Processing method and device for zero-code modeling and related equipment

Abstract

This application belongs to the field of model building technology and provides a method, apparatus, computer device, and computer-readable storage medium for zero-code modeling. To solve the problem of low efficiency in building multiple models based on zero-code modeling, this application obtains at least two starting components and their corresponding slave components selected by the user, connects the starting nodes and slave nodes using a directed acyclic graph to obtain the modeling link corresponding to each starting component, receives the deployment request submitted by the user to deploy the modeling link, and verifies whether the modeling link is complete. If the modeling link is complete, the modeling link is generated into a scheduling task model. Compared with traditional technologies, where a zero-code modeling platform can only deploy one model at a time, this application allows users to flexibly configure multiple modeling links to deploy multiple models at once, thereby meeting diverse modeling scenarios and improving the efficiency of building multiple models.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of model construction, and particularly relates to a zero-code modeling processing method and device, computer equipment and a computer readable storage medium. BACKGROUND

[0002] No-code (English for No-Code) is also called codeless in the industry. A zero-code development platform mainly builds an application scenario model through the standardization and visualization of a process, so as to realize application building without code development. Most zero-code modeling platforms are implemented based on an AutoML (English full name is Automated Machine Learning) framework, and the AutoML framework includes MLBox, AutoKeras or TPOT and the like. When zero-code modeling is implemented based on AutoML, the zero-code modeling platform can only deploy one model each time, and for a complex business scenario, flexible modeling of multiple models cannot be realized, thereby reducing the modeling efficiency of zero-code modeling. SUMMARY

[0003] The present application provides a zero-code modeling processing method and device, computer equipment and a computer readable storage medium, which can solve the technical problem of low efficiency of building multiple models based on zero-code modeling in the prior art.

[0004] In a first aspect, the present application provides a zero-code modeling processing method, including: based on a modeling canvas, acquiring at least two starting components selected by a user, and determining a starting node corresponding to each starting component according to a position of each starting component on the modeling canvas, wherein the starting component is a first component of a modeling link; acquiring a slave component corresponding to each starting component selected by the user, and determining a slave node corresponding to the slave component, wherein the slave component is a subsequent component of the starting component; connecting the starting node and the slave node by using a directed acyclic graph to obtain a modeling link corresponding to each starting component; receiving a deployment request of deploying the modeling link submitted by the user, and verifying whether the modeling link is complete according to the deployment request; and if the modeling link is complete, generating a scheduling task model of the modeling link.

[0005] In a second aspect, the present application also provides a processing device for zero-code modeling, comprising: a first obtaining unit, configured to obtain at least two starting components selected by a user based on a modeling canvas, and determine starting nodes corresponding to the starting components according to positions at which the user places each of the starting components on the modeling canvas, wherein the starting components are first components of modeling links; a second obtaining unit, configured to obtain slave components corresponding to each of the starting components selected by the user, and determine slave nodes corresponding to the slave components, wherein the slave components are subsequent components of the starting components; a connecting unit, configured to connect the starting nodes and the slave nodes using a directed acyclic graph to obtain a modeling link corresponding to each of the starting components; a checking unit, configured to receive a deployment request for deploying the modeling link submitted by the user, and check whether the modeling link is complete according to the deployment request; and a generating unit, configured to generate a scheduling task model of the modeling link if the modeling link is complete.

[0006] In a third aspect, the present application also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements steps of the processing method for zero-code modeling when executing the computer program.

[0007] In a fourth aspect, the present application also provides a computer readable storage medium, which stores a computer program, and the processor executes steps of the processing method for zero-code modeling when executing the computer program.

[0008] The present application provides a processing method, device, computer device and computer readable storage medium for zero-code modeling. The processing method obtains at least two starting components selected by a user based on a modeling canvas, and determines starting nodes corresponding to the starting components according to positions at which the user places each of the starting components on the modeling canvas. Then, the processing method obtains slave components corresponding to each of the starting components selected by the user, and determines slave nodes corresponding to the slave components. The processing method connects the starting nodes and the slave nodes using a directed acyclic graph to obtain a modeling link corresponding to each of the starting components. The processing method receives a deployment request for deploying the modeling link submitted by the user, and checks whether the modeling link is complete. If the modeling link is complete, the processing method generates a scheduling task model of the modeling link. If the modeling link is not complete, the processing method does not generate a scheduling task model of the modeling link. Compared with the traditional technology in which a zero-code modeling platform can only deploy one model each time, the embodiments of the present application can enable a user to flexibly configure multiple modeling links and implement deployment of multiple models at one time, thereby meeting diversified modeling scenarios and improving the construction efficiency of multiple models. BRIEF DESCRIPTION OF DRAWINGS

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0010] Figure 1 A flowchart of a processing method of zero-code modeling provided by an embodiment of the present application is shown in FIG. 1.

