Display processing method and device of artificial intelligence model, equipment and medium

Abstract

The invention relates to a display processing method and device of an artificial intelligence model, equipment and a medium, and the method comprises the steps: obtaining the model structure information of the artificial intelligence model; the model structure information comprises a plurality of layer class elements and connection relations among the layer class elements; generating a spiral line model according to the model structure information; the spiral line model comprises a plurality of nodes and edges connected with the nodes, the plurality of nodes are in one-to-one correspondence with the plurality of layer elements, and the edges are used for indicating the connection relationship; and displaying the spiral line model in a view. According to the technical scheme, the visual display effect of the artificial intelligence model is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of visualization display technology, and in particular to a display processing method, device, equipment and medium for an artificial intelligence model. Background Art

[0002] With the continuous development of artificial intelligence technology, artificial intelligence models are widely used in various fields. The deep learning framework is a collection of tools for building, training and deploying deep learning networks. People can build deep learning networks through code. Complex deep learning networks are composed of multiple layers of networks, and each layer of the network is composed of many parameters. As the number of layers increases, it becomes difficult to build deep learning networks through code, and it is difficult to maintain and adjust. By visualizing the model structure of the deep learning network, users can understand the complex model architecture and enhance the model's interpretability and maintainability.

[0003] Currently, there are two ways to visualize model structures: VisualDL and Tensorboard. The existing methods visualize the model in a horizontal and vertical tiled manner. For increasingly complex model structures, the tiled visualization effect is poor. Summary of the invention

[0004] In order to solve the above technical problems, the present disclosure provides a display processing method, device, equipment and medium for an artificial intelligence model.

[0005] In a first aspect, an embodiment of the present disclosure provides a display processing method of an artificial intelligence model, including:

[0006] Acquire model structure information of the artificial intelligence model; the model structure information includes a plurality of layer elements and connection relationships between the layer elements;

[0007] Generate a spiral line model according to the model structure information; the spiral line model includes a plurality of nodes and edges connecting the nodes, the plurality of nodes correspond to the plurality of layer class elements one by one, and the edges are used to indicate the connection relationship;

[0008] The helix model is displayed in a view.

[0009] In a second aspect, an embodiment of the present disclosure provides a display processing device for an artificial intelligence model, comprising:

[0010] An acquisition module, used to acquire model structure information of an artificial intelligence model; the model structure information includes a plurality of layer elements and connection relationships between the layer elements;

[0011] A generating module, used for generating a spiral line model according to the model structure information; the spiral line model comprises a plurality of nodes and edges connecting the nodes, the plurality of nodes correspond one-to-one to the plurality of layer class elements, and the edges are used for indicating the connection relationship;

[0012] A display module is used to display the helical line model in a view.

[0013] In a third aspect, an embodiment of the present disclosure provides an electronic device, comprising: a processor; a memory for storing executable instructions of the processor; the processor is used to read the executable instructions from the memory and execute the instructions to implement the display processing method of the artificial intelligence model described in the first aspect above.

[0014] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the display processing method of the artificial intelligence model described in the first aspect above is implemented.

[0015] Compared with the prior art, the technical solution provided by the disclosed embodiment has the following advantages: a spiral line model is generated according to the model structure information of the artificial intelligence model, so that the spiral line model includes multiple nodes and edges connecting the nodes, the multiple nodes correspond one-to-one to multiple layer-type elements of the artificial intelligence model, the edges are used to indicate the connection relationship between the layer-type elements, and the spiral line model is displayed in the view. Thus, the complete artificial intelligence model structure can be displayed in a limited display area without losing the key information of the model structure, which solves the problem of poor visualization display effect of complex artificial intelligence models in the tiled manner in the related art and improves the visualization display effect of the artificial intelligence model. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0018] Figure 1 A flowchart of a display processing method of an artificial intelligence model provided by an embodiment of the present disclosure;

[0019] Figure 2 A schematic diagram of converting a two-dimensional spiral line model into a three-dimensional spiral line model provided by an embodiment of the present disclosure;

[0020] Figure 3 A schematic diagram of generating a spiral line model provided by an embodiment of the present disclosure;

[0021] Figure 4 A schematic diagram showing an artificial intelligence model;

