining the model evaluation information of the initial business data processing model, the communication link between the executive body and the big data server in the embodiment of the present disclosure may be established in advance, and then the initial business data processing model is obtained from the big data server. Corresponding model evaluation information, alternatively, it is possible to obtain multiple reference business data in advance, and then use the initial business data processing model to process the multiple reference business data, and then analyze and obtain the corresponding model evaluation information according to the processing result. No restrictions.

S102: Determine a target operator according to the model evaluation information.

The operator refers to a machine learning algorithm, which may refer to the process of running on business data to create a business data processing model. The operator may be, for example, a K-proximity algorithm, a linear regression algorithm, etc., which is not limited.

The target operator can refer to the operator selected based on the model evaluation information. Since the target operator can be selected with reference to the model evaluation information of the initial business data processing model, the obtained target operator can be applied to the target to be constructed in the current round. Business data processing model.

In some embodiments, the target operator is determined according to the model evaluation information, which may be a pre-configured evaluation threshold, and then a candidate operator is randomly selected, and the candidate operator evaluation information corresponding to the candidate operator is determined based on the model evaluation information, and the candidate operator is determined. The evaluation information is analyzed and compared with the evaluation threshold. When the evaluation information of the candidate operator is greater than or equal to the evaluation threshold, the corresponding candidate operator is used as the target operator, which is not limited.

In other embodiments, the target operator is determined according to the model evaluation information, and the operator characteristics of each optional operator may also be obtained, and then the matching degree between the model evaluation information and the characteristics of each operator is analyzed, according to the matching degree of each operator. The degree determines the target operator from multiple optional operators, which is not limited.

Of course, any other possible methods can also be used to realize the determination of the target operator according to the model evaluation information, such as engineering methods, combination of numbers and shapes, etc., which are not limited.

In the embodiment of the present disclosure, since the number of operators that can be selected in the process of generating the target service data processing model may be multiple, and the performance information corresponding to different operators may be different, when the target operator is determined according to the model evaluation information It can guarantee the performance of the target operator and provide reliable reference information for the subsequent generation process of the target business data processing model.

S103: Determine target description information of the target operator according to the model evaluation information.

Among them, the description information can refer to the hyperparameters corresponding to the operator, that is, some parameters used to determine the model, different hyperparameters, different models (for example, different layers of convolutional neural network models are different model), the hyperparameters can be variables determined according to experience, or can be configured according to the application scenario, which is not limited.

For example, in deep learning, hyperparameters include: learning rate, number of iterations, number of layers, number of neurons in each layer, etc.

The target description information refers to the description information of the target operator, which is determined based on the model evaluation information and can be used to generate the target business data processing model. The target description information is, for example, the specified learning rate and the number of iterations. , model layers, etc.

In some embodiments, determining the target description information of the target operator according to the model evaluation information may be determining the initial description information of the target operator, generating a corresponding initial vector based on the initial description information, and assigning a value to the initial vector and changing the initial value. The value of the vector, based on the model evaluation information, evaluate the performance of the target operator corresponding to the initial vector of various values, and form the corresponding evaluation value, and select the value of the initial vector corresponding to the maximum evaluation value as the final optimization of the initial vector value, and then optimize the initial description information based on the final optimized value to obtain the target description information.

In other embodiments, the target description information of the target operator is determined according to the model evaluation information, or any other possible method may be used to determine the target description information of the target operator according to the model evaluation information, such as determining the corresponding information of the target operator. The initial description information is then optimized by the Bayesian optimization method to generate the target description information, which is not limited.

It can be understood that the same operator can use different description information, and the performance of models generated by combining different description information with operators may be different. Determining the target description information of the target operator based on the model evaluation information can guarantee the obtained target. The description effect of the description information can improve the reliability of the target description information in the process of generating the target business data processing model.

In this embodiment of the present disclosure, after the model evaluation information of the initial business data processing model is determined, the adapted target operator and the target description information can be respectively determined based on the model evaluation information, so as to realize the effective process of determining the target operator and the target description information. Decoupling, while ensuring the contribution of target operators and target description information to model performance, effectively improves the flexibility of the target business data processing model generation process, and improves the generation efficiency of the target business data processing model.

In this embodiment of the present disclosure, the number of target operators may also be multiple. Correspondingly, each target operator has corresponding target description information. In the process of determining the target operator according to the model evaluation information, the target operator may also be determined based on the model evaluation information. The model evaluation information simultaneously determines multiple target operators and target description information corresponding to each target operator, thereby greatly improving the search and selection efficiency of target operators and target description information.

S104: Generate a target business data processing model according to the target description information and the target operator.

The target business data processing model may refer to a business data processing model generated based on target description information and target operators.

That is to say, after determining the target description information and the target operator according to the model evaluation information, the embodiment of the present disclosure can perform description processing on the target operator based on the target description information, and generate target service data processing based on the described target operator. Model.

In some embodiments, the number of target operators may be multiple, and the target business data processing model is generated according to the target description information and the target operators, which may be to determine the priority information of multiple target operators, and then according to the priority information. The corresponding target operators are described and processed by using the target description information respectively, so as to obtain a plurality of post-description target operators, and a target business data processing model is generated based on the plurality of post-description target operators.

In other embodiments, the target business data processing model is generated according to the target description information and the target operator, and one or more target operators may be randomly selected from multiple target operators as the target operator to be described, and according to The target description information corresponding to the target operator to be described performs description processing on the target operator to be described to obtain a plurality of post-description target operators, and generates a target business data processing model based on the plurality of post-description target operators, which is not limited. .