[0011] Figure 2 A first sub-flowchart of the processing method of zero-code modeling provided by an embodiment of the present application is shown in FIG. 2.

[0012] Figure 3 A second sub-flowchart of the processing method of zero-code modeling provided by an embodiment of the present application is shown in FIG. 3.

[0013] Figure 4 A third sub-flowchart of the processing method of zero-code modeling provided by an embodiment of the present application is shown in FIG. 4.

[0014] Figure 5 A schematic block diagram of a processing device of zero-code modeling provided by an embodiment of the present application is shown in FIG. 5.

[0015] Figure 6 A schematic block diagram of a computer device provided by an embodiment of the present application is shown in FIG. 6. DETAILED DESCRIPTION

[0016] The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, but not all embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0017] It should be understood that when used in the specification and the appended claims, the terms "comprise" and "include" indicate the presence of described features, integers, steps, operations, elements, and / or components, but do not exclude one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0018] The embodiments of the present application provide a processing method of zero-code modeling, which can be applied to computer devices such as tablet computers, notebook computers, desktop computers, etc. Figure 1 , Figure 1 A flowchart of a processing method of zero-code modeling provided by an embodiment of the present application is shown in FIG. 1. Figure 1 As shown in the figure, the method comprises the following steps S11-S16:

[0019] S11, based on the modeling canvas, obtain at least two starting components selected by a user, and determine a starting node corresponding to each of the starting components according to a position of each of the starting components on the modeling canvas by the user, wherein the starting component is a first component of the modeling link.

[0020] S12, obtain a slave component corresponding to each of the starting components selected by the user, and determine a slave node corresponding to the slave component, wherein the slave component is a subsequent component of the starting component.

[0021] S13, connect the starting node and the slave node by using a directed acyclic graph to obtain a modeling link corresponding to each of the starting components.

[0022] Specifically, zero-code modeling generally uses a modeling canvas, and a user performs modeling by dragging and dropping components on the modeling canvas based on the visual modeling canvas. Embodiments of the present application are also based on the modeling canvas, and the user can select components from a preset component set and place the components on the modeling canvas by dragging and dropping the components. In order to implement the construction of multiple models at a time, the user selects at least two components as starting components, each of the starting components corresponds to a modeling link, and each of the modeling links corresponds to a constructed model. The user selects the starting components, wherein the starting component is the first component of the modeling link, that is, the first component of the modeling link. The user places the starting component at a certain position on the modeling canvas by dragging and dropping the component. The computer device obtains the starting components selected by the user, and determines a starting node corresponding to each of the starting components according to a position of each of the starting components on the modeling canvas by the user dragging and dropping the starting component, to obtain a starting node of the modeling link, wherein the starting node is the first node of the modeling link, that is, the first node of the modeling link. For each of the starting nodes, a slave component corresponding to each of the starting components is obtained according to the slave component selected by the user, the slave component is a subsequent component of the starting component, that is, the slave component is a component following the starting component. A node corresponding to the slave component is determined as a slave node, the slave node is a subsequent node of the starting node, that is, the slave node is a node following the starting node. The starting node and the slave node are connected by using a directed acyclic graph to obtain a modeling link corresponding to each of the starting components, the modeling link is a DAG graph, wherein DAG is an English abbreviation of Directed Acyclic Graph, and the DAG graph is a directed acyclic graph. The starting component and the slave component are respectively selected by the user from a preset component set according to specific modeling needs.

