Webpage generation method and device based on large model, equipment and storage medium

Abstract

The invention relates to the technical field of computers, is applied to the field of financial business science and technology, and discloses a webpage generation method, device and equipment based on a large model and a storage medium. The method comprises the steps that by designing the large model, initial design elements and initial design attributes, a target interaction mode and a target component are determined; and generating a target webpage according to a dynamic optimization mechanism, the target interaction mode and the target component. By means of the mode, manual cooperation between a designer and a developer is reduced through automatic design draft analysis and code generation, the requirement for later manual debugging is reduced, and the initial quality of codes is improved. The pre-trained large design model can understand and generate codes conforming to financial industry design specifications, and the standardization degree is improved. Accurate implementation of the design intention is ensured through an accurate analysis algorithm and large model training, and therefore the efficiency of generating the webpage according to the actual demand of the user in the financial field is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology and is applied to the field of financial business technology, and in particular to a web page generation method, device, equipment and storage medium based on a large model. Background Art

[0002] In the financial sector, front-end development is a key link, which involves converting the designer's creativity into actual code for the user interface. However, current practices often rely on manual collaboration between designers and developers, or use some basic automation tools to assist in the conversion of design drafts to code. There are some significant problems with this approach: the manual collaboration process is inefficient, designers and developers need to communicate frequently to ensure that the design intent is accurately implemented, and manual coding further prolongs the project cycle. This inefficient manual process not only increases the time cost, but may also lead to poor communication and affect the quality of the final product; the existing automation tools have limited standardization, and they can usually only handle some basic layouts and elements. When faced with the complexity and special requirements of financial web pages, these tools are often unable to cope with it. Financial web pages usually need to deal with complex data display, real-time updates, security requirements, etc., which are not what general automation tools are good at; the code quality varies, and due to the lack of intelligent optimization mechanisms, the code generated by automation tools often requires a lot of post-manual debugging. This not only affects the development efficiency, but also the quality of the final product. Therefore, how to improve the efficiency of generating web pages according to the actual needs of users in the financial field has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present application provides a web page generation method, apparatus, device and storage medium based on a large model to improve the efficiency of generating web pages according to actual user needs in the financial field.

[0004] In a first aspect, the present application provides a method for generating a web page based on a large model, the method comprising:

[0005] Determining initial design elements and initial design attributes of the initial document based on a preset document parsing algorithm and the initial document;

[0006] Determine a target interaction mode and at least one target component through a pre-trained design macro model, the initial design elements, and the initial design attributes;

[0007] A target webpage is generated according to a preset dynamic optimization mechanism, the target interaction mode and each of the target components.

[0008] In a second aspect, the present application further provides a webpage generation device based on a large model, the device comprising:

[0009] An element and attribute determination module, used to determine the initial design elements and initial design attributes of the initial document based on a preset document parsing algorithm and the initial document;

[0010] An interaction mode and component determination module, configured to determine a target interaction mode and at least one target component through a pre-trained design macro model, the initial design elements and the initial design attributes;

[0011] The target webpage generation module is used to generate a target webpage according to a preset dynamic optimization mechanism, the target interaction mode and each of the target components.

[0012] In a third aspect, the present application also provides a computer device, comprising a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and implement the large model-based web page generation method as described above when executing the computer program.

[0013] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the large model-based web page generation method as described above.

[0014] The present application discloses a web page generation method, device, equipment and storage medium based on a large model, wherein the web page generation method based on the large model includes determining the initial design elements and initial design attributes of the initial manuscript based on a preset manuscript parsing algorithm and an initial manuscript; determining the target interaction mode and at least one target component through a pre-trained design large model, the initial design elements and the initial design attributes; generating a target web page according to a preset dynamic optimization mechanism, the target interaction mode and each of the target components. In the above manner, the present application reduces the manual collaboration between designers and developers through automated design draft parsing and code generation, reduces the need for later manual debugging, and improves the initial quality of the code. The pre-trained design large model can understand and generate code that complies with the design specifications of the financial industry, thereby improving the degree of standardization. Accurate parsing algorithms and large model training ensure the accurate realization of design intent, thereby improving the efficiency of generating web pages in the financial field according to actual user needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1is a schematic flow chart of a web page generation method based on a large model provided in the first embodiment of the present application;