In this embodiment, the model evaluation information of the initial business data processing model is determined, the target operator is determined according to the model evaluation information, the target description information of the target operator is determined according to the model evaluation information, and the target description information and the target operator are determined according to the target description information and the target operator. , to generate the target business data processing model, thus, the target operator and target description information can be determined respectively based on the model evaluation information, so as to realize the decoupling of the determination process of the target operator and the target description information, and effectively improve the generation of target business data processing model. The flexibility of the process improves the generation efficiency of the target business data processing model.

FIG. 2 is a schematic diagram of a second embodiment according to the present disclosure.

As shown in Figure 2, the generation method of the business data processing model includes:

S201: Determine model evaluation information of an initial business data processing model.

The description of S201 may refer to the above-mentioned embodiments for details, and details are not repeated here.

S202: Determine multiple candidate operators.

The candidate operator refers to multiple operators that may be used as target operators.

In the embodiment of the present disclosure, to determine multiple candidate operators, it may be to predetermine the feature information of the application scenario and the above-mentioned initial business data processing model, and to determine, based on the feature information, multiple operators that may be applicable to the business data processing model as The candidate operator, thus, can provide a reliable analysis object for the subsequent determination of the target operator.

S203: Determine candidate operator evaluation information corresponding to each candidate operator according to the model evaluation information.

The candidate operator evaluation information refers to the evaluation information of the candidate operator determined based on the model evaluation information. The candidate operator evaluation information can be used to evaluate the performance of the candidate operator to determine different candidate operators. performance differences. The candidate operator evaluation information may include the iteration participation times corresponding to the candidate operator, operator performance information, etc., which is not limited.

In some embodiments, determining the candidate operator evaluation information corresponding to each candidate operator according to the model evaluation information may be to determine the time resources and system memory resources consumed by each candidate operator in the execution process based on the model evaluation information, and The time resource and system memory resource are used as the corresponding candidate operator evaluation information.

In other embodiments, the candidate operator evaluation information corresponding to each candidate operator is determined according to the model evaluation information, and any other possible information may be determined based on the model evaluation information as the candidate operator evaluation information, for example, each candidate operator There are no restrictions on the robustness of abnormal input and the legibility of each candidate operator.

In the embodiment of the present disclosure, by determining the candidate operator evaluation information corresponding to each candidate operator according to the model evaluation information, the obtained candidate operator evaluation information can effectively represent the evaluation information corresponding to one or more dimensions of the candidate operator, which can be Subsequent determination of the target operator from multiple candidate operators provides a clear and accurate reference.

S204: Determine target operator evaluation information from multiple candidate operator evaluation information, and use the candidate operator corresponding to the target operator evaluation information as the target operator.

The target operator evaluation information may refer to one or more candidate operator evaluation information selected from multiple candidate operator evaluation information based on preset conditions. For example, priority information of multiple candidate operator evaluation information may be determined, and then one or more candidate operator evaluation information may be selected from the multiple candidate operator evaluation information as target operator evaluation information according to the priority information.

In some embodiments, the number of evaluation information contained in the candidate operator evaluation information may be multiple, and the multiple evaluation information corresponds to multiple dimension features of the candidate operator, and the target operator is determined from the multiple candidate operator evaluation information. Evaluation information, which can be weighted processing of multiple evaluation information in the evaluation information of a single candidate operator to obtain comprehensive evaluation information of multiple dimension features, and then analyze the comprehensive evaluation information corresponding to the evaluation information of multiple candidate operators Compare, and take the candidate operator evaluation information corresponding to the optimal comprehensive evaluation information as the target operator evaluation information.

In other embodiments, the target operator evaluation information is determined from multiple candidate operator evaluation information, and dimension evaluation information corresponding to one or more dimension features adapted to the application scenario may also be selected from the candidate operator evaluation information. , and analyze and compare the dimension evaluation information corresponding to the multiple candidate operators to determine the target operator evaluation information suitable for the application scenario.

In the embodiment of the present disclosure, preferably, multiple candidate operators can be determined, and according to the model evaluation information, the candidate operator evaluation information corresponding to each candidate operator can be determined, and then the target operator can be determined from the multiple candidate operator evaluation information. The candidate operator evaluation information corresponding to the target operator evaluation information is used as the target operator. Therefore, the obtained candidate operator evaluation information can effectively represent the performance information of the corresponding candidate operator. Based on the candidate operator evaluation information, it can be accurately , Quickly determine a candidate operator with better performance from a plurality of candidate operators as a target operator to ensure the reliability of the determination process of the target operator.

S205: Determine initial description information of the target operator.

The initial description information refers to the unprocessed description information of the target operator in the initial state. The initial description information can be, for example, the hyperparameter value initially configured for the target operator, or the preset value of the hyperparameter. No restrictions.

In the embodiment of the present disclosure, to determine the initial description information of the target operator, the communication link between the executive body and the big data server in the embodiment of the present disclosure may be established in advance, and the better description information of the target operator is obtained from the big data server as The initial description information, thus, the initial description information can provide a reliable reference object for determining the target description information.

S206: Determine the information adjustment value of the initial description information according to the model evaluation information.

The information adjustment value refers to the adjustment information determined by the initial description information based on the model evaluation information, and the information adjustment value can be used as a reference for subsequent adjustment of the initial description information.

