Method and system for constructing a machine learning modeling process

Abstract

A method and system for constructing a machine learning modeling process are provided. The method includes: displaying the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process; executing at least one step in the machine learning modeling process in response to a user operation for executing the at least one step; while executing the at least one step, receiving a user modification operation for modifying the machine learning modeling process; and modifying the machine learning modeling process in response to the modification operation, wherein, when executing the modified portion of the machine learning modeling process, executing the process based on the modified machine learning modeling process. According to the method and system, the efficiency of machine learning modeling can be improved.

Description

[0001] This application is a divisional application of a patent application with application date of June 26, 2018, application number 201810669281.8, and titled “Method and system for constructing a machine learning modeling process”. Technical Field

[0002] The present invention generally relates to the field of machine learning, and more particularly, to a method and system for constructing a machine learning modeling process. Background Art

[0003] With the emergence of massive amounts of data, people are turning to machine learning techniques to extract value from this data. Machine learning is an inevitable product of the development of artificial intelligence research to a certain stage. It aims to improve system performance through computational means and the use of experience. In computer systems, "experience" usually exists in the form of "data." Machine learning algorithms can generate "models" from this data. In other words, empirical data is fed to the machine learning algorithm, which then generates a model based on this empirical data. When faced with new situations, the model provides corresponding judgments, i.e., predictions. Therefore, generating models based on empirical data (i.e., the machine learning modeling process) is the key to machine learning technology.

[0004] When building a machine learning modeling process, users need to constantly modify and run the machine learning modeling process until they confirm that the built machine learning modeling process can meet their requirements. However, when building a machine learning modeling process through a machine learning platform (for example, Microsoft's cloud computing system "Azure", etc.), during the running of the built machine learning modeling process, only the configuration of the steps in the machine learning modeling process can be viewed. If the user wants to modify the machine learning modeling process, he can only wait for the entire machine learning modeling process to finish running or manually stop the machine learning modeling process. On the one hand, waiting for the entire machine learning modeling process to finish running will take a long time, which makes it impossible to improve the machine learning modeling process in a timely manner, greatly affecting the modeling efficiency; on the other hand, manually stopping the machine learning modeling process may cause the steps that have been running for a long time to be forced to stop running, so that part of the running results obtained by the step are discarded, resulting in a waste of running work, which in turn affects the modeling efficiency. Summary of the Invention

[0005] An exemplary embodiment of the present invention is to provide a method and system for constructing a machine learning modeling process to solve the problem of low modeling efficiency in the prior art.

[0006] According to an exemplary embodiment of the present invention, a method for constructing a machine learning modeling process is provided, comprising: displaying the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process; executing at least one step in response to a user operation for executing the at least one step in the machine learning modeling process; while executing the at least one step, receiving a modification operation from a user for modifying the machine learning modeling process; and modifying the machine learning modeling process in response to the modification operation, wherein, when executing the modified portion of the machine learning modeling process, executing the process based on the modified machine learning modeling process.

[0007] Optionally, the step of displaying the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process includes: displaying a directed acyclic graph for representing the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process, wherein the nodes in the directed acyclic graph correspond one-to-one to the steps in the machine learning modeling process.

[0008] Optionally, the modification operation includes a configuration operation for modifying the configuration of a downstream step of a currently running step, wherein the step of modifying the machine learning modeling process in response to the modification operation includes: in response to the configuration operation, while running at least one step, modifying the configuration of the step targeted by the configuration operation, wherein when running to the modified step, the modified step is run based on the modified configuration.

[0009] Optionally, the modification operation includes a configuration operation for modifying the configuration of the currently running step or the configuration of the upstream step of the currently running step, wherein the step of modifying the machine learning modeling process in response to the modification operation includes: in response to the configuration operation, stopping the running of the currently running step and modifying the configuration of the step targeted by the configuration operation, wherein the method further includes: when a user operation for running at least one step in the machine learning modeling process is received again, starting to run from the most upstream modified step, wherein when running to the modified step, running the modified step based on the modified configuration.

[0010] Optionally, the modification operation includes a structural adjustment operation for adjusting the downstream structure of the step currently running in the machine learning modeling process, wherein the step of modifying the machine learning modeling process in response to the modification operation includes: in response to the structural adjustment operation, adjusting the structure of the machine learning modeling process while running at least one step, wherein when running to the part of the machine learning modeling process with the adjusted structure, running according to the adjusted structure.

[0011] Optionally, the modification operation includes a structural adjustment operation for adjusting the upstream structure of the step currently running in the machine learning modeling process, wherein the step of modifying the machine learning modeling process in response to the modification operation includes: stopping the currently running step in response to the structural adjustment operation, and adjusting the structure of the machine learning modeling process, wherein the method further includes: when a user operation for running at least one step in the machine learning modeling process is received again, starting to run from the part of the adjusted structure at the most upstream of the machine learning modeling process, wherein when running to the part of the adjusted structure of the machine learning modeling process, running according to the adjusted structure.

[0012] Optionally, the structural adjustment operation includes at least one of an operation for adding steps to the machine learning modeling process, an operation for deleting steps in the machine learning modeling process, and an operation for changing the logical relationship between steps in the machine learning modeling process.

[0013] Optionally, the method further includes: receiving a construction operation from a user to construct another machine learning modeling process in the graphical interface while running the at least one step; and constructing the other machine learning modeling process while running the at least one step in response to the construction operation.

[0014] Optionally, the step of receiving a user's modification operation for modifying the machine learning modeling process while running the at least one step includes: receiving a user's selection operation on a node in the directed acyclic graph while running the at least one step; in response to the selection operation, displaying to the user a control for configuring a configuration item of the step corresponding to the selected node; and receiving the user's input operation on the displayed control.

[0015] Optionally, the method further includes: while executing the at least one step, receiving a user selection operation on a node in the directed acyclic graph; in response to the selection operation, displaying at least one control around the selected node for displaying at least one output element of the step corresponding to the node, and in response to the user's selection operation on one of the at least one controls, displaying to the user an output result of the output element corresponding to the selected control.

[0016] Optionally, the step of receiving a user's modification operation for modifying the machine learning modeling process while running the at least one step includes: while running the at least one step, displaying a node list in a predetermined area of ​​the graphical interface, and receiving a user operation of selecting and dragging a node from the node list to connect to a node in the directed acyclic graph; and / or, while running the at least one step, in response to a user operation on a connection point of a node in the directed acyclic graph, recommending to the user the nodes and / or node combinations to which the node can be connected through the connection point; and receiving an operation of the user selecting a node or a node combination from the recommended nodes and / or node combinations to connect to the connection point.