[0023] For example, the user selects a first start component on the modeling canvas by dragging and dropping, and places the first start component at a position on the modeling canvas (the node corresponding to the position is the first start node, which is the start node of a modeling link). The computer device obtains the first start component selected by the user, determines the node corresponding to the first start component according to the motion trail of the first start component dragged and dropped by the user, and takes the node where the first start component is located as the first start node. The user then selects a plurality of slave components in sequence, and places the slave components on the modeling canvas. The positions of the slave components on the modeling canvas are slave nodes. The computer device also obtains the slave components and the slave nodes corresponding to the slave components in sequence, takes the slave components as first slave components, takes the slave nodes as first slave nodes, and connects the first start node and the first slave nodes by using a directed acyclic graph, that is, connects the first start component and the first slave components by using a directed acyclic graph, to obtain a first modeling link. The above process is repeated to obtain at least one other start component selected by the user, take the node of each other start component on the modeling canvas as an other start node, obtain other slave components corresponding to the other start components and other slave nodes corresponding to the other slave components, and connect the other start nodes and the other slave nodes by using a directed acyclic graph to obtain other modeling links, thereby obtaining a modeling link corresponding to each start component. Since the user selects at least two start components, at least two modeling links can be obtained. Compared with the conventional technology in which only one model can be deployed each time, the embodiments of the present application can enable the user to flexibly configure multiple modeling links, thereby meeting diversified modeling scenarios and improving the construction efficiency of multiple models.

[0024] Further, the start components and the slave components can be obtained from different data sources. The start components and the slave components can include custom scripts, that is, the user can obtain different components from different data sources, and the custom scripts can be uploaded in the form of a file to serve as the components selected by the user. Therefore, the data sources of the components can be diversified, and the flexibility of modeling and the efficiency of modeling can be improved by using the custom scripts as the components.

[0025] S14, receiving a deployment request for deploying the modeling link submitted by the user, and verifying whether the modeling link is complete according to the deployment request;

[0026] S15, generating a scheduling task model for the modeling link if the modeling link is complete;

[0027] S16, if the modeling link is not complete, do not generate a scheduling task model for the modeling link.

[0028] Specifically, after the user constructs the modeling link on the modeling canvas, the user submits a deployment request for deploying the modeling link. After the computer device receives the deployment request, the computer device initiates automatic checking of the completeness of the modeling link according to the deployment request. The completeness of the modeling link can be checked based on the principle of a DAG graph. The modeling link is checked by using a DAG graph. The completeness of the modeling link is mainly checked based on the running logic of the DAG graph. For example, the DAG graph requires that no node in the modeling link points to a node in front of it, and there cannot be a node in the DAG graph that is isolated and has no connection with other nodes. The modeling link is complete, and the subsequent modeling process is performed, that is, the parameters of the components of each node included in the modeling link and the running logic and other deployment information set by the user, such as the running resources and the scheduling period set by the user, are integrated, so that a scheduling task model is generated for the modeling link. The scheduling task model is a model for completing a task in a certain business by using a scheduling manner. The scheduling task model is deployed, and the deployment of the scheduling task model is completed. If the modeling link is not complete, the modeling link is not generated into a scheduling task model. The user can be prompted to modify the modeling link until the modeling link is complete. The completeness of the modeling link is checked, which can avoid the situation that the processing logic of the modeling link is missing and causes an error in the deployment and running of the model. The error correction of the modeling link is pre-processed, and the construction efficiency of the scheduling task model is improved.

[0029] In the embodiment of the application, based on the modeling canvas, at least two starting components selected by a user are obtained. Starting nodes corresponding to the starting components are determined according to positions of the starting components on the modeling canvas. A slave component corresponding to each starting component selected by the user is obtained, and a slave node corresponding to the slave component is determined. The starting nodes and the slave nodes are connected by using a directed acyclic graph to obtain a modeling link corresponding to each starting component. A deployment request for deploying the modeling link submitted by the user is received, and the completeness of the modeling link is checked. If the modeling link is complete, a scheduling task model is generated for the modeling link. If the modeling link is not complete, the modeling link is not generated into a scheduling task model. Compared with the prior art in which a zero-code modeling platform can only deploy one model each time, the embodiment of the application can enable the user to flexibly configure multiple modeling links and implement the deployment of multiple models at a time, so that diversified modeling scenarios can be met, and the construction efficiency of multiple models is improved.