The information adjustment value is, for example, the value of the hyperparameter initially configured for the target operator, or the value adjusted by the preset value of the hyperparameter, which is not limited.

In the embodiment of the present disclosure, determining the information adjustment value of the initial description information according to the model evaluation information may be to determine the performance trend of the initial description information under different values according to the model evaluation information, and then determine the initial description information according to the performance trend. Therefore, an accurate execution basis can be provided for the subsequent adjustment process of the initial description information.

S207: Adjust the initial description information according to the information adjustment value to obtain the target description information.

That is to say, in the embodiments of the present disclosure, the initial description information can be flexibly adjusted based on the information adjustment value, thereby ensuring the adaptability of the obtained target description information and the construction process of the target service data processing model.

In the embodiment of the present disclosure, preferably, the initial description information of the target operator can be determined, and the information adjustment value of the initial description information can be determined according to the model evaluation information, and then the initial description information can be adjusted according to the information adjustment value to obtain the target description information, Since the information adjustment value is obtained based on the model evaluation information, and the model evaluation information can effectively represent the performance information of the target operator, the obtained information adjustment value can provide a clear and reliable reference for the adjustment process of the initial description information, which can effectively improve the The description effect of the obtained target description information on the target operator.

S208: Describe the corresponding target operator according to the plurality of target description information, so as to obtain a plurality of described target operators.

The described target operator refers to an operator obtained by the target operator through the description process of the corresponding target description information.

In the embodiment of the present disclosure, by describing the corresponding target operators according to the multiple target description information, the synchronous description processing of the multiple target operators can be quickly and accurately realized, thereby effectively improving the work efficiency of constructing the target business data processing model.

S209: Build a target business data processing model according to the multiple described target operators.

In the embodiment of the present disclosure, by constructing a target business data processing model according to a plurality of described target operators, the performance of the target business data processing model can be effectively improved, so that it is suitable for different application environments.

In the embodiment of the present disclosure, preferably, corresponding target operators can be described in parallel according to multiple target description information, so as to obtain multiple described target operators, and a target business data processing model can be constructed according to the multiple described target operators. It realizes the parallel processing of the description process corresponding to multiple target operators, and reduces the time cost of the overall description process to a large extent. Based on multiple described target operators, the construction effect of the target business data processing model can be effectively improved, and the target business data obtained can be guaranteed. The working performance of the processing model in different application environments.

In this embodiment, by determining the model evaluation information of the initial business data processing model, multiple candidate operators are determined, and according to the model evaluation information, the candidate operator evaluation information corresponding to each candidate operator is determined, and the multiple candidate operators are selected from the multiple candidate operators. The target operator evaluation information is determined in the evaluation information, and the candidate operator corresponding to the target operator evaluation information is used as the target operator. Therefore, the obtained candidate operator evaluation information can effectively represent the performance information of the corresponding candidate operator. The operator evaluation information can accurately and quickly determine the candidate operator with better performance from multiple candidate operators as the target operator, to ensure the reliability of the target operator determination process, and then determine the initial description information of the target operator, According to the model evaluation information, the information adjustment value of the initial description information is determined, and the initial description information is adjusted according to the information adjustment value to obtain the target description information. Since the information adjustment value is obtained based on the model evaluation information, the model evaluation information can effectively represent the target calculation information. Therefore, the obtained information adjustment value can provide a clear and reliable reference for the adjustment process of the initial description information, which can effectively improve the description effect of the obtained target description information on the target operator. target operator to obtain multiple described target operators, and build a target business data processing model according to the multiple described target operators, which can realize parallel processing of the description process corresponding to multiple target operators, and greatly reduce the overall The time cost of the description process, based on multiple described target operators, can effectively improve the construction effect of the target business data processing model, and ensure the performance of the obtained target business data processing model in different application environments.

FIG. 3 is a schematic diagram of a third embodiment according to the present disclosure.

As shown in Figure 3, the generation method of the business data processing model includes:

S301: Determine model evaluation information of an initial business data processing model.

S302: Determine multiple candidate operators.

For the description of S301-S302, reference may be made to the foregoing embodiments, and details are not repeated here.

S303: Determine the iteration participation times of each candidate operator according to the model evaluation information.

It can be understood that, before the target operator is determined in the embodiment of the present disclosure, the above-mentioned multiple candidate operators may participate in multiple rounds of model evaluation to improve the performance of each candidate operator. The model evaluation participation rounds and performance information of the candidate operators are used to determine the candidate operators participating in the model evaluation in the next round.

The number of iterative participations refers to the number of participations of candidate operators in the above-mentioned multi-round model evaluation process.

In the embodiment of the present disclosure, when a candidate operator is iterating, it will consume time cost and computing power cost, and the consumption amount of time cost and computing power cost may increase with the increase of the number of iteration participations. The model evaluation information determines the iterative participation times of each candidate operator, which can effectively represent the cost information of each candidate operator, thereby providing a reliable reference for subsequent determination of candidate operator evaluation information and ensuring the obtained candidate operator evaluation information comprehensiveness.

S304: Determine operator performance information of each candidate operator according to the model evaluation information.

The operator performance information may refer to related information used to describe the performance of each candidate operator.

For example, a set of operator performance evaluation criteria may be formulated in advance, and then a comparative analysis is performed based on the model evaluation information corresponding to each candidate operator and the operator performance evaluation criteria to determine the operator performance information of each candidate operator.