[0017] Optionally, the method also includes: in response to a user operation for running at least one step in the other machine learning modeling process, running the at least one step in the other machine learning modeling process while running the at least one step in the other machine learning modeling process; while running the at least one step in the other machine learning modeling process, receiving a modification operation from the user for modifying the other machine learning modeling process; in response to the modification operation, modifying the other machine learning modeling process, wherein, when running to the modified part of the other machine learning modeling process, it is run based on the modified machine learning modeling process.

[0018] According to another exemplary embodiment of the present invention, there is provided a system for constructing a machine learning modeling process, comprising: a display device for displaying the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process; an operating device for executing at least one step in the machine learning modeling process in response to a user operation for executing the at least one step; and a constructing device for receiving a modification operation of the user for modifying the machine learning modeling process while the operating device executes the at least one step, and modifying the machine learning modeling process in response to the modification operation, wherein when the operating device executes to the modified part of the machine learning modeling process, it executes based on the modified machine learning modeling process.

[0019] Optionally, the display device displays a directed acyclic graph for representing the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process, wherein the nodes in the directed acyclic graph correspond one-to-one to the steps in the machine learning modeling process.

[0020] Optionally, the modification operation includes a configuration operation for modifying the configuration of a downstream step of a currently running step, wherein, in response to the configuration operation, while the running device runs the at least one step, the construction device modifies the configuration of the step targeted by the configuration operation, wherein, when the running device runs to the modified step, the modified step is run based on the modified configuration.

[0021] Optionally, the modification operation includes a configuration operation for modifying the configuration of the currently running step or the configuration of the upstream step of the currently running step, wherein, in response to the configuration operation, the running device stops running the currently running step, and the construction device modifies the configuration of the step targeted by the configuration operation, wherein, when the user operation for running at least one step in the machine learning modeling process is received again, the running device starts running from the most upstream modified step, wherein, when the running device runs to the modified step, the modified step is run based on the modified configuration.

[0022] Optionally, the modification operation includes a structural adjustment operation for adjusting the downstream structure of the step currently being run in the machine learning modeling process, wherein, in response to the structural adjustment operation, while the running device runs the at least one step, the construction device adjusts the structure of the machine learning modeling process, wherein when the running device runs to the part of the machine learning modeling process whose structure is adjusted, it runs according to the adjusted structure.

[0023] Optionally, the modification operation includes a structural adjustment operation for adjusting the upstream structure of the step currently being run in the machine learning modeling process, wherein, in response to the structural adjustment operation, the running device stops running the step currently being run, and the construction device adjusts the structure of the machine learning modeling process, wherein, when a user operation for running at least one step in the machine learning modeling process is received again, the running device starts running from the part of the adjusted structure at the most upstream of the machine learning modeling process, wherein, when the running device runs to the part of the machine learning modeling process with the adjusted structure, it runs according to the adjusted structure.

[0024] Optionally, the structural adjustment operation includes at least one of an operation for adding steps to the machine learning modeling process, an operation for deleting steps in the machine learning modeling process, and an operation for changing the logical relationship between steps in the machine learning modeling process.

[0025] Optionally, while the running device is running the at least one step, the construction device receives a construction operation from the user to construct another machine learning modeling process in the graphical interface, and in response to the construction operation, constructs the another machine learning modeling process while running the at least one step.

[0026] Optionally, the construction device receives a user selection operation on a node in the directed acyclic graph while the running device runs the at least one step; in response to the selection operation, displays to the user a control for configuring a configuration item of the step corresponding to the selected node; and receives the user's input operation on the displayed control.

[0027] Optionally, the construction device receives a user selection operation on a node in the directed acyclic graph while the running device runs the at least one step; in response to the selection operation, at least one control is displayed around the selected node for displaying at least one output element of the step corresponding to the node, and in response to the user's selection operation on one of the at least one control, the output result of the output element corresponding to the selected control is displayed to the user.

[0028] Optionally, while the running device is running the at least one step, the construction device displays a node list in a predetermined area of ​​the graphical interface and receives a user operation of selecting and dragging a node from the node list to connect to a node in the directed acyclic graph; and / or, while the running device is running the at least one step, the construction device recommends to the user the nodes and / or node combinations to which the node can be connected through the connection point in response to a user operation on a connection point of a node in the directed acyclic graph; and receives an operation of the user selecting a node or a node combination from the recommended nodes and / or node combinations to connect to the connection point.

[0029] Optionally, the running device responds to a user operation for running at least one step in the other machine learning modeling process, and runs the at least one step in the other machine learning modeling process while running the at least one step in the other machine learning modeling process; the building device receives a modification operation from the user for modifying the other machine learning modeling process while the running device runs the at least one step in the other machine learning modeling process, and modifies the other machine learning modeling process in response to the modification operation, wherein when the running device runs to the modified part of the other machine learning modeling process, it runs based on the modified machine learning modeling process.

[0030] According to another exemplary embodiment of the present invention, a computer-readable medium is provided, wherein a computer program for executing the method for constructing a machine learning modeling process as described above is recorded on the computer-readable medium.

[0031] According to another exemplary embodiment of the present invention, a computing device is provided, comprising a storage component and a processor, wherein a set of computer-executable instructions is stored in the storage component, and when the set of computer-executable instructions is executed by the processor, the method for constructing a machine learning modeling process as described above is executed.

[0032] According to the method and system for constructing a machine learning modeling process according to an exemplary embodiment of the present invention, the machine learning modeling process can be modified while the machine learning modeling process is running. In addition, it is also possible to construct and / or run another machine learning modeling process while running one machine learning modeling process.

[0033] According to the method and system for constructing a machine learning modeling process according to an exemplary embodiment of the present invention, the operation mode of the machine learning modeling process is made more flexible. On the one hand, it is convenient for users to continue to improve the machine learning modeling process (for example, modify the structure or step configuration, complete other steps of the machine learning modeling process, etc.) during the operation of a machine learning modeling process, or to construct another machine learning modeling process, so that users can use the running time of the machine learning modeling process to perform other modeling work, that is, the efficiency of the modeling work is improved by increasing the parallelism of the running work; on the other hand, it is convenient for users to adjust the configuration or downstream structure of the downstream step in a timely manner based on the running effect of the upstream step (for example, output results), so that the downstream part can better perform appropriate processing on the running results of the upstream step, thereby greatly improving the flexibility and efficiency of the modeling work.