[0017] Figure 2 is a schematic flow chart of a web page generation method based on a large model provided in the second embodiment of the present application;

[0018] Figure 3 is a schematic flow chart of a web page generation method based on a large model provided in the third embodiment of the present application;

[0019] Figure 4 A schematic block diagram of a web page generation device based on a large model provided in an embodiment of the present application;

[0020] Figure 5 A schematic block diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0022] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0023] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0024] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0025] The embodiments of the present application provide a web page generation method, device, equipment and storage medium based on a large model. Among them, the web page generation method based on a large model can be applied to a server, and the manual collaboration between designers and developers is reduced through automated design draft parsing and code generation, the need for later manual debugging is reduced, and the initial quality of the code is improved. The pre-trained design large model can understand and generate code that meets the design specifications of the financial industry, thereby improving the degree of standardization. Accurate parsing algorithms and large model training ensure the accurate implementation of design intent, thereby improving the efficiency of generating web pages in the financial field according to actual user needs. Among them, the server can be an independent server or a server cluster.

[0026] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0027] See also Figure 1 , Figure 1 This is a schematic flow chart of a web page generation method based on a large model provided by the first embodiment of the present application. The web page generation method based on the large model can be applied to a server to reduce the manual collaboration between designers and developers through automated design draft parsing and code generation, reduce the need for later manual debugging, and improve the initial quality of the code. The pre-trained design large model can understand and generate code that meets the design specifications of the financial industry, thereby improving the degree of standardization. Accurate parsing algorithms and large model training ensure the accurate realization of design intent, thereby improving the efficiency of generating web pages in the financial field according to actual user needs.

[0028] like Figure 1 As shown, the web page generation method based on the large model specifically includes steps S10 to S30.

[0029] Step S10: determining initial design elements and initial design attributes of the initial document based on a preset document parsing algorithm and the initial document;

[0030] Specifically, it identifies design elements such as images, texts, tables, etc. in documents, and extracts the attributes of each design element, such as font, color, layout, etc.

[0031] Step S20, determining a target interaction mode and at least one target component through the pre-trained design macro model, the initial design elements and the initial design attributes;

[0032] Specifically, select a pre-trained design model that can understand design elements and attributes, input the extracted design elements and attributes into the model, analyze the user's possible interaction methods through the model, and determine the target interaction mode based on the analysis results. According to the target interaction mode, identify the necessary UI components, such as buttons, sliders, input boxes, etc., and ensure that at least one target component is determined to achieve the interactive function.

[0033] Step S30: Generate a target webpage according to a preset dynamic optimization mechanism, the target interaction mode and each of the target components.

[0034] Specifically, a preset dynamic optimization mechanism is applied to dynamically adjust interaction modes and components based on user behavior and feedback.

[0035] In one embodiment, the initial document is parsed to extract design elements and attributes therein, such as color, font, layout, etc. The extracted design elements and attributes are input into a pre-trained design model to obtain a target interaction mode and at least one target component, and a target web page is generated according to the target interaction mode and each target component in combination with a preset dynamic optimization mechanism.

[0036] The present embodiment discloses a web page generation method, apparatus, device and storage medium based on a large model, wherein the web page generation method based on the large model includes determining the initial design elements and initial design attributes of the initial document based on a preset document parsing algorithm and an initial document; determining the target interaction mode and at least one target component through a pre-trained design large model, the initial design elements and the initial design attributes; generating a target web page according to a preset dynamic optimization mechanism, the target interaction mode and each of the target components. In the above manner, the present application reduces the manual collaboration between designers and developers through automated design draft parsing and code generation, reduces the need for later manual debugging, and improves the initial quality of the code. The pre-trained design large model can understand and generate code that complies with the design specifications of the financial industry, thereby improving the degree of standardization. Accurate parsing algorithms and large model training ensure the accurate realization of design intent, thereby improving the efficiency of generating web pages in the financial field according to actual user needs.