In the embodiment of the present disclosure, by determining the operator performance information of each candidate operator according to the model evaluation information, a suitable reference basis can be provided for the subsequent determination of the candidate operator evaluation information, so that the candidate operator evaluation information can effectively characterize each candidate operator child performance information.

S305: Use the iteration participation times and operator performance information as candidate operator evaluation information corresponding to the corresponding candidate operator.

That is to say, after obtaining the iteration participation times and operator performance information in the embodiments of the present disclosure, the iteration participation times and operator performance information may be used as candidate operator evaluation information corresponding to the corresponding candidate operators, so as to realize candidate operator evaluation. The information is comprehensively and accurately considered for the corresponding candidate operator, which can effectively improve the ability of the candidate operator evaluation information to represent the relevant information of the candidate operator.

For example, there are 3 candidate operators (candidate operator 1, candidate operator 2, candidate operator 3). At the current point in time, candidate operator 1 has iterated 10 times (the evaluation score is 0.1), and the candidate operator 2 has been iterated 5 times (evaluation score is 0.3), candidate operator 3 has been iterated 2 times (evaluation score is 0.4), and at this time candidate operator 1 performs better (evaluation score is 1), candidate operator 2 The performance of the candidate operator 3 is second (the evaluation score is 0.9), and the performance of the candidate operator 3 is poor (the evaluation score is 0.6), then the candidate operator 2 has the highest comprehensive evaluation score at the current time point (score=0.9+0.3= 1.2), so the candidate operator 2 can be selected as the target operator.

In the embodiment of the present disclosure, preferably, the iteration participation times of each candidate operator may be determined according to the model evaluation information, the operator performance information of each candidate operator may be determined according to the model evaluation information, and the iteration participation times and the operator performance may be calculated. The information is used as the candidate operator evaluation information corresponding to the corresponding candidate operator, which can greatly enrich the obtained candidate operator evaluation information from the two dimensions of iteration participation times and operator performance information, and realize comprehensive and accurate consideration of each candidate operator. , to improve the ability of the candidate operator evaluation information to represent the candidate operator related information.

S306: Determine target operator evaluation information from multiple candidate operator evaluation information, and use the candidate operator corresponding to the target operator evaluation information as the target operator.

The description of S306 may refer to the above-mentioned embodiment for details, and details are not repeated here.

S307: Determine multiple candidate description information of the target operator.

The candidate description information refers to a plurality of description information applicable to the target operator.

It can be understood that the target operator may correspond to multiple candidate description information, and the description effects of the multiple candidate description information on the target operator may be different.

In the embodiment of the present disclosure, by determining multiple candidate description information of the target operator, a reliable analysis object can be provided for the subsequent determination of the target description information.

S308: Determine a matching degree value between the model evaluation information and each candidate description information.

The matching degree value may refer to a value obtained by performing matching processing between each candidate description information and model evaluation information and used to represent the matching degree between each candidate description information and model evaluation information.

In the embodiment of the present disclosure, since the model evaluation information can effectively represent the best description information of the target operator in the model evaluation process, when the matching degree value between the model evaluation information and each candidate description information is determined, the matching degree value is the following value. Determining the target description information provides accurate judgment basis.

S309: Select the candidate description information corresponding to the largest matching degree value from the plurality of matching degree values, and use the corresponding candidate description information as the target description information.

That is to say, after determining the matching degree value between the model evaluation information and each candidate description information in this embodiment of the present disclosure, the candidate description information corresponding to the largest matching degree value may be selected from multiple matching degree values, and The corresponding candidate description information is used as the target description information, so as to improve the applicability of the obtained target description information in the process of generating the target business data processing model.

In the embodiment of the present disclosure, preferably, multiple candidate description information of the target operator may be determined, the matching degree value between the model evaluation information and each candidate description information may be determined, and then the largest matching degree may be selected from the multiple matching degree values. The candidate description information corresponding to the value, and the corresponding candidate description information is used as the target description information, because the obtained matching degree value can effectively represent the best description information obtained by the target operator in the model training process. When the target description information is determined based on the maximum matching degree value, the adaptability of the obtained target description information in the process of generating the target business data processing model can be effectively improved, and the description effect of the obtained target description information can be maximized.

S310: Generate a target business data processing model according to the target description information and the target operator.

For the description of S310, reference may be made to the foregoing embodiments, and details are not repeated here.

In this embodiment, multiple candidate operators are determined by determining the model evaluation information of the initial business data processing model, the iteration participation times of each candidate operator are determined according to the model evaluation information, and the number of iterations of each candidate operator is determined according to the model evaluation information. The operator performance information, and the iteration participation times and operator performance information are used as the candidate operator evaluation information corresponding to the corresponding candidate operators, which can greatly enrich the obtained candidate operators from the two dimensions of iteration participation times and operator performance information. Evaluation information, to achieve comprehensive and accurate consideration of each candidate operator, improve the ability of candidate operator evaluation information to represent candidate operator-related information, and then determine target operator evaluation information from multiple candidate operator evaluation information, and use it. The candidate operator corresponding to the target operator evaluation information is used as the target operator to determine multiple candidate description information of the target operator, determine the matching degree value between the model evaluation information and each candidate description information, and select from the multiple matching degree values. The candidate description information corresponding to the maximum matching degree value, and the corresponding candidate description information is used as the target description information, because the obtained matching degree value can effectively represent the best description information obtained by the target operator in the model training process and each candidate. The matching degree between the description information. When the target description information is determined based on the maximum matching degree value, the adaptability of the obtained target description information in the process of generating the target business data processing model can be effectively improved, and the description of the obtained target description information can be maximized. Effect.