[0034] Additional aspects and / or advantages of the present general inventive concept will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present general inventive concept. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above and other objects and features of exemplary embodiments of the present invention will become more apparent from the following description taken in conjunction with the accompanying drawings which exemplarily illustrate embodiments, in which:

[0036] Figure 1 A flowchart illustrating a method for constructing a machine learning modeling process according to an exemplary embodiment of the present invention;

[0037] Figure 2 An example of constructing a machine learning modeling process according to an exemplary embodiment of the present invention is shown;

[0038] Figure 3 shows an example of a control for displaying an output element of a step corresponding to a selected node according to an exemplary embodiment of the present invention;

[0039] Figure 4 An example of presenting an output result of a step corresponding to a selected node to a user according to an exemplary embodiment of the present invention is shown;

[0040] Figure 5 Another example of constructing a machine learning modeling process according to an exemplary embodiment of the present invention is shown;

[0041] Figure 6 A block diagram of a system for building a machine learning modeling process according to an exemplary embodiment of the present invention is shown. DETAILED DESCRIPTION

[0042] Reference will now be made in detail to the embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to like parts throughout. The embodiments will be described below with reference to the drawings in order to explain the present invention.

[0043] Figure 1 A flowchart of a method for constructing a machine learning modeling process according to an exemplary embodiment of the present invention is shown. Here, as an example, the method can be executed by a computer program, or by a hardware device or a collection of hardware and software resources specifically used to perform machine learning, big data computing, or data analysis. For example, the method can be executed by a machine learning platform used to implement machine learning-related services.

[0044] Reference Figure 1 In step S10, the constructed machine learning modeling process is displayed in a graphical interface for constructing a machine learning modeling process.

[0045] As an example, you can first enter a graphical interface for constructing a machine learning modeling process, and then, when you receive an operation from a user to open a file representing the machine learning modeling process, the machine learning modeling process defined by the content of the file is displayed in the graphical interface; or when you receive an operation from a user to request the creation of a machine learning modeling process, you can respond in real time to the user's construction operation for constructing the machine learning modeling process and display the constructed machine learning modeling process in the graphical interface.

[0046] Furthermore, before the constructed machine learning modeling process is run, the user can continue to construct the displayed machine learning modeling process through the graphical interface. For example, in response to a user's configuration operation for modifying the configuration of a step, the configuration of the step targeted by the configuration operation can be modified; in response to a user's structure adjustment operation for adjusting the structure of the machine learning modeling process, the structure of the machine learning modeling process can be adjusted. As an example, the structure adjustment operation may include at least one of an operation for adding a step to the machine learning modeling process, an operation for deleting a step in the machine learning modeling process, and an operation for changing the logical relationship between steps in the machine learning modeling process. Here, the logical relationship between the steps is also the input-output association relationship between the steps, for example, the output of one step serves as the input of another step.

[0047] As an example, the constructed machine learning modeling process may include at least one of the following steps: data import, data splicing, data splitting, feature extraction, model training, model testing and model evaluation. Specifically, the data import step is used to import one or more data sets (for example, data tables) containing historical data records; the data splicing step is used to splice the data records in the imported multiple data sets; the data splitting step is used to split the spliced ​​data records into a training set and a test set, or to split the data records in an imported data set into a training set and a test set, wherein the data records in the training set are used to be converted into training samples to train a model, and the data records in the test set are used to be converted into test samples to evaluate the model effect according to the test results of the trained model for the test samples; the feature extraction step is used to extract features from the training set and the test set to generate training samples and test samples; the model training step is used to train a machine learning model based on the training samples according to the machine learning algorithm; the model testing step is used to obtain the test results of the trained machine learning model for the test samples; the model evaluation step is used to evaluate the effect of the trained machine learning model based on the accuracy of the test results.

[0048] As an example, a directed acyclic graph (DAG) representing the machine learning modeling process can be displayed in a graphical interface for constructing the machine learning modeling process, wherein the nodes in the directed acyclic graph correspond one-to-one to the steps in the machine learning modeling process. The user can construct or run the corresponding machine learning modeling process by editing or running the displayed DAG.

[0049] In step S20, in response to a user operation for executing at least one step in the machine learning modeling process, the at least one step is executed.

[0050] Here, as an example, a user's selection of one or more executable steps in the machine learning modeling process may be received, and the corresponding steps may be sequentially executed according to the user's selection.

[0051] In step S30, while executing at least one of the steps, a modification operation of the user for modifying the machine learning modeling process is received.

[0052] As an example, the modification operation may include: a configuration operation for modifying the configuration of a step in the machine learning modeling process and / or a structural adjustment operation for adjusting the structure of the machine learning modeling process. As an example, the structural adjustment operation may include: at least one of an operation for adding a step to the machine learning modeling process, an operation for deleting a step in the machine learning modeling process, and an operation for changing the logical relationship between steps in the machine learning modeling process. As an example, when a directed acyclic graph for representing the constructed machine learning modeling process is displayed in a graphical interface for constructing the machine learning modeling process, the structural adjustment operation may include: at least one of adding a node to the directed acyclic graph, deleting a node from the directed acyclic graph, and changing the connection relationship between nodes in the directed acyclic graph.

[0053] In step S40, in response to the modification operation, the machine learning modeling process is modified, wherein when running to the modified part of the machine learning modeling process, it is run based on the modified machine learning modeling process.

[0054] As an example, when the modification operation is a configuration operation for modifying the configuration of a downstream step of a currently running step, in response to the configuration operation, while running the at least one step, the configuration of the step targeted by the configuration operation is modified, wherein when the modified step is run, the modified step is run based on the modified configuration.

[0055] Here, the downstream step of the currently running step refers to a step that should be executed after the currently running step in the execution order. It should be understood that the step targeted by the configuration operation can be a step that will be executed in the current execution (i.e., a step in the at least one step) or a step that will not be executed in the current execution (i.e., a step not in the at least one step).

[0056] As an example, when the modification operation is a structural adjustment operation for adjusting the downstream structure of a step currently being run in the machine learning modeling process, in response to the structural adjustment operation, the structure of the machine learning modeling process is adjusted while running at least one step, wherein when the part of the machine learning modeling process with the adjusted structure is run, it is run according to the adjusted structure.

[0057] Here, the downstream structure of the currently running step in the machine learning modeling process refers to the structure of the portion of the machine learning modeling process that should be executed after the currently running step. It should be understood that the structure adjustment operation can be used to adjust the structure of the portion that will be executed in the current execution, and can also be used to adjust the structure of the portion that will not be executed in the current execution.

[0058] According to the above-described exemplary embodiments of the present invention, it is possible to flexibly adjust the configuration or downstream structure of the downstream steps of the currently running step without affecting the operation of the machine learning modeling process.

[0059] As an example, when the modification operation is a configuration operation for modifying the configuration of the currently running step or the configuration of the upstream step of the currently running step, in response to the configuration operation, the currently running step is stopped, and the configuration of the step targeted by the configuration operation is modified, wherein the method for constructing a machine learning modeling process according to an exemplary embodiment of the present invention also includes: when a user operation for running at least one step in the machine learning modeling process is received again, running starts from the most upstream modified step, wherein when running to the modified step, the modified step is run based on the modified configuration.