[0022] Figure 5 A schematic diagram showing an artificial intelligence model provided by an embodiment of the present disclosure;

[0023] Figure 6 A schematic diagram of a display interface provided by an embodiment of the present disclosure;

[0024] Figure 7 A schematic diagram of a spiral line model expansion design provided in an embodiment of the present disclosure;

[0025] Figure 8 A schematic diagram of the structure of a display processing device for an artificial intelligence model provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0026] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0027] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0028] Figure 1 This is a flow chart of a display processing method for an artificial intelligence model provided in an embodiment of the present disclosure. The method provided in an embodiment of the present disclosure can be executed by a display processing device for an artificial intelligence model, which can be implemented in software and / or hardware and can be integrated on any electronic device with computing capabilities.

[0029] like Figure 1 As shown, the display processing method of the artificial intelligence model provided by the embodiment of the present disclosure may include:

[0030] Step 101, obtaining model structure information of the artificial intelligence model.

[0031] The method of the embodiment of the present disclosure is used for visual display of artificial intelligence models, which include machine learning models and deep learning models.

[0032] In this embodiment, the model structure information of the artificial intelligence model includes multiple layer class elements and the connection relationship between the layer class elements, wherein the layer class elements include layer elements and class elements, and the layer elements include but are not limited to convolutional layers, normalization layers, etc. Layer elements are, for example, Layer in the deep learning framework Keras, and class elements are, for example, Nodes in the deep learning framework TensorFlow. The connection relationship is directional. For example, if the output of layer class element one is the input of layer class element two, there is a connection relationship from layer class element one to layer class element two.

[0033] Step 102: Generate a helical line model according to the model structure information.

[0034] In this embodiment, the spiral line model includes multiple nodes and edges connecting the nodes. The multiple nodes correspond to multiple layer elements one by one. The edges are used to indicate the connection relationship. The spiral line model includes a two-dimensional spiral line model and a three-dimensional spiral line model. For example Figure 2 As shown, Figure 2 The left side (top view / plan view) shows a two-dimensional helix model, Figure 2 The right side (side view / elevation view) shows a three-dimensional spiral model, which adopts any one of the Fermat spiral, equiangular spiral, hyperbolic spiral, Archimedean spiral, Conue spiral, and Euler spiral.

[0035] Optionally, generating a two-dimensional spiral model includes: generating a spiral according to an expression of a preset spiral, connecting each layer-class element in the artificial intelligence model through the spiral, wherein for all N layer-class elements of the artificial intelligence model, N nodes are generated correspondingly, and the N layer-class elements correspond to the N nodes one by one, and then, in the order from input to output of the artificial intelligence model, the position of each node on the spiral is determined according to the connection relationship between each layer-class element and the number of nodes N, so as to add N nodes to the spiral. For example, the center of the spiral pattern can be used as the starting point of the ray to determine multiple rays, and the angle between each two adjacent rays is the same, and a node is added to each intersection position of each ray with the spiral, thereby determining the position of each node to improve the display effect of the view. After adding N nodes to the spiral, the spiral portion between two adjacent nodes is used as an edge, and the direction of each edge is determined according to the connection relationship between the layer-class elements to generate a directed graph. Furthermore, in practical applications, the output of a layer-like element may be used as the input of multiple other layer-like elements, and the input of a layer-like element may also come from the output of multiple other layer-like elements. In the case where a layer-like element has multiple inputs and / or multiple outputs, on the basis of the spiral, directed edges between nodes are further generated according to the connection relationship between the layer-like elements. For example, a schematic diagram of generating a two-dimensional spiral model is shown as follows: Figure 3 shown.

[0036] Optionally, generating a three-dimensional spiral model includes: generating a two-dimensional spiral model according to the model structure information, stretching the two-dimensional spiral model in a direction perpendicular to the plane where the two-dimensional spiral model is located to convert the two-dimensional spiral model into a three-dimensional spiral model. For example, a schematic diagram of converting the two-dimensional spiral model into the three-dimensional spiral model is as follows: Figure 2 shown.

[0037] Step 103: Display the helix model in the view.