FIG. 4 is a schematic diagram of a fourth embodiment according to the present disclosure.

As shown in Figure 4, the generating device 40 of the business data processing model includes:

The first determination module 401 is used to determine the model evaluation information of the initial business data processing model;

The second determination module 402 is configured to determine the target operator according to the model evaluation information;

The third determining module 403 is configured to determine the target description information of the target operator according to the model evaluation information; and

The generating module 404 is configured to generate a target business data processing model according to the target description information and the target operator.

In some embodiments of the present disclosure, as shown in FIG. 5 , which is a schematic diagram according to a fifth embodiment of the present disclosure, the apparatus 50 for generating a business data processing model includes: a first determination module 501 and a second determination module 502, a third determination module 503, a generation module 504, wherein the second determination module 502 includes:

The first determination submodule 5021 is used to determine a plurality of candidate operators;

The second determination submodule 5022 is configured to determine candidate operator evaluation information corresponding to each candidate operator according to the model evaluation information;

The third determination sub-module 5023 is configured to determine target operator evaluation information from multiple candidate operator evaluation information, and use the candidate operator corresponding to the target operator evaluation information as the target operator.

In some embodiments of the present disclosure, the third determining module 503 is specifically configured to:

Determine the initial description information of the target operator;

Based on the model evaluation information, determine the information adjustment value of the initial description information; and

Adjust the initial description information according to the information adjustment value to obtain the target description information.

In some embodiments of the present disclosure, the third determining module 503 is further configured to:

Determine multiple candidate description information of the target operator;

determining a value for the degree of match between the model evaluation information and each candidate description; and

The candidate description information corresponding to the largest matching degree value is selected from the plurality of matching degree values, and the corresponding candidate description information is used as the target description information.

In some embodiments of the present disclosure, the second determination sub-module 5022 is specifically configured to:

Determine the iteration participation times of each candidate operator according to the model evaluation information;

Determine the operator performance information of each candidate operator according to the model evaluation information; and

The iteration participation times and operator performance information are used as candidate operator evaluation information corresponding to the corresponding candidate operator.

In some embodiments of the present disclosure, the number of target operators is multiple;

Wherein, the generating module 504 is specifically used for:

Describe the corresponding target operator according to the plurality of target description information, so as to obtain a plurality of described target operators;

According to a plurality of described target operators, a target business data processing model is constructed.

It can be understood that, the generating apparatus 50 of the business data processing model in FIG. 5 of this embodiment and the generating apparatus 40 of the business data processing model in the above-mentioned embodiment, the first determining module 501 is the same as the first determining module 501 in the above-mentioned embodiment. module 401, the second determination module 502 and the second determination module 402 in the above embodiment, the third determination module 503 and the third determination module 403 in the above embodiment, the generation module 504 and the generation module 404 in the above embodiment, can have the same function and structure.

It should be noted that, the foregoing explanations on the method for generating a business data processing model are also applicable to the apparatus for generating a business data processing model in this embodiment.

In this embodiment, the model evaluation information of the initial business data processing model is determined, the target operator is determined according to the model evaluation information, the target description information of the target operator is determined according to the model evaluation information, and the target description information and the target operator are determined according to the target description information and the target operator. , to generate the target business data processing model, thus, the target operator and target description information can be determined respectively based on the model evaluation information, so as to realize the decoupling of the determination process of the target operator and the target description information, and effectively improve the generation of target business data processing model. The flexibility of the process improves the generation efficiency of the target business data processing model.

The method for generating a business data processing model in the embodiment of the present disclosure may be applied to the following apparatus for generating a business data processing model, and the framework structure of the apparatus for generating a business data processing model may also be as follows:

FIG. 6 is a schematic diagram of a sixth embodiment according to the present disclosure.

As shown in FIG. 6, the generating apparatus 60 of the business data processing model may further include: an operator selection module 601 and an operator description module 602; wherein,

The operator selection module 601 is used to determine the model evaluation information of the initial business data processing model, and determine the target operator according to the model evaluation information, and transmit the model evaluation information to the operator description module;

The operator description module 602 is configured to determine target description information of the target operator according to the model evaluation information, wherein the target operator and the target description information are used to generate a target service data processing model.

In some embodiments of the present disclosure, as shown in FIG. 7 , which is a schematic diagram according to a seventh embodiment of the present disclosure, the apparatus 70 for generating a business data processing model may further include: an operator selection module 701 , an operator Describe the module 702, wherein the operator selection module 701 is specifically used for:

Determine multiple candidate operators;

Determine candidate operator evaluation information corresponding to each candidate operator according to the model evaluation information; and

The target operator evaluation information is determined from the plurality of candidate operator evaluation information, and the candidate operator corresponding to the target operator evaluation information is used as the target operator.

In this embodiment of the present disclosure, the operator selection module 701 may be used to perform a dispatch supervisor for the model generation process, and thus may also be referred to as a dispatch supervisor module (Dispatcher Supervisor), and a process can be started in the dispatch supervisor module (Dispatcher Supervisor). , to receive and send messages based on the process, determine the target operator, and control a single scheduling module to perform corresponding operations according to the received command message.

In some embodiments of the present disclosure, the operator description module 702 is specifically configured to:

Determine the initial description information of the target operator;

Determine the information adjustment value of the initial description information based on the model evaluation information; and

Adjust the initial description information according to the information adjustment value to obtain the target description information.