[0037] See also Figure 2 , Figure 2This is a schematic flow chart of a web page generation method based on a large model provided in the second embodiment of the present application. The web page generation method based on the large model can be applied to a server to reduce the manual collaboration between designers and developers through automated design draft parsing and code generation, reduce the need for later manual debugging, and improve the initial quality of the code. The pre-trained design large model can understand and generate code that meets the design specifications of the financial industry, thereby improving the degree of standardization. Accurate parsing algorithms and large model training ensure the accurate realization of design intent, thereby improving the efficiency of generating web pages in the financial field according to actual user needs.

[0038] based on Figure 1 The embodiment shown, this embodiment Figure 2 As shown, step S10 includes steps S101 to S103.

[0039] Step S101, performing a standardization processing operation on the initial image document to generate a standard image in a standard format, wherein the standardization processing operation includes a scaling operation, a cropping operation and / or a denoising operation;

[0040] Specifically, the initial image document is imported into the system, the image is resized to meet the standard size, unnecessary edges or areas in the image are removed, the core content is retained, a denoising algorithm is used to reduce noise in the image, and the scaled, cropped and denoised image is saved in a standard format.

[0041] Step S102: extracting image elements in the standard image by using the preset document parsing algorithm, wherein the image elements include button elements, text box elements and / or icon elements;

[0042] Specifically, buttons in the image are identified and their attributes such as shape, color, and position are extracted; text boxes in the image are identified and their attributes such as font, size, color, and alignment are extracted; icons in the image are identified and their attributes such as shape, color, and size are extracted. Based on the extracted attributes, the elements are classified as buttons, text boxes, or icons.

[0043] Step S103: classify the standard image according to the button element, the text box element and / or the icon element to determine the initial design element and the initial design attribute.

[0044] In one embodiment, the initial image document is subjected to standardization processing operations, such as scaling, cropping, denoising, etc., to generate a standard image in a standard format. Image elements in the standard image, such as button elements, text box elements, icon elements, etc., are extracted through a preset document parsing algorithm. Based on the extracted image elements, the standard image is classified to determine the initial design elements and initial design attributes.

[0045] The present embodiment discloses a webpage generation method, device, equipment and storage medium based on a large model, wherein the webpage generation method based on the large model includes performing a standard processing operation on the initial image document to generate a standard image in a standard format, wherein the standard processing operation includes a scaling operation, a cropping operation and / or a denoising operation; extracting image elements in the standard image through the preset document parsing algorithm, wherein the image elements include button elements, text box elements and / or icon elements; classifying the standard image according to the button elements, the text box elements and / or the icon elements to determine the initial design elements and the initial design attributes; determining the target interaction mode and at least one target component through the pre-trained design large model, the initial design elements and the initial design attributes; generating a target webpage according to the preset dynamic optimization mechanism, the target interaction mode and each of the target components. In the above manner, the present application reduces the manual collaboration between designers and developers through automated design draft parsing and code generation, reduces the need for later manual debugging, and improves the initial quality of the code. The pre-trained design large model can understand and generate code that meets the design specifications of the financial industry, thereby improving the degree of standardization. Precise parsing algorithms and large model training ensure the accurate realization of design intent, thereby improving the efficiency of generating web pages based on actual user needs in the financial field.

[0046] See also Figure 3 , Figure 3 This is a schematic flow chart of a web page generation method based on a large model provided in the third embodiment of the present application. The web page generation method based on the large model can be applied to a server to reduce the manual collaboration between designers and developers through automated design draft parsing and code generation, reduce the need for later manual debugging, and improve the initial quality of the code. The pre-trained design large model can understand and generate code that meets the design specifications of the financial industry, thereby improving the degree of standardization. Accurate parsing algorithms and large model training ensure the accurate realization of design intent, thereby improving the efficiency of generating web pages in the financial field according to actual user needs.

[0047] based on Figure 1 The embodiment shown, this embodiment Figure 3 As shown, step S10 includes steps S104 to S107.

[0048] Step S104, extracting text information from the initial text document by using the preset document parsing algorithm;

[0049] Step S105, determining the RGB information of the text information according to the color channel information RGB extraction algorithm;

[0050] Step S106, determining the character spacing, character size and / or character layout of each character in the text information;

[0051] Step S107: determining the initial design elements and the initial design attributes according to the RGB information, the character spacing, the character size and / or the character layout.