[0038] In this embodiment, displaying a spiral model in a view includes: displaying a two-dimensional spiral model in the view; and / or, in response to a switching request, converting the two-dimensional spiral model into a three-dimensional spiral model, and displaying the three-dimensional spiral model in the view.

[0039] When visualizing an artificial intelligence model, the related technology uses a linear tiling method from top to bottom and from left to right to achieve visualization. For example, Figure 4 As shown, for complex artificial intelligence models, it is impossible to fully visualize the artificial intelligence models in the display interface. Users need to drag the scroll bar or zoom out to view the complete model. Zooming out the view will lead to unclearness, loss of valid information and other problems, and the display readability and interpretability are poor. In this embodiment, by generating a spiral line model, the spiral line model includes multiple nodes and edges connecting the nodes. The multiple nodes correspond one-to-one to multiple layer elements of the artificial intelligence model, and the edges are used to indicate the connection relationship between the layer elements, such as Figure 5 As shown, the complete artificial intelligence model structure can be displayed within a limited canvas and display screen without losing key information of the model structure, thereby improving the visualization display effect of the artificial intelligence model.

[0040] According to the technical solution of the embodiment of the present disclosure, a spiral model is generated according to the model structure information of the artificial intelligence model, so that the spiral model includes multiple nodes and edges connecting the nodes, the multiple nodes correspond one-to-one to multiple layer-type elements of the artificial intelligence model, the edges are used to indicate the connection relationship between the layer-type elements, and the spiral model is displayed in the view. In this way, the complete artificial intelligence model structure can be displayed in a limited display area without losing the key information of the model structure, which solves the problem of poor visualization display effect of complex artificial intelligence models in the tiling method in the related art and improves the visualization display effect of the artificial intelligence model.

[0041] Based on the above embodiments, in order to further improve the visualization effect of the artificial intelligence model, the embodiments of the present disclosure also provide a variety of interactive operation methods to meet the user needs of the artificial intelligence model visualization display scenario. The interactive operation is explained below.

[0042] In one embodiment of the present disclosure, after the spiral model is displayed in the view, if the three-dimensional spiral model is currently displayed, in response to a first interactive operation in the view, the viewing angle parameters of the current camera of the three-dimensional spiral model are obtained, and then the viewing angle parameters are adjusted according to the first interactive operation to display the three-dimensional spiral model according to the adjusted viewing angle parameters. The viewing angle parameters are used to indicate the orientation of the three-dimensional spiral model by simulating the camera setting to determine the orientation, scale, etc. of the displayed spiral model.

[0043] As an example, click in the view with the left mouse button, and while holding down the left mouse button, adjust the viewing angle parameters in real time by moving the mouse, and display the three-dimensional spiral model according to the adjusted viewing angle parameters. For example, while holding down the left mouse button and moving the mouse to the right, the three-dimensional spiral model is adjusted from displaying the front side to displaying the left side.

[0044] In one embodiment of the present disclosure, after the spiral model is displayed in the view, in response to a second interactive operation in the view, the scaling of the spiral model is adjusted, and the spiral model is displayed according to the adjusted scaling.

[0045] As an example, by adjusting the mouse wheel, the zoom ratio of the spiral model is adjusted accordingly. When the mouse wheel is detected to be rolled in a first direction, the view of the spiral model is zoomed in. When the mouse wheel is detected to be rolled in a second direction, the view of the spiral model is zoomed out.

[0046] The display interface is described below.

[0047] In one embodiment of the present disclosure, the view is composed of a UI area, a layer class name area, a node name area, a canvas area, a canvas operation area, a node detail information area, and an eagle eye area. Figure 6 As shown, Figure 6 Mark 1 is the UI area, mark 2 is the layer class name area, mark 3 is the node name area, mark 4 is the canvas area, mark 5 is the canvas operation area, mark 6 is the node details information area, and mark 7 is the eagle eye area.

[0048] The UI area is used to display the native UI (User Interface) of the software. For example, if the view is implemented based on the user interface of JupiterLab (a Notebook application), the UI area is used for the native UI of JupiterLab.

[0049] The layer class name area is used to display the layer class name. For example, taking the deep learning framework Keras as an example, in the Keras framework, nodes are composed of class elements and layer elements. The layer class name area is used to display the layer class name of Keras. The layer class name is used to distinguish the categories of layer class elements. Optionally, for different categories of layer class elements, nodes can be distinguished by color in the canvas area.