[0060] Here, the upstream step of the currently running step refers to: the step that should be run before the currently running step in the order of execution. It should be understood that the step targeted by the configuration operation can be a step that has been run in this operation or a step that is being run (that is, a step in the at least one step), or a step that will not be run in this operation (that is, not a step in the at least one step). Here, the most upstream modified step is the step with the highest running order among the modified steps. Since the upstream step is modified, its output result will inevitably change accordingly and affect the operation of the downstream steps. Therefore, when the user operation for running at least one step in the machine learning modeling process is received again, even if the at least one step does not include the most upstream modified step, it is necessary to start running from the most upstream modified step.

[0061] As an example, when the modification operation is a structural adjustment operation for adjusting the upstream structure of the step currently being run in the machine learning modeling process, in response to the structural adjustment operation, the currently running step is stopped and the structure of the machine learning modeling process is adjusted, wherein the method for constructing a machine learning modeling process according to an exemplary embodiment of the present invention also includes: when a user operation for running at least one step in the machine learning modeling process is received again, running is started from the part of the adjusted structure at the most upstream of the machine learning modeling process, wherein when running to the part of the adjusted structure of the machine learning modeling process, it is run according to the adjusted structure.

[0062] Here, the upstream structure of the currently running step in the machine learning modeling process refers to the structure of the portion of the machine learning modeling process that should be executed before the currently running step. It should be understood that the structure adjustment operation can be used to adjust the structure of a portion that has already been executed in the current run, and can also be used to adjust the structure of a portion that has not been executed in the current run.

[0063] As an example, steps S20 to S40 can be implemented in the following manner: the method can be executed by a machine learning platform for executing a machine learning process, and in response to a user operation for running at least one step in the displayed machine learning modeling process, a task for executing the corresponding step can be submitted to an operating device of the machine learning platform (for example, a server for executing task workflows located at the back end of the machine learning platform) to sequentially execute the at least one step according to the configuration of the at least one step and the logical relationship between the steps. While the operating device is executing the at least one step, a configuration operation is received, and in response to the configuration operation, it is determined which of the following types the step targeted by the configuration operation belongs to: a to-be-executed step downstream of the currently running step (i.e., a step in the running queue), a non-to-be-executed step downstream of the currently running step (i.e., a step not in the running queue), the currently running step and its upstream step. When it is determined that there is a step to be run that is downstream of the currently running step, the configuration of the step targeted by the configuration operation is modified, and the operating device is notified that the configuration of the step has been modified, so that the operating device runs the step according to the modified configuration when running the step; when it is determined that there is a non-step to be run that is downstream of the currently running step, the configuration of the step targeted by the configuration operation is modified; when it is determined that there is a step that is currently running and its upstream step, the operating device is notified to stop running at least one step, and the configuration of the step targeted by the configuration operation is modified.

[0064] Accordingly, while the operating device is executing the at least one step, a structural adjustment operation is received. In response to the structural adjustment operation, it is determined whether the portion targeted by the structural adjustment operation belongs to the following types: a downstream structure of the currently running step including a step to be executed downstream of the currently running step, a downstream structure of the currently running step excluding a step to be executed downstream of the currently running step, or an upstream structure of the currently running step. If it is determined that the portion targeted by the structural adjustment operation belongs to the downstream structure of the currently running step including a step to be executed downstream of the currently running step, the structure of the portion targeted by the structural adjustment operation is adjusted, and the operating device is notified that the structure of the portion has been modified, so that the operating device operates according to the modified structure when operating the portion. If it is determined that the portion targeted by the structural adjustment operation belongs to the downstream structure of the currently running step excluding a step to be executed downstream of the currently running step, the structure of the portion targeted by the structural adjustment operation is adjusted. If it is determined that the portion targeted by the structural adjustment operation belongs to the upstream structure of the currently running step, the operating device is notified to stop executing the at least one step, and the structure of the portion targeted by the structural adjustment operation is adjusted.

[0065] As an example, when a directed acyclic graph representing the constructed machine learning modeling process is displayed in a graphical interface for constructing a machine learning modeling process, and a step corresponding to at least one node in the directed acyclic graph is executed, a configuration operation can be received while executing the at least one step in the following manner: while executing the at least one step, a user selection operation on a node in the directed acyclic graph is received, and in response to the selection operation, a control for configuring a configuration item of the step corresponding to the selected node is displayed to the user, and an input operation of the user on the displayed control is received. For example, the node selection operation can be an operation of clicking the node with the left button of the mouse. For example, a control for configuring the configuration item of the step corresponding to the selected node can be displayed in a predetermined area of ​​the graphical interface.

[0066] As an example, when a directed acyclic graph representing the constructed machine learning modeling process is displayed in a graphical interface for constructing a machine learning modeling process, and a step corresponding to at least one node in the directed acyclic graph is executed, a structural adjustment operation can be received while executing the at least one step in the following manner: while executing the at least one step, a node list is displayed in a predetermined area of ​​the graphical interface, and a user operation of selecting and dragging a node from the node list to connect to a node in the directed acyclic graph is received; and / or, while executing the at least one step, in response to a user operation on a connection point of a node in the directed acyclic graph, nodes and / or node combinations to which the node can be connected through the connection point are recommended to the user, and an operation of the user selecting a node or a node combination from the recommended nodes and / or node combinations to connect to the connection point is received. For example, the nodes and / or node combinations to which the node can be connected through the connection point can be displayed around the connection point. For example, the user operation on a connection point of a node in the directed acyclic graph may include: hovering over a connection point of a node in the directed acyclic graph, and clicking the connection point after the connection point enters a waiting-to-connect state in response to the hovering operation.

[0067] like Figure 2As shown, a DAG diagram representing the machine learning modeling process is displayed in a graphical interface for constructing a machine learning modeling process, and the steps corresponding to the "HE-TreeNet" node in the DAG diagram are currently being run. While continuing to run the steps corresponding to the "HE-TreeNet" node, controls for configuring the model test steps can be displayed in the right area of ​​the graphical interface in response to the user's selection operation on the model test node, and the configuration of the model test step is modified in response to the user's input operation on the displayed control, thereby achieving adjustment of the configuration of the model test step before running to run the model test step, so as to run the model test step based on the adjusted configuration. In addition, while running the steps corresponding to the "HE-TreeNet" node, in response to the user's selection operation on the feature extraction node, a control for configuring the configuration items of the feature extraction step can be displayed in the right area of ​​the graphical interface. Then, in response to the user's input operation on the displayed control, the step corresponding to the "HE-TreeNet" node is stopped, and the configuration of the feature extraction step is modified. When the user operation for running the DAG graph is received again, since the upstream steps of the feature extraction step have not changed, the operation can be started from the feature extraction step, and the feature extraction step can be run based on the modified configuration. In addition, a node list can be displayed in the left area of ​​the graphical interface. While running the steps corresponding to the "HE-TreeNet" node, a structural adjustment operation of selecting and dragging a node from the node list to connect to a node in the directed acyclic graph can be received.