In some embodiments of the present disclosure, the operator description module 702 is specifically configured to:

Determine multiple candidate description information of the target operator;

determining a value for the degree of match between the model evaluation information and each candidate description; and

The candidate description information corresponding to the largest matching degree value is selected from the plurality of matching degree values, and the corresponding candidate description information is used as the target description information.

In this embodiment of the present disclosure, the operator description module 702 may be used to perform scheduling optimization on the description information of a single target operator, and thus may also be referred to as a single dispatch module (Single Dispatcher), a single dispatch module (Single Dispatcher) You can also start a process, and based on the process, initialize the target operator, tune the description information of the target operator, and collect the evaluation indicators of the target operator.

It can be understood that the above-mentioned operator selection module 701 and operator description module 702 can be encapsulated in a tuner module (Tuner) at the same time, so as to improve the degree of integration of the generating device of the business data processing model, and at the same time, it can improve the operator selection. The interaction efficiency between the module and the operator description module ensures the working performance of the operator selection module and the operator description module.

In the corresponding process of the Tuner module, the scheduling supervisor sub-module is the main thread, and the single-scheduling sub-module is the sub-thread. A shared queue can be used to communicate between the scheduler sub-module and the single-scheduler sub-module.

In some embodiments of the present disclosure, the operator selection module 701 is specifically configured to:

Determine the iteration participation times of each candidate operator according to the model evaluation information;

Determine the operator performance information of each candidate operator according to the model evaluation information; and

The iteration participation times and operator performance information are used as candidate operator evaluation information corresponding to the corresponding candidate operator.

In some embodiments of the present disclosure, wherein the apparatus further comprises:

The model generation module 703 is used to obtain the target operator identifier corresponding to the target operator, and generate a model construction message according to the target operator identifier, and transmit the model construction message to the operator description module 702;

In this embodiment of the present disclosure, the model generation module 703 may be used to drive multiple processes of the business data processing model generation process, and thus may also be referred to as a driver module (Driver), and the driver module (Driver) can also be started process, and based on the process perform receiving messages, sending messages, etc.

It can be understood that, in the device for generating the business data processing model, the above-mentioned Tuner module and the Driver module pass through interprocess communication (interprocess communication, IPC), such as pipe communication, and the communication content can be composed of command and It consists of data, the transmission format is (command, data), command represents the type of command, and data represents the data object corresponding to the command operation.

Among them, the command type can include but is not limited to:

#in

Initialize=b'IN', indicating that the search space is initialized.

RequestTrialJobs=b'GE', indicating that the experiment is requested.

ReportMetricData=b'ME', indicating reporting metrics.

UpdateSearchSpace=b'SS', which means to update the search space.

ImportData=b'FD', which means import data, which is used for breakpoint recovery.

TrialEnd=b'EN', indicating the end of a single experiment.

Terminate=b'TE', which means to terminate the task.

#out

Initialized=b'ID', indicating that the search space has been initialized.

NewTrialJob=b'TR', indicating that the experimental configuration has been generated.

NoMoreTrialJobs=b'NO', indicating that there is no more configuration information in the search space.

KillTrialJob=b'KI', indicating termination of the experiment.

Among them, the data format can include but is not limited to:

#in

Initialize data: Used to initialize the search space for various machine learning algorithms.

The hyperparameter search space for multiple algorithms can be, for example,

Among them, id represents different operators (types), and parameters represents the description information (type and value range) corresponding to the operators.

RequestTrialJobs data: The number of Trials (experiments) requested.

ReportMetricData data: Trial metrics reported.

UpdateSearchSpace data: The updated search space.

ImportData data: Import the previous data list for experimental recovery.

TrialEnd data: Trial status reported.

Terminate data: Empty.

#out

Initialized data: Empty.

NewTrialJob data: Trial configuration parameters.

NoMoreTrialJobs data: Empty.

KillTrialJob data: Experiment id.

The operator description module 702 is further configured to receive the model construction message, obtain the target operator identifier by parsing from the model construction message, and describe the target operator according to the target description information of the target operator to which the target operator identifier belongs, and obtain the Describe the target operator, wherein the described target operator is used to construct the target business data processing model.

In some embodiments of the present disclosure, wherein,

The number of operator description modules 702 is multiple, and each operator description module 702 is used to determine the target description information of the corresponding target operator; wherein, the multiple operator description modules 702 are used to describe the corresponding target operator, and obtain a plurality of all the target operators. Describe the target operator, and multiple described target operators are used together to construct the target business data processing model.

In the embodiment of the present disclosure, in the process of constructing the target business data processing model, the task processes of the multiple operator description modules 702 are independent of each other, that is, the description information tuning process of the multiple target operators can be implemented in parallel, so as to improve the multiple goals Optimization efficiency of operator description information.

For example, FIG. 8 is a block diagram of an apparatus for generating an exemplary business data processing model in an embodiment of the present disclosure. As shown in FIG. 8 , the apparatus for generating an exemplary business data processing model includes: a driver module (Driver) and a Tuner module (Tuner), Driver module and Tuner module can generate corresponding processes respectively. The Tuner module may be divided into a dispatch supervisor sub-module (Dispatcher Supervisor) and a plurality of single dispatch sub-modules (Single Dispatcher). The number of the single dispatch sub-module may be one or more, which is not limited.