[0030] Please refer to Figure 2 ,Figure 2 A first sub-flow diagram of the zero-code modeling processing method provided by the embodiment of the present application is shown in the following table. Figure 2 As shown in the table, in this embodiment, the receiving of the user's deployment request for deploying the modeling link, and the checking of whether the modeling link is complete according to the deployment request, include:

[0031] S21, obtaining the target modeling link selected by the user from all the modeling links, and receiving the deployment request corresponding to the target modeling link;

[0032] S22, checking whether the target modeling link is complete according to the deployment request.

[0033] Specifically, since multiple models can be deployed at a time, in order to further improve the flexibility of modeling, the user is allowed to select part of the modeling links as target modeling links from all the modeling links, and the target modeling links are deployed preferentially, so that the user can select the target modeling link from all the modeling links based on a selection method similar to text selection, and submit the deployment request corresponding to the target modeling link. After the computer device obtains the target modeling link and the deployment request corresponding to the target modeling link, it checks whether the target modeling link is complete according to the deployment request. Here, when checking the completeness of the target modeling link, since the target modeling link is selected by the user from all the modeling links, in order to avoid the problem of incomplete target modeling link caused by node omission in the selection process, the number of nodes contained in the target modeling link can also be checked to be consistent with the number of components of the modeling link selected by the user. If the number of nodes contained in the target modeling link is consistent with the number of components of the modeling link selected by the user, it indicates that the user has completely selected the entire target modeling link, and the target modeling link selected by the user satisfies the completeness in the number of nodes. If the number of nodes contained in the target modeling link is not consistent with the number of components of the modeling link selected by the user, it indicates that the user has selected part of the nodes of the target modeling link and omitted other nodes of the target modeling link, and the target modeling link selected by the user does not satisfy the completeness in the number of nodes, and the modeling link is incomplete.

[0034] Further, the target modeling link is multiple, and the checking of whether the target modeling link is complete includes:

[0035] checking whether each of the target modeling links is complete.

[0036] Specifically, under the premise that the user constructs multiple modeling links, the user can select multiple modeling links at a time and deploy multiple models at the same time. The multiple modeling links selected by the user are taken as target modeling links, the target modeling links are multiple, and when the user submits a deployment request, the multiple target modeling links selected by the user from all the modeling links and the deployment request corresponding to the target modeling links are obtained, and then according to the deployment request, it is checked whether each target modeling link is complete. If each target modeling link is complete, multiple models can be deployed at the same time subsequently, thereby supporting the deployment request of all complete links. If there is an incomplete target modeling link, the user needs to be reminded to modify it until the target modeling link to be deployed is complete, and then the target modeling link is deployed. Compared with the traditional technology that can only deploy one model at a time, the diversified modeling scenarios can be met, and the construction efficiency of multiple models is improved.

[0037] In an embodiment, the generating the modeling link into a scheduling task model comprises:

[0038] The calling part of each component contained in the modeling link is identified, and the calling part is labeled with a calling identifier, and the calling identifier is integrated into the scheduling task model, wherein the calling part is the running logic of each component that needs to call the corresponding script for processing.

[0039] Specifically, each component contained in the modeling link is divided into a calling part and a non-calling part. The calling part can also be called a calling running part or a calling running logic. It is the running logic of each component that needs to call the corresponding script for processing. The calling part is labeled with a calling identifier, and the calling part of each component corresponding to the modeling link is identified. The calling part is labeled with a calling identifier, and the calling identifier is integrated into the scheduling task model, so that the running logic of the calling part can be directly processed according to the calling identifier subsequently. The non-calling part can also be called a non-calling running part or a non-calling running logic. It is the running logic of each component that does not need to call the corresponding script and can be directly processed. The non-calling part can be labeled with a non-calling identifier. The non-calling part can be understood as the framework of the component running logic, which is fixed and cannot be changed. The calling part can be understood as the specific filling content in the framework, which is a flexible filling script part. By dividing each component into two parts of calling part and non-calling part, unnecessary processing can be reduced in the deployment link of the scheduling task model (i.e. the link of online deployment in the production environment), such as saving the out-parameter data, statistical information of the result data, saving the inheritance parameter file, and the like, thereby improving the running efficiency of the scheduling task model.

[0040] In an embodiment, the modeling link is used to generate a scheduling task model, which includes:

[0041] The parameters and running logic of each component contained in the modeling link are obtained, and the parameters and running logic of all the components are integrated to obtain a deployment engineering file.