[0068] Furthermore, as an example, the method for constructing a machine learning modeling process according to an exemplary embodiment of the present invention may further include: when a directed acyclic graph representing the constructed machine learning modeling process is displayed in a graphical interface for constructing the machine learning modeling process, and a step corresponding to at least one node in the directed acyclic graph is executed, receiving a user selection operation on a node in the directed acyclic graph, and in response to the selection operation, displaying at least one control around the selected node for displaying at least one output element of the step corresponding to the node, and in response to the user selecting one of the at least one control, displaying the output result of the output element corresponding to the selected control to the user. The at least one output element of a step is the at least one element output by the step. As an example, the at least one control may have a one-to-one correspondence with the at least one output element. As an example, the at least one control may be applied with a corresponding visual effect based on the type of the corresponding output element, where different visual effects correspond to different types. As an example, the output result of the output element may include the current output result and / or historical output results of the output element. Here, the current output result refers to the output result obtained after the step corresponding to the selected node is currently executed, and the historical output result refers to the output result obtained after the step corresponding to the selected node was executed before the current execution. As an example, the visual effect of the at least one control being displayed can also be used to distinguish whether the corresponding output element has the result of the current execution.

[0069] As an example, the output result of the output element can be the specific output content of the output element itself, or it can be related information of the specific output content, for example, it can be the size of the specific output content, the channel entrance for accessing the specific output content, etc. It should be understood that the types of multiple output elements of the same step can be the same or different, and the types of output elements of different steps can be the same or different. As an example, the type of the output element may include at least one of the following types: a data table, information for defining a machine learning model, an evaluation report, and an analysis report. For example, with respect to the data table, it can be a data table as a training set and a data table as a test set output by the data splitting step, it can be a data table as a training sample and a data table as a test sample output by the feature extraction step, and it can also be a data table indicating the test results output by the model testing step; the information used to define the machine learning model can be the parameters of the machine learning model; the evaluation report can be a report for evaluating the test effect of the machine learning model; the analysis report can be a report on the analysis performed during the running step, for example, a report on the feature importance analysis performed during the running feature extraction step.

[0070] like Figure 3As shown, in response to a user's selection operation on a node in a running directed acyclic graph, at least one control for displaying at least one output element of the step corresponding to the node may be displayed around the selected node. The types of the multiple output elements of the step may be the same or different, and the controls corresponding to the different types of output elements are displayed separately. Figure 4 As shown, in response to the user's selection operation on the controls displayed around the data splitting node, the output result of the output element corresponding to the selected control can be displayed in the right area of ​​the graphical interface, and the connection between the data splitting node and the subsequent node (i.e., the feature extraction node) to which the output element corresponding to the selected control is applied can be highlighted (e.g., highlighted). The output result of the data splitting step can be displayed to the user by displaying the size of the specific output content of the output element corresponding to the selected control, the channel entrance for accessing the specific output content, etc. In addition, the display of the current output result and the historical output result can be switched according to the user's selection. It should be understood that the specific interaction scenarios and operation details of the exemplary embodiments of the present invention when displaying the output result of the step corresponding to the selected node to the user are not limited to Figure 3 and Figure 4 Example shown.

[0071] According to the above exemplary embodiments of the present invention, it is convenient for users to view the output results of the upstream steps and to adjust the configuration or downstream structure of the downstream steps in a timely manner so that the downstream part can better process the output results of the upstream steps appropriately, thereby greatly improving the flexibility and efficiency of the modeling work.

[0072] In addition, as an example, the method for constructing a machine learning modeling process according to an exemplary embodiment of the present invention may also include: while running the at least one step, receiving a construction operation of the user to construct another machine learning modeling process in the graphical interface; in response to the construction operation, while running the at least one step, constructing the other machine learning modeling process.

[0073] Furthermore, as an example, the method for constructing a machine learning modeling process according to an exemplary embodiment of the present invention may further include: in response to a user operation for running at least one step in the other machine learning modeling process, running the at least one step in the other machine learning modeling process while running the at least one step in the machine learning modeling process. According to an exemplary embodiment of the present invention, multiple machine learning modeling processes can be run simultaneously, and the operations of each other do not affect each other.

[0074] Further, as an example, the method for constructing a machine learning modeling process according to an exemplary embodiment of the present invention may also include: while running the at least one step in the other machine learning modeling process, receiving a modification operation from a user for modifying the other machine learning modeling process; in response to the modification operation, modifying the other machine learning modeling process, wherein, when running to the modified part of the other machine learning modeling process, running based on the modified machine learning modeling process.

[0075] like Figure 5 As shown, a DAG graph representing the machine learning modeling process is displayed in a graphical interface for constructing a machine learning modeling process, and the steps corresponding to the "HE-TreeNet" node in the DAG graph are currently being executed. A node list can be displayed in the left area of ​​the graphical interface. While continuing to execute the DAG graph, a user operation of selecting and adding nodes from the node list to the canvas area of ​​the graphical interface and connecting the newly added nodes to each other can be received, and another DAG graph is created in response to the user operation. In addition, in response to the user operation, the newly created DAG graph can be executed while the original DAG graph is being executed. Furthermore, while the newly created DAG graph is being executed, a modification operation of the user for modifying the newly created DAG graph can be received; in response to the modification operation, the newly created DAG graph is modified, wherein when the modified part of the newly created DAG graph is executed, the operation is executed based on the modified DAG graph. It should be understood that the specific interaction scenarios and operation details of the operation of the machine learning modeling process according to the exemplary embodiment of the present invention are not limited to Figure 2 and Figure 5 Example shown.

[0076] Figure 6 1 shows a block diagram of a system for constructing a machine learning modeling process according to an exemplary embodiment of the present invention. Figure 6 As shown, a system for constructing a machine learning modeling process according to an exemplary embodiment of the present invention includes: a display device 10, an operating device 20 and a constructing device 30.

[0077] Specifically, the display device 10 is used to display the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process.

[0078] As an example, the display device 10 may display a directed acyclic graph for representing the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process, wherein the nodes in the directed acyclic graph correspond one-to-one to the steps in the machine learning modeling process.

[0079] The running device 20 is used to run at least one step in response to a user operation for running at least one step in the machine learning modeling process.