[0052] In one embodiment, an initial text document is imported into the system, a preset algorithm suitable for text parsing is selected, the text content in the document is identified using the algorithm, the identified text content is extracted, the RGB color value of the text information is extracted according to the algorithm, and the RGB value of the extracted text color is recorded.

[0053] Analyze the spacing between characters in the text, analyze the size of each character in the text, analyze the arrangement and position of characters in the text, and record the layout information of the characters.

[0054] Based on the extracted text information, design elements such as titles, text, lists, etc. are identified, and the properties of each design element are extracted, including but not limited to RGB color value, character spacing, character size, and character layout.

[0055] In one embodiment, the text information in the initial text document, such as text content, font, font size, etc., is extracted by a preset document parsing algorithm, and the RGB information of the text information, such as the values ​​of the red, green, and blue channels, is determined according to the color channel information RGB extraction algorithm. The character spacing, character size, and / or character layout of each character in the text information are determined, such as the spacing between characters, the size of characters, the position of characters, etc. The initial design elements and initial design attributes, such as color, font, font size, etc., are determined according to the RGB information, character spacing, character size, and / or character layout.

[0056] The present embodiment discloses a webpage generation method, device, equipment and storage medium based on a large model, wherein the webpage generation method based on the large model includes extracting text information from the initial text document through the preset document parsing algorithm; determining the RGB information of the text information according to the color channel information RGB extraction algorithm; determining the character spacing, character size and / or character layout of each character in the text information; determining the initial design element and the initial design attribute according to the RGB information, the character spacing, the character size and / or the character layout; determining the target interaction mode and at least one target component through the pre-trained design large model, the initial design element and the initial design attribute; generating the target webpage according to the preset dynamic optimization mechanism, the target interaction mode and each target component. In the above manner, the present application reduces the manual collaboration between designers and developers through automated design draft parsing and code generation, reduces the need for later manual debugging, and improves the initial quality of the code. The pre-trained design large model can understand and generate code that meets the design specifications of the financial industry, thereby improving the degree of standardization. Accurate parsing algorithms and large model training ensure the accurate implementation of design intent, thereby improving the efficiency of generating webpages in the financial field according to actual user needs.

[0057] based on Figure 1 In the embodiment shown, in this embodiment, before step S20, the following steps are included:

[0058] Obtain a historical interaction mode training set and a historical component training set as a training set;

[0059] The training set is labeled by using the design macro model to obtain each historical design element and each historical design attribute in the training set;

[0060] Manual annotations of the training set are obtained, and the design macro model is pre-trained through the manual annotations, each of the historical design elements and each of the historical design attributes to obtain the pre-trained design macro model.

[0061] In one embodiment, a historical interaction mode training set and a historical component training set are obtained as training sets, such as historical webpage interaction modes, historical components, etc. The training set is annotated by the design macro model to obtain various historical design elements and various historical design attributes in the training set, such as color, font, font size, etc., and manual annotations of the training set are obtained, such as manually annotated color, font, font size, etc., and the design macro model is pre-trained by manual annotation, various historical design elements, and various historical design attributes to obtain a pre-trained design macro model.

[0062] The present embodiment discloses a web page generation method, device, equipment and storage medium based on a large model, wherein the web page generation method based on the large model includes determining the initial design elements and initial design attributes of the initial document based on a preset document parsing algorithm and an initial document; obtaining a historical interaction mode training set and a historical component training set as a training set; annotating the training set through the design large model to obtain each historical design element and each historical design attribute in the training set; obtaining manual annotations of the training set, and pre-training the design large model through the manual annotations, each historical design element and each historical design attribute to obtain the pre-trained design large model; determining a target interaction mode and at least one target component through the pre-trained design large model, the initial design element and the initial design attribute; generating a target web page according to a preset dynamic optimization mechanism, the target interaction mode and each target component. In the above manner, the present application reduces the manual collaboration between designers and developers through automated design draft parsing and code generation, reduces the need for later manual debugging, and improves the initial quality of the code. The pre-trained design large model can understand and generate code that meets the design specifications of the financial industry, thereby improving the degree of standardization. Precise parsing algorithms and large model training ensure the accurate realization of design intent, thereby improving the efficiency of generating web pages based on actual user needs in the financial field.