[0050] The node name area is used to display the node name. For example, nodes belonging to the same layer class name may have different node names, such as "input" and "dropout". Optionally, in response to a drag operation on an icon in the node name area, the icon can be dragged onto the spiral model to add a corresponding node to the spiral model, or the icon can be dragged onto any node in the spiral model to replace the node, for example Figure 5 As shown, by dragging the circular icon in the node name area on the left side of the display interface, you can drag the circular icon to any node in the canvas area. At this time, a prompt message pops up to confirm whether to replace the node. When the confirmation message is received, the node in the canvas area is replaced by the node corresponding to the circular icon.

[0051] The canvas area is used to display the spiral model. As an example, the first letter of the node name and the Keras-specific layer name can be displayed on each node in the canvas area.

[0052] The canvas operation area is used to display operation controls, and the operation controls are used to adjust the display of the spiral model. The operation controls include but are not limited to displaying zoom-in controls, zoom-out controls, and switch controls.

[0053] As an example, when a trigger operation on a zoom-in control is detected, the view of the spiral model displayed in the canvas area is zoomed in; when a trigger operation on a zoom-out control is detected, the view of the spiral model displayed in the canvas area is zoomed out.

[0054] As another example, when a trigger operation on a switching control is detected, the two-dimensional spiral model is converted into a three-dimensional spiral model, or the three-dimensional spiral model is converted into a two-dimensional spiral model.

[0055] The node details area is used to display node parameter details information of the currently selected node. Optionally, in response to an edit operation on the node details area, the node parameter details information of the currently selected node is edited, wherein the edit operation includes adding, deleting, and modifying.

[0056] The Eagle Eye area is used to display an eagle eye view of the helix model.

[0057] In one embodiment of the present disclosure, displaying a spiral line model in a view includes: determining multiple target nodes belonging to the same group from multiple nodes according to preset grouping information; and grouping and displaying target nodes of different groups. Among them, several nodes can be divided into a group according to actual needs, and target nodes belonging to the same group can be marked and distinguished by a preset display method, for example Figure 7 As shown, Figure 7 The left side shows two nodes belonging to the same group, and the two nodes can be marked and distinguished by a dotted bounding box.

[0058] In one embodiment of the present disclosure, a spiral line model is displayed in a view, including: obtaining association information of each layer element, generating a nested structure according to the association information corresponding to the node for the node with association information, and then displaying the association information corresponding to the node through the nested structure. For example, a layer element has multiple outputs, where output one is used as the input of another layer element, and output two and output three are one of the outputs of the artificial intelligence model, then the association information includes output two and output three, and a nested structure is generated according to output two and output three. The nested structure can be implemented in a tree-like branch shape, and each branch corresponds to a piece of association information, for example Figure 7 Shown on the right.

[0059] There are many conditions for triggering the display of nested structures. For example, in response to a trigger operation on a certain node, if the node has corresponding associated information, a nested structure is generated according to the associated information corresponding to the node, and the nested structure is displayed to display the corresponding associated information through the nested structure. Figure 7 As shown in the figure for a single node on the right. For another example, Figure 7 The two target nodes on the left belong to the same group. In response to the trigger operation for the group, after the trigger, Figure 7 As shown on the right, the target nodes of the group are expanded and displayed. When expanded and displayed, the associated information of each layer class element is obtained. For target nodes with associated information, a nested structure is generated according to the associated information corresponding to the target nodes, and the associated information corresponding to the target nodes is displayed through the nested structure.

[0060] Figure 8 A schematic diagram of a display processing device for an artificial intelligence model provided by an embodiment of the present disclosure is shown in FIG. Figure 8 As shown, the display processing device of the artificial intelligence model includes: an acquisition module 81, a generation module 82, and a display module 83.

[0061] The acquisition module 81 is used to acquire the model structure information of the artificial intelligence model; the model structure information includes a plurality of layer elements and the connection relationship between the layer elements;

[0062] A generating module 82, configured to generate a spiral line model according to the model structure information; the spiral line model comprises a plurality of nodes and edges connecting the nodes, the plurality of nodes correspond one-to-one to the plurality of layer class elements, and the edges are used to indicate the connection relationship;

[0063] The display module 83 is used to display the helical line model in a view.