[0042] Specifically, after the modeling link is obtained, if the modeling link is complete, the parameters and running logic of each component contained in the modeling link are obtained, and the parameters and running logic of all the components are integrated according to the sequence of the components contained in the modeling link to obtain a deployment engineering file, and then the scheduling task model is generated by using the deployment engineering file.

[0043] Referring to Figure 3 , Figure 3 The second sub-process diagram of the zero-code modeling processing method provided in the embodiment of the present application is shown in Figure 3 In the embodiment, the modeling link is used to generate a scheduling task model, which further includes:

[0044] S31, display a filling prompt of deployment information, and obtain information content corresponding to the deployment information filled by the user;

[0045] S32, integrate the information content and the deployment engineering file into a scheduling task model.

[0046] Specifically, the user can be prompted to fill in the deployment information, for example, a prompt interface is displayed to prompt the user to fill in the deployment information such as running resources and scheduling period, and information content corresponding to the deployment information filled by the user is obtained, and then the information content and the deployment engineering file are integrated into a scheduling task model.

[0047] Referring to Figure 4 , Figure 4 The third sub-process diagram of the zero-code modeling processing method provided in the embodiment of the present application is shown in Figure 4 In the embodiment, after the modeling link is used to generate a scheduling task model, the scheduling task model is further deployed into a task training model and a task running model.

[0048] S41, the scheduling task model is deployed into a task training model and a task running model, respectively;

[0049] S42, the task running model is used to process a task to obtain a task processing result, and actual task data corresponding to the task processing result is obtained;

[0050] S43, training the task training model with the actual task data to update the task training model.

[0051] Specifically, the scheduling task model is deployed into a task training model and a task running model respectively, that is, the generated scheduling task model is duplicated into two copies, one of which is used for model training to update the model, that is, the task training model is used for training, and the other is used for task running in the production environment, that is, the task running model is used for task running in the production environment, which can also be called a task prediction model or a task reasoning model. The task running model is used for specific business processing in actual business, and the task training model and the task running model are deployed online at the same time, and then the task running model is used for specific task processing of business to obtain a task processing result. According to the task processing result, corresponding actual task data is generated, and the task training model is trained with the actual task data to realize timely updating of the task training model. Especially when the business data changes, the task training model can be updated in time. For example, if the scheduling task model is an insurance premium rate prediction model, the insurance premium rate prediction model is duplicated into two copies, one of which is used as a task training model, and the other is used as an insurance premium rate prediction model for predicting the insurance premium rate in the insurance business. The task training model and the insurance premium rate prediction model are deployed online at the same time. Then, the insurance premium rate prediction model is used to predict the insurance premium rate of a certain insurance customer to obtain a premium rate prediction result, and the insurance product of the insurance customer is marketed according to the premium rate prediction result. According to the actual marketing effect of the insurance product of the insurance customer, corresponding insurance actual marketing business data is generated, and the task training model is trained with the insurance actual marketing business data to realize self-updating of the task training model. Especially when the business data changes, the task training model can be updated in time, and the task training model can replace the insurance premium rate prediction model, so that the insurance premium rate prediction model is more suitable for the current changes in the insurance business, improves the adaptability of the insurance premium rate prediction model and the insurance marketing business, reduces the cost of insurance marketing, and improves the utilization rate of insurance marketing resources.

[0052] It should be noted that the zero-code modeling processing method of each embodiment described above can combine the technical features contained in different embodiments as needed to obtain a combined embodiment, but all within the scope of protection claimed by the present application.

[0053] Please refer to Figure 5 , Figure 5An illustrative block diagram of a processing apparatus for zero-code modeling is provided for embodiments of the present application. Corresponding to the above-described processing method for zero-code modeling, embodiments of the present application further provide a processing apparatus for zero-code modeling. As shown in Figure 5 the processing apparatus for zero-code modeling includes units for performing the above-described processing method for zero-code modeling, and the processing apparatus for zero-code modeling can be configured in a computer device. Specifically, referring to Figure 5 the processing apparatus 50 for zero-code modeling, the processing apparatus 50 for zero-code modeling includes a first acquisition unit 51, a second acquisition unit 52, a connection unit 53, a verification unit 54, and a generation unit 55.

[0054] The first acquisition unit 51 is configured to acquire at least two starting components selected by a user based on a modeling canvas, and determine starting nodes corresponding to the starting components according to positions of each of the starting components on the modeling canvas by the user, wherein the starting components are first components of the modeling links.