Based on the generation device and communication protocol of the exemplary business data processing model, the experimental process of determining the target operator and the target description information corresponding to the target operator can be as follows:

①The Driver module receives and forwards the initialize request message (Initialize); after the Dispatcher Supervisor of the Tuner module receives the Initialize request, multiple Single Dispatchers can be initialized (each Single Dispatcher corresponds to an operator). After all Single Dispatchers are initialized, the Dispatcher Supervisor can send a message (Initialized) to the Driver module indicating that the initialization has been completed; when the Driver module receives Initialized, the entire initialization task ends.

②The Driver module sends a message requesting the experiment (RequestTrialJobs); after the Dispatcher Supervisor of the Tuner module receives the RequestTrialJobs request, for each experiment request, it can be obtained according to the corresponding multi-operator selection strategy (flexibly supporting various strategies) to obtain more information in the next experiment. A target operator that may generate better indicators, thereby generating the next experiment configuration (target description information) in the corresponding Single Dispatcher, and making the Dispatcher Supervisor send a new experiment request message (NewTrialJob) to the Driver module. The NewTrialJob information contains The target operator and its target description information selected by the Dispatcher Supervisor, when the Driver module receives the complete NewTrialJob information, it indicates that the experimental task is requested to be completed.

③The Driver module receives and forwards the message (ReportMetricData) requesting the report metric data; when the Dispatcher Supervisor of the Tuner module receives the ReportMetricData request, according to the data information of the ReportMetricData (including the operator type and operator evaluation information corresponding to the experiment), the corresponding Single Dispatcher The operator evaluation information of the experiment can be collected (the hyperparameter tuning strategy will generate corresponding operator description information for the next experiment request based on the collected operator evaluation information).

④ After the stop conditions are met (for example: the number of experiments reaches the preset number, or the time consumed has exceeded the time budget, etc.), the Driver module receives and forwards the request termination message (Terminate); after the Dispatcher Supervisor of the Tuner module receives the Terminate request, The Terminate signal can be sent to multiple SingleDispatchers in the Tuner module. When all the Single Dispatcher threads stop running and the Dispatcher Supervisor thread also stops running, it means that the Tuner module process stops and completes (that is, the target operator selection and the target operator corresponding to the target description information optimization task Finish).

It can be understood that, the generating apparatus 70 of the business data processing model in FIG. 7 of this embodiment and the generating apparatus 60 of the business data processing model in the above embodiment, the operator selection module 701 and the operator selection in the above embodiment The module 601 and the operator description module 702 may have the same function and structure as the operator description module 602 in the above embodiment.

It should be noted that the foregoing explanations on the method for generating a business data processing model are also applicable to the apparatus for generating a business data processing model in this embodiment.

In this embodiment, the model evaluation information of the initial business data processing model is determined, the target operator is determined according to the model evaluation information, the target description information of the target operator is determined according to the model evaluation information, and the target description information and the target operator are determined according to the target description information and the target operator. , to generate the target business data processing model, thus, the target operator and target description information can be determined respectively based on the model evaluation information, so as to realize the decoupling of the determination process of the target operator and the target description information, and effectively improve the generation of target business data processing model. The flexibility of the process improves the generation efficiency of the target business data processing model.

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. 9 shows a schematic block diagram of an example electronic device that can be used to implement the business data processing model generation method 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, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are by way of example only, and are not intended to limit implementations of the disclosure described and/or claimed herein.

As shown in FIG. 9 , the device 900 includes a computing unit 901 that can be executed according to a computer program stored in a read only memory (ROM) 902 or a computer program loaded from a storage unit 908 into a random access memory (RAM) 903 Various appropriate actions and handling. In the RAM 903, various programs and data necessary for the operation of the device 900 can also be stored. The computing unit 901 , the ROM 902 , and the RAM 903 are connected to each other through a bus 904 . An input/output (I/O) interface 905 is also connected to bus 904 .

Various components in the device 900 are connected to the I/O interface 905, including: an input unit 906, such as a keyboard, mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a magnetic disk, an optical disk, etc. ; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 909 allows the device 900 to exchange information/data with other devices through a computer network such as the Internet and/or various telecommunication networks.

Computing unit 901 may be various general-purpose and/or special-purpose processing components with processing and computing capabilities. Some examples of computing units 901 include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), various specialized artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processing processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 executes the various methods and processes described above, for example, executes the generation method of the business data processing model. For example, in some embodiments, a method of generating a business data processing model may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 908 . In some embodiments, part or all of the computer program may be loaded and/or installed on device 900 via ROM 902 and/or communication unit 909 . When the computer program is loaded into the RAM 903 and executed by the computing unit 901, one or more steps of the above-described method for performing the generation of a business data processing model may be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to perform the generation method of the business data processing model by any other suitable means (eg, by means of firmware).

Various implementations of the systems and techniques described herein above may be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips system (SOC), load programmable logic device (CPLD), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include being implemented in one or more computer programs executable and/or interpretable on a programmable system including at least one programmable processor that The processor, which may be a special purpose or general-purpose programmable processor, may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device an output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes 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, performs the functions/functions specified in the flowcharts and/or block diagrams. Action is implemented. The program code may execute entirely on the machine, partly on the machine, partly on the machine and partly on a remote machine as a stand-alone software package or entirely on the remote machine or server.

In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in connection with the instruction execution system, apparatus or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media may include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include one or more wire-based electrical connections, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), fiber optics, compact disk read only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination of the foregoing.