[0064] In one embodiment of the present disclosure, the display module 83 is specifically used for:

[0065] displaying a two-dimensional helix model in said view; and / or,

[0066] In response to a switching request, converting the two-dimensional helical line model into a three-dimensional helical line model;

[0067] The three-dimensional helical model is displayed in the view.

[0068] In one embodiment of the present disclosure, the device further includes:

[0069] A viewing angle adjustment module, configured to obtain viewing angle parameters of a current camera of the three-dimensional helical model in response to a first interactive operation in the view;

[0070] The viewing angle parameter is adjusted according to the first interactive operation to display the three-dimensional spiral model according to the adjusted viewing angle parameter.

[0071] In one embodiment of the present disclosure, the device further includes:

[0072] a scale adjustment module, configured to adjust a scale of the spiral model in response to a second interactive operation in the view;

[0073] The helical model is displayed at the adjusted zoom ratio.

[0074] In one embodiment of the present disclosure, the view is composed of a UI area, a layer class name area, a node name area, a canvas area, a canvas operation area, a node detail information area, and an eagle eye area;

[0075] Among them, the UI area is used to display the native UI of the software;

[0076] The layer class name area is used to display the layer class name;

[0077] The node name area is used to display the node name;

[0078] The canvas area is used to display the spiral model;

[0079] The canvas operation area is used to display a zoom-in control, a zoom-out control, and a switch control;

[0080] The node details area is used to display node parameter details information of the currently selected node;

[0081] The eagle-eye area is used to display the eagle-eye view of the spiral model;

[0082] The proportion adjustment module is specifically used for: when a trigger operation on the zoom-in control is detected, performing a view zoom-in process on the spiral line model displayed in the canvas area; when a trigger operation on the zoom-out control is detected, performing a view zoom-out process on the spiral line model displayed in the canvas area;

[0083] The display module 83 is specifically used for: when a trigger operation on the switching control is detected, converting the two-dimensional spiral line model into a three-dimensional spiral line model;

[0084] The device also includes:

[0085] The information editing module is used to edit the node parameter detail information of the currently selected node in response to an editing operation on the node detail area.

[0086] In one embodiment of the present disclosure, the display module 83 is specifically used for:

[0087] According to the preset grouping information, determining a plurality of target nodes belonging to the same group among the plurality of nodes;

[0088] Displaying target nodes of different groups in groups; and / or,

[0089] Acquire association information of each layer class element, and for a target node with the association information, generate a nested structure according to the association information corresponding to the target node;

[0090] The associated information corresponding to the target node is displayed through the nested structure.

[0091] In one embodiment of the present disclosure, the spiral model adopts any one of a Fermat spiral, an equiangular spiral, a hyperbolic spiral, an Archimedean spiral, a Connolly spiral, and an Euler spiral.

[0092] The display processing device of the artificial intelligence model provided in the embodiment of the present disclosure can execute the display processing method of any artificial intelligence model provided in the embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method. For the contents not described in detail in the embodiment of the device of the present disclosure, reference can be made to the description in any method embodiment of the present disclosure.

[0093] The present disclosure also provides an electronic device, which includes one or more processors and a memory. The processor may be a central processing unit (CPU) or other forms of processing units with data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions. The memory may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on a computer-readable storage medium, and the processor may run the program instructions to implement the method of the above embodiment of the present disclosure and / or other desired functions. Various contents such as input signals, signal components, noise components, etc. may also be stored in the computer-readable storage medium.

[0094] In one example, the electronic device may further include: an input device and an output device, and these components are interconnected through a bus system and / or other forms of connection mechanisms. In addition, the input device may also include, for example, a keyboard, a mouse, etc. The output device may output various information to the outside, including determined distance information, direction information, etc. The output device may include, for example, a display, a speaker, a printer, and a communication network and a remote output device connected thereto, etc. In addition, according to the specific application, the electronic device may also include any other appropriate components such as a bus, an input / output interface, etc.