[0055] The second acquisition unit 52 is configured to acquire slave components corresponding to each of the starting components selected by the user, and determine slave nodes corresponding to the slave components, wherein the slave components are subsequent components of the starting components.

[0056] The connection unit 53 is configured to connect the starting nodes and the slave nodes by a directed acyclic graph to obtain a modeling link corresponding to each of the starting components.

[0057] The verification unit 54 is configured to receive a deployment request for deploying the modeling link submitted by the user, and verify whether the modeling link is complete according to the deployment request.

[0058] The generation unit 55 is configured to generate a scheduling task model of the modeling link if the modeling link is complete.

[0059] In an embodiment, the verification unit 54 includes:

[0060] A first acquisition sub-unit is configured to acquire a target modeling link selected by the user from all the modeling links, and receive a deployment request corresponding to the target modeling link.

[0061] A verification sub-unit is configured to verify whether the target modeling link is complete according to the deployment request.

[0062] In an embodiment, the target modeling link is multiple, and the verification sub-unit is configured to verify whether each of the target modeling links is complete.

[0063] In an embodiment, the generating unit 55 is configured to identify a calling part of each component included in the modeling link, mark the calling part with a calling identifier, and integrate the calling identifier into a scheduling task model, wherein the calling part is a running logic of each component for calling a corresponding script for processing.

[0064] In an embodiment, the generating unit 55 is configured to obtain parameters and running logics of each component included in the modeling link, integrate the parameters and running logics of all the components, and obtain a deployment engineering file.

[0065] In an embodiment, the generating unit 55 further includes:

[0066] a displaying sub-unit configured to display a filling prompt of deployment information, and obtain information content corresponding to the deployment information filled by the user;

[0067] an integrating sub-unit configured to integrate the information content and the deployment engineering file into a scheduling task model.

[0068] In an embodiment, the zero-code modeling processing apparatus 50 further includes:

[0069] a deploying unit configured to deploy the scheduling task model into a task training model and a task running model, respectively;

[0070] a processing unit configured to perform task processing by using the task running model, obtain a task processing result, and obtain corresponding actual task data according to the task processing result;

[0071] a training unit configured to train the task training model by using the actual task data, so as to update the task training model.

[0072] It should be noted that the specific implementation process of the above-mentioned zero-code modeling processing apparatus and each unit can be clearly understood by those skilled in the art, which can be referred to the corresponding description in the foregoing method embodiments. For the convenience and brevity of description, it will not be repeated here.

[0073] Meanwhile, the division and connection mode of each unit in the above-mentioned zero-code modeling processing apparatus are only used for example description. In other embodiments, the zero-code modeling processing apparatus can be divided into different units as needed, or each unit in the zero-code modeling processing apparatus can adopt different connection order and mode, so as to complete all or part of the functions of the above-mentioned zero-code modeling processing apparatus.

[0074] The above-mentioned zero-code modeling processing apparatus can be realized in the form of a computer program, which can run on a computer device as shown in the accompanying drawings. Figure 6 The above-mentioned zero-code modeling processing apparatus can be realized in the form of a computer program, which can run on a computer device as shown in the accompanying drawings.

[0075] Referring to Figure 6 , Figure 6 is a schematic block diagram of a computer device provided by an embodiment of the present application. The computer device 500 can be a desktop computer or a server computer, or can be a component or part of other devices.

[0076] Referring to Figure 6 , the computer device 500 includes a processor 502, a memory, and a network interface 505 connected through a system bus 501, wherein the memory can include a non-volatile storage medium 503 and an internal memory 504, and the memory can also be a volatile storage medium.

[0077] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032, when executed, can cause the processor 502 to perform a processing method of zero code modeling.

[0078] The processor 502 is configured to provide computing and control capabilities to support the operation of the entire computer device 500.

[0079] The internal memory 504 provides an environment for the execution of the computer program 5032 in the non-volatile storage medium 503, and the computer program 5032, when executed by the processor 502, can cause the processor 502 to perform a processing method of zero code modeling.

[0080] The network interface 505 is configured to perform network communication with other devices. Those skilled in the art can understand that Figure 6 the structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device 500 to which the scheme of the present application is applied. The specific computer device 500 can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. For example, in some embodiments, the computer device can only include the memory and the processor, and in such embodiments, the structure and function of the memory and the processor are consistent with the embodiment shown in Figure 6 , which will not be described here.