[0063] In a more preferred embodiment, step S20 includes:

[0064] Converting the initial design element and the initial design attribute into a design element annotation and a design attribute annotation respectively;

[0065] Matching the design element annotations and the design attribute annotations with a preset interaction mode-component library through the pre-trained design macro model;

[0066] An interaction mode that matches the design element annotation and the design attribute annotation is determined as the target interaction mode, and a component that matches the design element annotation and the design attribute annotation is determined as the target component.

[0067] In one embodiment, the initial design elements and initial design attributes are converted into design element annotations and design attribute annotations, such as converting colors, fonts, font sizes, etc. into annotations. The design element annotations and design attribute annotations are matched with a preset interaction mode-component library through a pre-trained design macro model, such as matching the annotations with preset interaction modes and components. The interaction mode that matches the design element annotations and design attribute annotations is determined as a target interaction mode, and the component that matches the design element annotations and design attribute annotations is determined as a target component.

[0068] The present embodiment discloses a web page generation method, device, equipment and storage medium based on a large model. The web page generation method based on the large model includes determining the initial design elements and initial design attributes of the initial document based on a preset document parsing algorithm and an initial document; obtaining a historical interaction mode training set and a historical component training set as a training set; annotating the training set through the design large model to obtain each historical design element and each historical design attribute in the training set; obtaining manual annotations of the training set, and pre-training the design large model through the manual annotations, each historical design element and each historical design attribute to obtain the pre-trained design large model; converting the initial design element and the initial design attribute into design element annotations and design attribute annotations respectively; matching the design element annotations and the design attribute annotations with a preset interaction mode-component library through the pre-trained design large model; determining the interaction mode matching the design element annotations and the design attribute annotations as the target interaction mode, and determining the component matching the design element annotations and the design attribute annotations as the target component; generating a target web page according to a preset dynamic optimization mechanism, the target interaction mode and each target component. Through the above methods, this application reduces the manual collaboration between designers and developers through automated design draft parsing and code generation, reduces the need for later manual debugging, and improves the initial quality of the code. The pre-trained design large model can understand and generate code that meets the design specifications of the financial industry, improving the degree of standardization. Accurate parsing algorithms and large model training ensure the accurate realization of design intent, thereby improving the efficiency of generating web pages in the financial field according to actual user needs.

[0069] based on Figure 1 In the embodiment shown, in this embodiment, step S30 includes:

[0070] Determine the current interaction demand through the preset dynamic optimization mechanism and the target interaction mode;

[0071] Based on the current interaction requirements, generate at least one set of front-end codes;

[0072] The target web page is generated based on each of the front-end codes and each of the target components.

[0073] In one embodiment, the current interaction demand is determined by a preset dynamic optimization mechanism and a target interaction mode, such as determining the current interaction demand based on factors such as user behavior and device type, and at least one set of front-end codes is generated based on the current interaction demand, such as generating corresponding front-end codes based on the current interaction demand. A target webpage is generated based on each front-end code and each target component, such as combining the front-end code and the target component to generate a target webpage.

[0074] The present embodiment discloses a web page generation method, device, equipment and storage medium based on a large model, wherein the web page generation method based on the large model includes determining the initial design elements and initial design attributes of the initial document based on a preset document parsing algorithm and an initial document; determining the target interaction mode and at least one target component through a pre-trained design large model, the initial design elements and the initial design attributes; determining the current interaction requirements through the preset dynamic optimization mechanism and the target interaction mode; generating at least one set of front-end codes based on the current interaction requirements; generating the target web page based on each of the front-end codes and each of the target components. In the above manner, the present application reduces the manual collaboration between designers and developers through automated design draft parsing and code generation, reduces the need for later manual debugging, and improves the initial quality of the code. The pre-trained design large model can understand and generate code that meets the design specifications of the financial industry, thereby improving the degree of standardization. Accurate parsing algorithms and large model training ensure the accurate realization of design intent, thereby improving the efficiency of generating web pages in the financial field according to actual user needs.