[0095] In addition to the above methods and devices, the embodiments of the present disclosure may also be a computer program product, which includes computer program instructions. When the computer program instructions are executed by a processor, the processor executes any method provided by the embodiments of the present disclosure.

[0096] The computer program product may be written in any combination of one or more programming languages ​​to write program code for performing the operations of the disclosed embodiments, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0097] In addition, the embodiments of the present disclosure may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes any method provided by the embodiments of the present disclosure.

[0098] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0099] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0100] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A display processing method for an artificial intelligence model, characterized in that: include: Acquire model structure information of the artificial intelligence model; the model structure information includes a plurality of layer elements and connection relationships between the layer elements; Generate a spiral line model according to the model structure information; the spiral line model includes a plurality of nodes and edges connecting the nodes, the plurality of nodes correspond to the plurality of layer class elements one by one, and the edges are used to indicate the connection relationship; The helix model is displayed in a view.

2. The method according to claim 1, characterized in that The step of displaying the spiral line model in a view comprises: displaying a two-dimensional helix model in said view; and / or, In response to a switching request, converting the two-dimensional helical line model into a three-dimensional helical line model; The three-dimensional helical model is displayed in the view.

3. The method according to claim 2, characterized in that After displaying the three-dimensional helical model in the view, the method further comprises: In response to a first interactive operation in the view, obtaining a viewing angle parameter of a current camera of the three-dimensional helical model; The viewing angle parameter is adjusted according to the first interactive operation to display the three-dimensional spiral model according to the adjusted viewing angle parameter.

4. The method according to claim 3, characterized in that After displaying the spiral line model in the view, the method further comprises: In response to a second interactive operation in the view, adjusting a zoom ratio of the spiral model; The helical model is displayed at the adjusted zoom ratio.

5. The method according to claim 4, characterized in that The view consists of a UI area, a layer class name area, a node name area, a canvas area, a canvas operation area, a node detail information area, and an eagle eye area; Among them, the UI area is used to display the native UI of the software; The layer class name area is used to display the layer class name; The node name area is used to display the node name; The canvas area is used to display the spiral model; The canvas operation area is used to display a zoom-in control, a zoom-out control, and a switch control; The node details area is used to display node parameter details information of the currently selected node; The eagle-eye area is used to display the eagle-eye view of the spiral model; In response to the second interactive operation in the view, adjusting the scaling of the spiral model comprises: When a trigger operation on the zoom control is detected, a view zooming process is performed on the spiral line model displayed in the canvas area; When a trigger operation on the zoom-out control is detected, a view zoom-out process is performed on the spiral line model displayed in the canvas area; The step of converting the two-dimensional helical line model into a three-dimensional helical line model in response to the switching request comprises: When a trigger operation on the switching control is detected, converting the two-dimensional spiral line model into a three-dimensional spiral line model; The method further comprises: In response to the editing operation on the node detail area, the node parameter detail information of the currently selected node is edited.

6. The method according to claim 1, characterized in that The step of displaying the spiral line model in a view comprises: According to the preset grouping information, determining a plurality of target nodes belonging to the same group among the plurality of nodes; Displaying target nodes of different groups in groups; and / or, Acquire association information of each layer class element, and for a target node with the association information, generate a nested structure according to the association information corresponding to the target node; The associated information corresponding to the target node is displayed through the nested structure.

7. The method according to claim 1, characterized in that The spiral model adopts any one of Fermat spiral, equiangular spiral, hyperbolic spiral, Archimedean spiral, Connolly spiral and Euler spiral.

8. A display processing device for an artificial intelligence model, characterized in that: include: An acquisition module, used to obtain model structure information of an artificial intelligence model; The model structure information includes a plurality of layer-type elements and connection relationships between the layer-type elements; A generating module, used for generating a spiral line model according to the model structure information; the spiral line model comprises a plurality of nodes and edges connecting the nodes, the plurality of nodes correspond to the plurality of layer class elements one by one, and the edges are used for indicating the connection relationship; The display module is used to display the helical line model in a view.

9. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is used to read the executable instructions from the memory and execute the instructions to implement the display processing method of the artificial intelligence model described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the display processing method of the artificial intelligence model described in any one of claims 1 to 7 is implemented.

Images