[0081] The processor 502 is configured to run the computer program 5032 stored in the memory to implement the following steps: based on a modeling canvas, obtaining at least two starting components selected by a user, and determining starting nodes corresponding to the starting components according to positions of the starting components on the modeling canvas by the user, wherein the starting components are first components of the modeling links; obtaining from components corresponding to each of the starting components selected by the user, and determining from nodes corresponding to the from components, wherein the from components are subsequent components of the starting components; connecting the starting nodes and the from nodes by using a directed acyclic graph to obtain a modeling link corresponding to each of the starting components; receiving a deployment request of deploying the modeling link submitted by the user, and verifying whether the modeling link is complete according to the deployment request; and if the modeling link is complete, generating a scheduling task model of the modeling link.

[0082] In an embodiment, when implementing the receiving of the deployment request of deploying the modeling link submitted by the user and the verifying of whether the modeling link is complete according to the deployment request, the processor 502 specifically implements the following steps:

[0083] obtaining a target modeling link selected by the user from all the modeling links, and receiving a deployment request corresponding to the target modeling link;

[0084] verifying whether the target modeling link is complete according to the deployment request.

[0085] In an embodiment, the target modeling link is multiple, and when implementing the verifying of whether the target modeling link is complete, the processor 502 specifically implements the following steps:

[0086] verifying whether each of the target modeling links is complete.

[0087] In an embodiment, when implementing the generating of the scheduling task model of the modeling link, the processor 502 specifically implements the following steps:

[0088] identifying calling parts of each component included in the modeling link, and labeling the calling parts by using calling identifiers, and integrating the calling identifiers into the scheduling task model, wherein the calling parts are running logics of each component for calling corresponding scripts for processing.

[0089] In an embodiment, when implementing the generating of the scheduling task model of the modeling link, the processor 502 specifically implements the following steps:

[0090] obtaining parameters and running logics of each component included in the modeling link, integrating the parameters and the running logics of all the components to obtain a deployment engineering file.

[0091] In an embodiment, the processor 502, in implementing the modeling link generation of the scheduling task model, further implements the following steps:

[0092] displaying a filling prompt of deployment information, and obtaining information content corresponding to the deployment information filled by the user;

[0093] integrating the information content and the deployment engineering file into a scheduling task model.

[0094] In an embodiment, after the processor 502 implements the modeling link generation of the scheduling task model, it further implements the following steps:

[0095] deploying the scheduling task model into a task training model and a task running model respectively;

[0096] adopting the task running model to process a task to obtain a task processing result, and obtaining corresponding actual task data according to the task processing result;

[0097] training the task training model by using the actual task data to update the task training model.

[0098] It should be understood that, in the embodiments of the present application, the processor 502 can be a central processing unit (CPU), and the processor 502 can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0099] Those of ordinary skill in the art can understand that all or part of the processes in the method of the above embodiments can be completed by a computer program, which can be stored in a computer readable storage medium. The computer program is executed by at least one processor in the computer system to implement the process steps of the above method embodiments.

[0100] Therefore, the present application also provides a computer readable storage medium. The computer readable storage medium can be a non-volatile computer readable storage medium or a volatile computer readable storage medium, and the computer readable storage medium stores a computer program, and the computer program is executed by a processor to make the processor execute the following steps:

[0101] A computer program product, when running on a computer, causes the computer to execute the steps of the processing method of the zero-code modeling described in the above embodiments.

[0102] The computer readable storage medium can be an internal storage unit of the device, such as a hard disk or a memory of the device. The computer readable storage medium can also be an external storage device of the device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the computer readable storage medium can include both the internal storage unit and the external storage device of the device.

[0103] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device, the apparatus and the unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0104] The storage medium is an entity, a non-transient storage medium, which can be a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, etc. Various entity storage media that can store computer programs.

[0105] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been described in the above description in general terms. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0106] In several embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other manners. For example, the embodiments of the apparatus described above are merely schematic. For example, the division of the units is merely a logical function division. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In a possible implementation process, the steps of the disclosed method can be implemented by using a program, and the program can be stored in a nonvolatile storage medium (such as a ROM, a flash memory, a reliable memory, etc.) or a storage device (such as a disk, a compact disc, etc.). When the program is executed by a processor, the steps of the above method are performed. The storage medium or the storage device is non-volatile, and the program is executed by the processor, thereby realizing the steps of the method.