To provide interaction with a user, the systems and techniques described herein may be implemented on a computer having a display device (eg, a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user ); and a keyboard and pointing device (eg, a mouse or trackball) through which a user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (eg, visual feedback, auditory feedback, or tactile feedback); and can be in any form (including acoustic input, voice input, or tactile input) to receive input from the user.

The systems and techniques described herein may be implemented on a computing system that includes back-end components (eg, as a data server), or a computing system that includes middleware components (eg, an application server), or a computing system that includes front-end components (eg, a user's computer having a graphical user interface or web browser through which a user may interact with implementations of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system. The components of the system may be interconnected by any form or medium of digital data communication (eg, a communication network). Examples of communication networks include: Local Area Networks (LANs), Wide Area Networks (WANs), the Internet, and blockchain networks.

A computer system can include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as a cloud computing server or a cloud host. It is a host product in the cloud computing service system to solve the traditional physical host and VPS service ("Virtual Private Server", or "VPS" for short). , there are the defects of difficult management and weak business expansion. The server can also be a server of a distributed system, or a server combined with a blockchain.

It should be understood that steps may be reordered, added or deleted using the various forms of flow shown above. For example, the steps described in the present disclosure can be executed in parallel, sequentially, or in different orders. As long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, there is no limitation herein.

The above-mentioned specific embodiments do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present disclosure should be included within the protection scope of the present disclosure.

Claims (15)

1. A method for generating a business data processing model, comprising: Determine model evaluation information for the initial business data processing model; Determine the target operator according to the model evaluation information; determining target description information of the target operator according to the model evaluation information; and According to the target description information and the target operator, a target business data processing model is generated. 2. The method according to claim 1, wherein the determining a target operator according to the model evaluation information comprises: Determine multiple candidate operators; According to the model evaluation information, determine candidate operator evaluation information corresponding to each of the candidate operators; and The target operator evaluation information is determined from the plurality of candidate operator evaluation information, and the candidate operator corresponding to the target operator evaluation information is used as the target operator. 3. The method according to claim 1, wherein the determining the target description information of the target operator according to the model evaluation information comprises: determining the initial description information of the target operator; determining an information adjustment value of the initial description information according to the model evaluation information; and The initial description information is adjusted according to the information adjustment value to obtain the target description information. 4. The method according to claim 1, wherein the determining the target description information of the target operator according to the model evaluation information comprises: determining multiple candidate descriptions of the target operator; determining a matching degree value between the model evaluation information and each of the candidate description information; and The candidate description information corresponding to the largest matching degree value is selected from a plurality of the matching degree values, and the corresponding candidate description information is used as the target description information. 5. The method according to claim 2, wherein the determining the candidate operator evaluation information corresponding to each of the candidate operators according to the model evaluation information comprises: According to the model evaluation information, determine the iteration participation times of each of the candidate operators; determining operator performance information of each of the candidate operators according to the model evaluation information; and The iteration participation times and the operator performance information are used as candidate operator evaluation information corresponding to the corresponding candidate operator. 6. The method according to any one of claims 1-5, the number of the target operator is multiple; Wherein, generating the target business data processing model according to the target description information and the target operator includes: Describe corresponding target operators according to a plurality of the target description information, so as to obtain a plurality of described target operators; The target business data processing model is constructed according to the plurality of described target operators. 7. A device for generating a business data processing model, comprising: a first determination module, used for determining model evaluation information of the initial business data processing model; a second determination module, configured to determine a target operator according to the model evaluation information; a third determining module, configured to determine target description information of the target operator according to the model evaluation information; and A generating module is configured to generate a target business data processing model according to the target description information and the target operator. 8. The apparatus according to claim 7, wherein the second determining module comprises: a first determination submodule for determining multiple candidate operators; a second determination submodule, configured to determine candidate operator evaluation information corresponding to each of the candidate operators according to the model evaluation information; and The third determination submodule is configured to determine target operator evaluation information from a plurality of candidate operator evaluation information, and use the candidate operator corresponding to the target operator evaluation information as the target operator. 9. The apparatus according to claim 7, wherein the third determining module is specifically configured to: determining the initial description information of the target operator; determining an information adjustment value of the initial description information according to the model evaluation information; and The initial description information is adjusted according to the information adjustment value to obtain the target description information. 10. The apparatus according to claim 7, wherein the third determining module is further configured to: determining multiple candidate descriptions of the target operator; determining a matching degree value between the model evaluation information and each of the candidate description information; and The candidate description information corresponding to the largest matching degree value is selected from a plurality of the matching degree values, and the corresponding candidate description information is used as the target description information. 11. The apparatus according to claim 8, wherein the second determination sub-module is specifically configured to: According to the model evaluation information, determine the iteration participation times of each of the candidate operators; determining operator performance information of each of the candidate operators according to the model evaluation information; and The iteration participation times and the operator performance information are used as candidate operator evaluation information corresponding to the corresponding candidate operator. 12. The device according to any one of claims 7-11, wherein the number of the target operators is multiple; Wherein, the generating module is specifically used for: Describe corresponding target operators according to a plurality of the target description information, so as to obtain a plurality of described target operators; The target business data processing model is constructed according to the plurality of described target operators. 13. 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, the instructions being executed by the at least one processor to enable the at least one processor to perform the execution of any of claims 1-6 Methods. 14. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any of claims 1-6. 15. A computer program product comprising a computer program which, when executed by a processor, implements the steps of the method according to any of claims 1-6.

Images