[0080] The construction device 30 is used to receive a modification operation from a user for modifying the machine learning modeling process while the running device 20 runs the at least one step, and modify the machine learning modeling process in response to the modification operation, wherein when the running device 20 runs to the modified part of the machine learning modeling process, it runs based on the modified machine learning modeling process.

[0081] As an example, the modification operation may include a configuration operation for modifying the configuration of a downstream step of a currently running step, wherein, in response to the configuration operation, while the running device 20 runs the at least one step, the construction device 30 may modify the configuration of the step targeted by the configuration operation, wherein, when the running device 20 runs to the modified step, the modified step is run based on the modified configuration.

[0082] As an example, the modification operation may include a configuration operation for modifying the configuration of a currently running step or the configuration of an upstream step of the currently running step, wherein, in response to the configuration operation, the running device 20 stops running the currently running step, and the construction device 30 modifies the configuration of the step targeted by the configuration operation, wherein, when a user operation for running at least one step in the machine learning modeling process is received again, the running device 20 starts running from the most upstream modified step, wherein, when the running device 20 runs to the modified step, the modified step is run based on the modified configuration.

[0083] As an example, the modification operation may include a structural adjustment operation for adjusting the downstream structure of a step currently being run in the machine learning modeling process, wherein, in response to the structural adjustment operation, while the running device 20 runs the at least one step, the construction device 30 adjusts the structure of the machine learning modeling process, wherein, when the running device 20 runs to the part of the machine learning modeling process whose structure is adjusted, it runs according to the adjusted structure.

[0084] As an example, the modification operation may include a structural adjustment operation for adjusting the upstream structure of a step currently being run in the machine learning modeling process, wherein, in response to the structural adjustment operation, the running device 20 stops running the currently running step, and the construction device 30 adjusts the structure of the machine learning modeling process, wherein, when a user operation for running at least one step in the machine learning modeling process is received again, the running device 20 starts running from the most upstream adjusted structure of the machine learning modeling process, wherein, when the running device 20 runs to the adjusted structure of the machine learning modeling process, it runs according to the adjusted structure.

[0085] As an example, the structural adjustment operation may include at least one of an operation for adding steps to the machine learning modeling process, an operation for deleting steps in the machine learning modeling process, and an operation for changing the logical relationship between steps in the machine learning modeling process.

[0086] As an example, the construction device 30 may receive a user selection operation on a node in the directed acyclic graph while the running device 20 runs the at least one step; in response to the selection operation, display to the user a control for configuring a configuration item of the step corresponding to the selected node; and receive the user's input operation on the displayed control.

[0087] As an example, the construction device 30 may receive a user selection operation on a node in the directed acyclic graph while the running device 20 runs the at least one step; in response to the selection operation, at least one control is displayed around the selected node for displaying at least one output element of the step corresponding to the node, and in response to the user's selection operation on one of the at least one controls, the output result of the output element corresponding to the selected control is displayed to the user.

[0088] As an example, the construction device 30 may display a node list in a predetermined area of ​​the graphical interface while the running device 20 runs the at least one step, and receive a user operation of selecting and dragging a node from the node list to connect to a node in the directed acyclic graph.

[0089] As an example, the construction device 30 may, in response to a user operation on a connection point of a node in the directed acyclic graph while the operation device 20 is executing the at least one step, recommend to the user nodes and / or node combinations to which the node can be connected through the connection point; and receive an operation of the user selecting a node or a node combination from the recommended nodes and / or node combinations to connect to the connection point.

[0090] As an example, the construction device 30 may receive a construction operation from the user to construct another machine learning modeling process in the graphical interface while the running device 20 runs the at least one step, and in response to the construction operation, construct the other machine learning modeling process while running the at least one step.

[0091] As an example, the running device 20 may run the at least one step in the other machine learning modeling process while running the at least one step in the machine learning modeling process in response to a user operation for running the at least one step in the other machine learning modeling process.

[0092] As an example, the construction device 30 may receive a modification operation from a user for modifying the other machine learning modeling process while the running device 20 is running the at least one step in the other machine learning modeling process, and modify the other machine learning modeling process in response to the modification operation, wherein when the running device 20 runs to the modified part of the other machine learning modeling process, it runs based on the modified machine learning modeling process.

[0093] It should be understood that the specific implementation of the system for constructing a machine learning modeling process according to an exemplary embodiment of the present invention can be referred to in conjunction with Figures 1 to 5 The relevant specific implementation methods described are implemented and will not be repeated here.

[0094] The devices included in the system for constructing a machine learning modeling process according to an exemplary embodiment of the present invention can be configured as software, hardware, firmware, or any combination of the above items to perform specific functions. For example, these devices may correspond to dedicated integrated circuits, pure software codes, or modules that combine software and hardware. In addition, one or more functions implemented by these devices may also be uniformly executed by components in physical entity devices (e.g., processors, clients, or servers, etc.).

[0095] It should be understood that the method for constructing a machine learning modeling process according to an exemplary embodiment of the present invention can be implemented by a program recorded on a computer-readable medium. For example, according to an exemplary embodiment of the present invention, a computer-readable medium for constructing a machine learning modeling process can be provided, wherein a computer program for executing the following method steps is recorded on the computer-readable medium: displaying the constructed machine learning modeling process in a graphical interface for constructing a machine learning modeling process; running the at least one step in response to a user operation for running the at least one step in the machine learning modeling process; while running the at least one step, receiving a modification operation from the user for modifying the machine learning modeling process; and modifying the machine learning modeling process in response to the modification operation, wherein, when running to the modified part of the machine learning modeling process, running based on the modified machine learning modeling process.

[0096] The computer program in the computer-readable medium can be run in an environment deployed in a computer device such as a client, a host, an agent device, a server, etc. It should be noted that the computer program can also be used to perform additional steps in addition to the above steps or perform more specific processing when performing the above steps. The contents of these additional steps and further processing have been referred to in the Figures 1 to 5 To avoid repetition, it will not be described again here.

[0097] It should be noted that the system for constructing a machine learning modeling process according to an exemplary embodiment of the present invention can completely rely on the operation of a computer program to realize the corresponding functions, that is, each device corresponds to each step in the functional architecture of the computer program, so that the entire system is called through a special software package (for example, lib library) to realize the corresponding functions.

[0098] On the other hand, the various devices included in the system for constructing a machine learning modeling process according to an exemplary embodiment of the present invention may also be implemented by hardware, software, firmware, middleware, microcode, or any combination thereof. When implemented in software, firmware, middleware, or microcode, the program code or code segment for performing the corresponding operation may be stored in a computer-readable medium such as a storage medium, so that the processor can perform the corresponding operation by reading and running the corresponding program code or code segment.