[0107] The steps in the method embodiments provided in the present application can be adjusted, combined and deleted according to actual needs. The units in the device embodiments provided in the present application can be combined, divided and deleted according to actual needs. In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0108] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a storage medium. Based on such understanding, the technical solutions of the present application essentially or say the part of the prior art that makes a contribution, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing an electronic device (which can be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application.

[0109] The above description is merely a specific implementation of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be covered in the scope of protection of the present application. Therefore, the scope of protection of the present application should be subject to the scope of protection of the claims.

Claims

1. A processing method of zero-code modeling, characterized by, The method comprises the following steps: Based on a modeling canvas, at least two starting components selected by a user are acquired, and a starting node corresponding to each starting component is determined according to the position of each starting component on the modeling canvas, wherein the starting component is the first component of a modeling link; A slave component corresponding to each starting component selected by the user is acquired, and a slave node corresponding to the slave component is determined, wherein the slave component is a subsequent component of the starting component; The starting node and the slave node are connected by a directed acyclic graph to obtain a modeling link corresponding to each starting component; A deployment request for deploying the modeling link submitted by the user is received, and whether the modeling link is complete is verified according to the deployment request; If the modeling link is complete, a scheduling task model is generated from the modeling link.

2. The processing method of zero-code modeling according to claim 1, characterized in that, The method of receiving the deployment request for deploying the modeling link submitted by the user and verifying whether the modeling link is complete according to the deployment request comprises: A target modeling link selected by the user from all the modeling links is acquired, and a deployment request corresponding to the target modeling link is received; Whether the target modeling link is complete is verified according to the deployment request.

3. The processing method of zero-code modeling according to claim 2, wherein, The target modeling link is multiple, and the method of verifying whether the target modeling link is complete comprises: Whether each target modeling link is complete is verified.

4. The processing method of zero-code modeling according to claim 1, wherein, The method of generating a scheduling task model from the modeling link comprises: An invocation part of each component included in the modeling link is identified, the invocation part is marked by an invocation identifier, and the invocation identifier is integrated into a scheduling task model, wherein the invocation part is a running logic of each component for invoking a corresponding script for processing.

5. The method of claim 1-4, wherein, The method of generating a scheduling task model from the modeling link comprises: Parameters and running logics of each component included in the modeling link are acquired, the parameters and running logics of all the components are integrated to obtain a deployment engineering file.

6. The processing method of zero-code modeling according to claim 5, wherein, The method of generating a scheduling task model from the modeling link further comprises: A filling prompt of deployment information is displayed, and information content corresponding to the deployment information filled by the user is acquired; The information content and the deployment engineering file are integrated into a scheduling task model.

7. The processing method of zero-code modeling according to claim 1, wherein, After the method of generating a scheduling task model from the modeling link, the method further comprises: The scheduling task model is deployed into a task training model and a task running model respectively; Task processing is performed by using the task running model to obtain a task processing result, and actual task data corresponding to the task processing result is obtained; The task training model is trained by using the actual task data to update the task training model.

8. A processing device for zero-code modeling, the processing device comprising: The method comprises the following steps: A first acquisition unit is configured to acquire at least two starting components selected by a user based on a modeling canvas, and determine a starting node corresponding to each starting component according to the position of each starting component on the modeling canvas, wherein the starting component is the first component of a modeling link. A second obtaining unit, configured to obtain a slave component corresponding to each of the start components selected by the user, and determine a slave node corresponding to the slave component, wherein the slave component is a subsequent component of the start component; A connecting unit, configured to connect the start node and the slave node by using a directed acyclic graph to obtain a modeling link corresponding to each of the start components; A checking unit, configured to receive a deployment request of deploying the modeling link submitted by the user, and check whether the modeling link is complete according to the deployment request; A generating unit, configured to generate a scheduling task model of the modeling link if the modeling link is complete.

9. A computer device, comprising: The computer device comprises a memory and a processor connected to the memory; the memory is configured to store a computer program; and the processor is configured to run the computer program to perform the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, and the computer program is executable by a processor to implement the steps of the method according to any one of claims 1-7.

Images