[0075] Based on any of the above embodiments, in this embodiment, the following steps are performed after step S30:

[0076] Get user feedback;

[0077] In response to the user feedback, iteratively training the pre-trained design macro model;

[0078] The target webpage is optimized by iteratively training the pre-trained design large model, and the optimized webpage is updated as the target webpage.

[0079] See also Figure 4 , Figure 4 The embodiment of the present application provides a schematic block diagram of a webpage generation device based on a large model, wherein the webpage generation device based on a large model is used to execute the aforementioned webpage generation method based on a large model. The webpage generation device based on a large model can be configured on a server.

[0080] like Figure 4 As shown, the webpage generation device 400 based on the large model includes:

[0081] The element and attribute determination module 410 is used to determine the initial design elements and initial design attributes of the initial document based on a preset document parsing algorithm and the initial document;

[0082] An interaction mode and component determination module 420, configured to determine a target interaction mode and at least one target component through a pre-trained design macro model, the initial design elements and the initial design attributes;

[0083] The target webpage generation module 430 is used to generate a target webpage according to a preset dynamic optimization mechanism, the target interaction mode and each of the target components.

[0084] Furthermore, the element and attribute determination module 410 includes:

[0085] A standard image generating unit, configured to perform a standard processing operation on the initial image document to generate a standard image in a standard format, wherein the standard processing operation includes a scaling operation, a cropping operation and / or a denoising operation;

[0086] An image element extraction unit, configured to extract image elements in the standard image by using the preset document parsing algorithm, wherein the image elements include button elements, text box elements and / or icon elements;

[0087] The element and attribute determination unit is used to classify the standard image according to the button element, the text box element and / or the icon element to determine the initial design element and the initial design attribute.

[0088] Furthermore, the element and attribute determination module 410 includes:

[0089] A text information extraction unit, used to extract text information from the initial text manuscript by using the preset manuscript parsing algorithm;

[0090] An RGB information determination unit, used to determine the RGB information of the text information according to a color channel information RGB extraction algorithm;

[0091] A character information determination unit, used to determine the character spacing, character size and / or character layout of each character in the text information;

[0092] The element and attribute determination unit is used to determine the initial design element and the initial design attribute according to the RGB information, the character spacing, the character size and / or the character layout.

[0093] Furthermore, the webpage generation device 400 based on the large model includes:

[0094] A training set acquisition module, used for acquiring a historical interaction mode training set and a historical component training set as a training set;

[0095] A historical data generation module, used to annotate the training set through the design macro model to obtain each historical design element and each historical design attribute in the training set;

[0096] The design big model generation module is used to obtain the manual annotations of the training set, and pre-train the design big model through the manual annotations, each of the historical design elements and each of the historical design attributes to obtain the pre-trained design big model.

[0097] Furthermore, the interaction mode and component determination module 420 includes:

[0098] An annotation conversion unit, used to convert the initial design element and the initial design attribute into a design element annotation and a design attribute annotation respectively;

[0099] A matching unit, used for matching the design element annotation and the design attribute annotation with a preset interaction mode-component library through the pre-trained design macro model;

[0100] The target component determination unit is used to determine the interaction mode that matches the design element annotation and the design attribute annotation as the target interaction mode, and determine the component that matches the design element annotation and the design attribute annotation as the target component.

[0101] Furthermore, the target webpage generation module 430 includes:

[0102] A current interaction demand determining unit, configured to determine a current interaction demand through the preset dynamic optimization mechanism and the target interaction mode;

[0103] A front-end code generating unit, configured to generate at least one set of front-end codes based on the current interaction requirement;

[0104] The target webpage generating unit is used to generate the target webpage based on each of the front-end codes and each of the target components.

[0105] Furthermore, the webpage generation device 400 based on the large model includes:

[0106] User feedback acquisition module, used to obtain user feedback;

[0107] An iterative training module, configured to iteratively train the pre-trained design macro model in response to the user feedback;

[0108] The target web page updating module is used to optimize the target web page by iteratively training the pre-trained design large model, and update the optimized web page to the target web page.