[0099] For example, an exemplary embodiment of the present invention can also be implemented as a computing device, which includes a storage component and a processor, wherein a set of computer-executable instructions is stored in the storage component, and when the set of computer-executable instructions is executed by the processor, a method for constructing a machine learning modeling process is executed.

[0100] Specifically, the computing device may be deployed in a server or client, or may be deployed on a node device in a distributed network environment. In addition, the computing device may be a PC, tablet device, personal digital assistant, smartphone, web application, or other device capable of executing the above-mentioned instruction set.

[0101] Here, the computing device is not necessarily a single computing device, but may be any collection of devices or circuits capable of executing the above instructions (or instruction sets) individually or in combination. The computing device may also be part of an integrated control system or system manager, or may be configured as a portable electronic device that is interconnected with a local or remote (e.g., via wireless transmission) interface.

[0102] In the computing device, the processor may include a central processing unit (CPU), a graphics processing unit (GPU), a programmable logic device, a dedicated processor system, a microcontroller, or a microprocessor. By way of example and not limitation, the processor may also include an analog processor, a digital processor, a microprocessor, a multi-core processor, a processor array, a network processor, etc.

[0103] According to an exemplary embodiment of the present invention, some operations described in the method for constructing a machine learning modeling process can be implemented by software, some operations can be implemented by hardware, and furthermore, these operations can also be implemented by a combination of software and hardware.

[0104] The processor may execute instructions or codes stored in one of the memory components, wherein the memory component may also store data. Instructions and data may also be sent and received over a network via a network interface device, wherein the network interface device may employ any known transmission protocol.

[0105] The storage component can be integrated with the processor, for example, by placing RAM or flash memory within an integrated circuit microprocessor or the like. Furthermore, the storage component can include a separate device, such as an external disk drive, a storage array, or any other storage device usable by a database system. The storage component and the processor can be operatively coupled or can communicate with each other, for example, via an I / O port, a network connection, or the like, such that the processor can access files stored in the storage component.

[0106] In addition, the computing device may also include a video display (such as a liquid crystal display) and a user interaction interface (such as a keyboard, mouse, touch input device, etc.) All components of the computing device may be connected to each other via a bus and / or a network.

[0107] The operations involved in the method for constructing a machine learning modeling process according to an exemplary embodiment of the present invention can be described as various interconnected or coupled functional blocks or functional diagrams. However, these functional blocks or functional diagrams can be equally integrated into a single logical device or operate according to non-precise boundaries.

[0108] For example, as described above, a computing device for constructing a machine learning modeling process according to an exemplary embodiment of the present invention may include a storage component and a processor, wherein a set of computer-executable instructions is stored in the storage component, and when the set of computer-executable instructions is executed by the processor, the following steps are performed: displaying the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process; executing at least one step in response to a user operation for executing the at least one step in the machine learning modeling process; while executing the at least one step, receiving a modification operation from the user for modifying the machine learning modeling process; modifying the machine learning modeling process in response to the modification operation, wherein when the modified part of the machine learning modeling process is executed, the process is executed based on the modified machine learning modeling process.

[0109] While various exemplary embodiments of the present invention have been described above, it should be understood that the foregoing description is merely illustrative and not exhaustive, and the present invention is not limited to the disclosed exemplary embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for constructing a machine learning modeling process, comprising: Displaying the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process; In response to a user operation for executing at least one step in the machine learning modeling process, executing the at least one step, wherein the at least one step includes at least one of the following steps: data import, data splicing, data splitting, feature extraction, model training, model testing, and model evaluation; While executing the at least one step, receiving a modification operation from a user for modifying the machine learning modeling process; In response to the modification operation, modifying the machine learning modeling process, wherein, when running to the modified part of the machine learning modeling process, running based on the modified machine learning modeling process; The modification operation includes a configuration operation for modifying the configuration of a currently running step or a configuration of an upstream step of the currently running step, and the step of modifying the machine learning modeling process in response to the modification operation includes: In response to the configuration operation, the currently running step is stopped and the configuration of the step targeted by the configuration operation is modified. The method further includes: when receiving a user operation for executing at least one step in the machine learning modeling process again, executing the process from the most upstream modified step, When the modified step is reached, the modified step is executed based on the modified configuration.

2. The method according to claim 1, wherein The steps of displaying the constructed machine learning modeling process in the graphical interface for constructing the machine learning modeling process include: A directed acyclic graph representing the constructed machine learning modeling process is displayed in a graphical interface for constructing the machine learning modeling process, wherein the nodes in the directed acyclic graph correspond one-to-one to the steps in the machine learning modeling process.

3. The method according to claim 1, wherein The modification operation includes a configuration operation for modifying the configuration of a downstream step of a currently running step, wherein the step of modifying the machine learning modeling process in response to the modification operation includes: In response to the configuration operation, while executing the at least one step, the configuration of the step targeted by the configuration operation is modified, When the modified step is reached, the modified step is executed based on the modified configuration.

4. The method according to claim 1, wherein The modification operation includes a structure adjustment operation for adjusting a downstream structure of a currently running step in the machine learning modeling process, wherein the step of modifying the machine learning modeling process in response to the modification operation includes: In response to the restructuring operation, while executing the at least one step, the structure of the machine learning modeling process is adjusted, Among them, when running to the part of the machine learning modeling process whose structure is adjusted, it runs according to the adjusted structure.

5. The method according to claim 1, wherein The modification operation includes a structure adjustment operation for adjusting an upstream structure of a currently running step in the machine learning modeling process, wherein the step of modifying the machine learning modeling process in response to the modification operation includes: In response to the structure adjustment operation, stopping the currently running step and adjusting the structure of the machine learning modeling process, The method further includes: when receiving a user operation for executing at least one step in the machine learning modeling process again, executing the process from the most upstream portion of the adjusted structure of the machine learning modeling process, Among them, when running to the part of the machine learning modeling process whose structure is adjusted, it runs according to the adjusted structure.

6. The method according to claim 4 or 5, wherein: The structural adjustment operation includes at least one of an operation for adding steps to the machine learning modeling process, an operation for deleting steps in the machine learning modeling process, and an operation for changing the logical relationship between steps in the machine learning modeling process.

7. The method according to claim 1, wherein The method further comprises: While executing the at least one step, receiving a construction operation from a user to construct another machine learning modeling process in the graphical interface; In response to the building operation, the other machine learning modeling process is built while executing the at least one step.

8. The method according to claim 2, wherein: The step of receiving a modification operation from a user for modifying the machine learning modeling process while executing the at least one step includes: While executing the at least one step, receiving a user's selection operation on a node in the directed acyclic graph; In response to the selection operation, displaying to the user a control for configuring a configuration item of the step corresponding to the selected node; Receive user input operations on displayed controls.

9. The method according to claim 2, wherein: The method further comprises: While executing the at least one step, receiving a user's selection operation on a node in the directed acyclic graph; In response to the selection operation, at least one control for displaying at least one output element of the step corresponding to the node is displayed around the selected node, and in response to the user's selection operation on one of the at least one controls, the output result of the output element corresponding to the selected control is displayed to the user.

10. The method according to claim 2, wherein: The step of receiving a modification operation from a user for modifying the machine learning modeling process while executing the at least one step includes: While executing the at least one step, displaying a node list in a predetermined area of ​​the graphical interface, and receiving a user operation of selecting and dragging a node from the node list to connect to a node in the directed acyclic graph; And / or, while running at least one of the steps, in response to a user operation on a connection point of a node in the directed acyclic graph, recommend to the user the nodes and / or node combinations to which the node can be connected through the connection point; and receive the user's operation of selecting a node or a node combination from the recommended nodes and / or node combinations to connect to the connection point.

11. The method according to claim 7, wherein: The method further comprises: In response to a user operation for executing at least one step in the other machine learning modeling process, executing the at least one step in the other machine learning modeling process concurrently with executing the at least one step in the machine learning modeling process; While executing the at least one step in the another machine learning modeling process, receiving a modification operation from a user for modifying the another machine learning modeling process; In response to the modification operation, modifying the other machine learning modeling process, Wherein, when running to the modified part of the said another machine learning modeling process, it runs based on the modified machine learning modeling process.

12. A system for constructing a machine learning modeling process, comprising: A display device, configured to display the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process; An operating device, configured to execute at least one step in the machine learning modeling process in response to a user operation for executing the at least one step, wherein the at least one step comprises at least one of the following steps: data import, data splicing, data splitting, feature extraction, model training, model testing, and model evaluation; Constructing means for receiving a modification operation from a user for modifying the machine learning modeling process while the operating means is executing the at least one step, and modifying the machine learning modeling process in response to the modification operation, When the operating device reaches the modified part of the machine learning modeling process, the operating device operates based on the modified machine learning modeling process; The modification operation includes a configuration operation for modifying the configuration of the currently running step or the configuration of the upstream step of the currently running step, wherein, in response to the configuration operation, the running device stops running the currently running step, and the construction device modifies the configuration of the step targeted by the configuration operation. When a user operation for executing at least one step in the machine learning modeling process is received again, the executing device starts executing from the modified step at the most upstream end. When the operating device runs to the modified step, the modified step is run based on the modified configuration.

13. The system according to claim 12, wherein: The display device displays a directed acyclic graph for representing the constructed machine learning modeling process in a graphical interface for constructing the machine learning modeling process, wherein the nodes in the directed acyclic graph correspond one-to-one to the steps in the machine learning modeling process.

14. The system according to claim 12, wherein: The modification operation includes a configuration operation for modifying the configuration of a step downstream of the currently running step, wherein, in response to the configuration operation, while the running device runs the at least one step, the construction device modifies the configuration of the step targeted by the configuration operation. When the operating device runs to the modified step, the modified step is run based on the modified configuration.

15. The system according to claim 12, wherein: The modification operation includes a structure adjustment operation for adjusting a downstream structure of a step currently being executed in the machine learning modeling process, wherein, in response to the structure adjustment operation, while the execution device executes the at least one step, the construction device adjusts the structure of the machine learning modeling process. Among them, when the running device runs to the part of the machine learning modeling process whose structure is adjusted, it runs according to the adjusted structure.

16. The system of claim 12, wherein: The modification operation includes a structure adjustment operation for adjusting an upstream structure of a step currently being executed in the machine learning modeling process, wherein, in response to the structure adjustment operation, the execution device stops executing the step currently being executed, and the construction device adjusts the structure of the machine learning modeling process. When a user operation for executing at least one step in the machine learning modeling process is received again, the executing device starts executing from the most upstream portion of the adjusted structure in the machine learning modeling process. Among them, when the running device runs to the part of the machine learning modeling process whose structure is adjusted, it runs according to the adjusted structure.

17. The system according to claim 15 or 16, wherein: The structural adjustment operation includes at least one of an operation for adding steps to the machine learning modeling process, an operation for deleting steps in the machine learning modeling process, and an operation for changing the logical relationship between steps in the machine learning modeling process.

18. The system of claim 12, wherein: The construction device receives a construction operation from the user to construct another machine learning modeling process in the graphical interface while the running device runs the at least one step, and in response to the construction operation, constructs the another machine learning modeling process while running the at least one step.

19. The system of claim 13, wherein: The construction device receives a user's selection operation on a node in the directed acyclic graph while the running device runs the at least one step; in response to the selection operation, displays to the user a control for configuring a configuration item of the step corresponding to the selected node; and receives the user's input operation on the displayed control.

20. The system of claim 13, wherein: The construction device receives a user's selection operation on a node in the directed acyclic graph while the running device runs the at least one step; in response to the selection operation, at least one control is displayed around the selected node, each of which is used to display at least one output element of the step corresponding to the node; and in response to the user's selection operation on one of the at least one control, the output result of the output element corresponding to the selected control is displayed to the user.

21. The system of claim 13, wherein: The construction device displays a node list in a predetermined area of ​​the graphical interface while the execution device executes the at least one step, and receives a user operation of selecting and dragging a node from the node list to connect to a node in the directed acyclic graph; And / or, while the running device is running the at least one step, the construction device recommends to the user the nodes and / or node combinations to which the node can be connected through the connection point in response to a user operation on a connection point of a node in the directed acyclic graph; and receives the user's operation of selecting a node or a node combination from the recommended nodes and / or node combinations to connect to the connection point.

22. The system of claim 18, wherein: The execution device executes the at least one step in the other machine learning modeling process simultaneously with the at least one step in the machine learning modeling process in response to a user operation for executing the at least one step in the other machine learning modeling process; The construction device receives a modification operation from a user for modifying the other machine learning modeling process while the execution device executes the at least one step in the other machine learning modeling process, and modifies the other machine learning modeling process in response to the modification operation. In which, when the running device runs to the modified part of the other machine learning modeling process, it runs based on the modified machine learning modeling process.

23. A computer-readable medium, wherein: Recorded on the computer-readable medium is a computer program for executing the method for constructing a machine learning modeling process as described in any one of claims 1 to 11.

24. A computing device comprising a storage component and a processor, wherein: The storage component stores a set of computer-executable instructions. When the set of computer-executable instructions is executed by the processor, the method for constructing a machine learning modeling process as described in any one of claims 1 to 11 is executed.

Images