[0109] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device and each module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0110] The above-mentioned device can be implemented in the form of a computer program. Figure 5 Runs on the computer device shown.

[0111] See also Figure 5 , Figure 5 1 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device may be a server.

[0112] See also Figure 5 The computer device includes a processor, a memory and a network interface connected through a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0113] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any one of the webpage generation methods based on the large model.

[0114] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0115] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any web page generation method based on the large model.

[0116] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0117] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0118] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0119] Determining initial design elements and initial design attributes of the initial document based on a preset document parsing algorithm and the initial document;

[0120] Determine a target interaction mode and at least one target component through a pre-trained design macro model, the initial design elements, and the initial design attributes;

[0121] A target webpage is generated according to a preset dynamic optimization mechanism, the target interaction mode and each of the target components.

[0122] In one embodiment, based on a preset manuscript parsing algorithm and an initial manuscript, initial design elements and initial design attributes of the initial manuscript are determined to achieve:

[0123] Performing a standardization operation on the initial image document to generate a standard image in a standard format, wherein the standardization operation includes a scaling operation, a cropping operation and / or a denoising operation;

[0124] Extracting image elements in the standard image by using the preset document parsing algorithm, wherein the image elements include button elements, text box elements and / or icon elements;

[0125] The standard image is classified according to the button element, the text box element and / or the icon element to determine the initial design element and the initial design attribute.

[0126] In one embodiment, based on a preset manuscript parsing algorithm and an initial manuscript, initial design elements and initial design attributes of the initial manuscript are determined to achieve:

[0127] Extracting text information from the initial text manuscript using the preset manuscript parsing algorithm;

[0128] Determine the RGB information of the text information according to the color channel information RGB extraction algorithm;

[0129] Determining character spacing, character size and / or character layout of each character in the text information;

[0130] The initial design elements and the initial design attributes are determined according to the RGB information, the character spacing, the character size and / or the character layout.

[0131] In one embodiment, before determining the target interaction mode and at least one target component through the pre-trained design macro model, the initial design elements and the initial design attributes, it is used to achieve:

[0132] Obtain a historical interaction mode training set and a historical component training set as a training set;

[0133] The training set is labeled by using the design macro model to obtain each historical design element and each historical design attribute in the training set;

[0134] Manual annotations of the training set are obtained, and the design macro model is pre-trained through the manual annotations, each of the historical design elements and each of the historical design attributes to obtain the pre-trained design macro model.

[0135] In one embodiment, a target interaction mode and at least one target component are determined by using a pre-trained design macro model, the initial design elements, and the initial design attributes to achieve:

[0136] Converting the initial design element and the initial design attribute into a design element annotation and a design attribute annotation respectively;

[0137] Matching the design element annotations and the design attribute annotations with a preset interaction mode-component library through the pre-trained design macro model;

[0138] An interaction mode that matches the design element annotation and the design attribute annotation is determined as the target interaction mode, and a component that matches the design element annotation and the design attribute annotation is determined as the target component.

[0139] In one embodiment, a target webpage is generated according to a preset dynamic optimization mechanism, the target interaction mode and each of the target components, for achieving:

[0140] Determine the current interaction demand through the preset dynamic optimization mechanism and the target interaction mode;

[0141] Based on the current interaction requirements, generate at least one set of front-end codes;

[0142] The target web page is generated based on each of the front-end codes and each of the target components.

[0143] In one embodiment, after the target webpage is generated according to the preset dynamic optimization mechanism, the target interaction mode and each of the target components, it is used to achieve:

[0144] Get user feedback;

[0145] In response to the user feedback, iteratively training the pre-trained design macro model;

[0146] The target webpage is optimized by iteratively training the pre-trained design large model, and the optimized webpage is updated as the target webpage.

[0147] A computer-readable storage medium is also provided in an embodiment of the present application, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and the processor executes the program instructions to implement any one of the large model-based web page generation methods provided in the embodiments of the present application.

[0148] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (Smart Medi a Card, SMC), a secure digital (Secure Digital, SD) card, a flash memory card (Flash Card), etc., equipped on the computer device.

[0149] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A web page generation method based on a large model, characterized in that: include: Determining initial design elements and initial design attributes of the initial document based on a preset document parsing algorithm and the initial document; Determine a target interaction mode and at least one target component through a pre-trained design macro model, the initial design elements, and the initial design attributes; A target webpage is generated according to a preset dynamic optimization mechanism, the target interaction mode and each of the target components.

2. The webpage generation method based on a large model according to claim 1, characterized in that: The initial manuscript includes an initial image manuscript, and determining the initial design elements and initial design attributes of the initial manuscript based on a preset manuscript parsing algorithm and the initial manuscript includes: Performing a standardization operation on the initial image document to generate a standard image in a standard format, wherein the standardization operation includes a scaling operation, a cropping operation and / or a denoising operation; Extracting image elements in the standard image by using the preset document parsing algorithm, wherein the image elements include button elements, text box elements and / or icon elements; The standard image is classified according to the button element, the text box element and / or the icon element to determine the initial design element and the initial design attribute.

3. The webpage generation method based on a large model according to claim 1, characterized in that: The initial manuscript includes an initial text manuscript, and determining the initial design elements and initial design attributes of the initial manuscript based on a preset manuscript parsing algorithm and the initial manuscript includes: Extracting text information from the initial text manuscript using the preset manuscript parsing algorithm; Determine the RGB information of the text information according to the color channel information RGB extraction algorithm; Determining character spacing, character size and / or character layout of each character in the text information; The initial design elements and the initial design attributes are determined according to the RGB information, the character spacing, the character size and / or the character layout.

4. The webpage generation method based on a large model according to claim 1, characterized in that: Before determining the target interaction mode and at least one target component by using the pre-trained design macro model, the initial design elements and the initial design attributes, the method includes: Obtain a historical interaction mode training set and a historical component training set as a training set; The training set is labeled by using the design macro model to obtain each historical design element and each historical design attribute in the training set; Manual annotations of the training set are obtained, and the design macro model is pre-trained through the manual annotations, each of the historical design elements and each of the historical design attributes to obtain the pre-trained design macro model.

5. The webpage generation method based on a large model according to claim 4 is characterized in that: The determining of a target interaction mode and at least one target component by using the pre-trained design macro model, the initial design elements and the initial design attributes comprises: Converting the initial design element and the initial design attribute into a design element annotation and a design attribute annotation respectively; Matching the design element annotations and the design attribute annotations with a preset interaction mode-component library through the pre-trained design macro model; An interaction mode that matches the design element annotation and the design attribute annotation is determined as the target interaction mode, and a component that matches the design element annotation and the design attribute annotation is determined as the target component.

6. The webpage generation method based on a large model according to claim 1, characterized in that: The generating of the target webpage according to the preset dynamic optimization mechanism, the target interaction mode and each of the target components includes: Determine the current interaction demand through the preset dynamic optimization mechanism and the target interaction mode; Based on the current interaction requirements, generate at least one set of front-end codes; The target web page is generated based on each of the front-end codes and each of the target components.

7. The webpage generation method based on a large model according to any one of claims 1 to 6, characterized in that: After the target webpage is generated according to the preset dynamic optimization mechanism, the target interaction mode and each target component, the method includes: Get user feedback; In response to the user feedback, iteratively training the pre-trained design macro model; The target webpage is optimized by iteratively training the pre-trained design large model, and the optimized webpage is updated as the target webpage.

8. A web page generation device based on a large model, characterized in that: include: An element and attribute determination module, used to determine the initial design elements and initial design attributes of the initial document based on a preset document parsing algorithm and the initial document; An interaction mode and component determination module, configured to determine a target interaction mode and at least one target component through a pre-trained design macro model, the initial design elements and the initial design attributes; The target webpage generation module is used to generate a target webpage according to a preset dynamic optimization mechanism, the target interaction mode and each of the target components.

9. A computer device, characterized in that: The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is used to execute the computer program and implement the web page generation method based on a large model as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the web page generation method based on a large model according to any one of claims 1 to 7.

Images