AI training data display method and terminal
By type conversion and visualizing the data during AI training, the complexity and accuracy of data processing in AI training are solved, data visualization and correction are realized, data readability and reliability are improved, and data sharing and decision-making are supported.
Patent Information
- Application Number
- CN202510275289.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-10
- Publication Date
- 2025-08-01
AI Technical Summary
The existing AI training technology has complex data processing, simple user needs, insufficient data views, and it is difficult to ensure accuracy and fluency of cleaned data, resulting in inefficient data processing.
Provides a method for displaying AI training data. By obtaining process event data and performing type conversion, generating view data, saving it as a view file for visual display, and reconverting the data type when dirty data is found to correct errors.
It improves the visual adaptability and readability of data, enhances the maintainability and reliability of data, supports data sharing and collaboration, ensures the accuracy and consistency of data, and facilitates data analysis and decision-making.
Smart Images
Figure CN120407518A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data processing, and particularly relates to a method for displaying AI training data and a terminal. Background Art
[0002] With the rapid development of information technology, data processing and analysis technologies have increasingly become the core tools in various fields, especially in AI training and data model applications. Among them, AI training uses a large amount of data and algorithms to teach machine learning models to perform tasks, specifically including collecting a large amount of data from various sources (these data include multiple data types such as text, pictures, audio, and video), cleaning the collected data to remove errors, duplicates, and irrelevant information in the data, and learning and training the data through building a model, and finally forming a corresponding machine model framework for decision-making and constructing neural networks, etc. During the training process, the program can verify the data through output or according to the data model to ensure the integrity and correctness of the data. However, this AI training method has certain limitations. The data training of AI is based on language types, which requires high requirements for programmers, and users urgently need a simple data view in conventional scenarios, and it is also difficult to ensure the accuracy and smooth process of the cleaned data. The problem of "dirty data" still exists, which has a negative impact on the effectiveness and efficiency of data processing. Summary of the Invention
[0003] The technical problem to be solved by the present invention is: to provide a method for displaying AI training data and a terminal, which can realize the visualization of the data in the AI training process, improve the readability of the data, and can correct the view data to facilitate data management.
[0004] In order to solve the above technical problem, the technical solution adopted by the present invention is: A method for displaying AI training data, comprising the steps of: S1. Obtain the process event data in the AI training process, and perform type conversion on the process event data to obtain view data; S2. Save the view data as a view file, and use the view file to visually display the view data; S3. If dirty data appears during the visual display process, re-perform type conversion on the process event data.
[0005] In order to solve the above technical problem, another technical solution adopted by the present invention is: An AI training data display terminal, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements each step of the above-mentioned method for displaying AI training data.
[0006] The beneficial effects of the present invention are as follows: An AI training data display method and terminal provided by the present invention can handle data of multiple data types, improve the adaptability of the data visualization process, and convert different types of process event data into unified view data for data processing and analysis by obtaining the process event data during the AI training process and performing type conversion on the process event data; save the view data as a view file, and use the view file to visually display the view data, enhancing the readability and maintainability of the data, facilitating data sharing and collaboration among different personnel, and providing strong support for further data analysis, model optimization, and decision-making; if dirty data appears during the visual display process, re-perform type conversion on the process event data to ensure the accuracy and consistency of the data, thereby enhancing the reliability of the entire visual display process. BRIEF DESCRIPTION OF THE DRAWINGS
[0007] Figure 1 is a flowchart of an AI training data display method according to an embodiment of the present invention; Figure 2 is a composition structure diagram of AI training data in an embodiment of the present invention; Figure 3 is a flowchart of type conversion in an embodiment of the present invention; Figure 4 is another flowchart of an AI training data display method according to an embodiment of the present invention; Figure 5 is a flowchart of data correction in an embodiment of the present invention; Figure 6 is a schematic diagram of an AI training data display terminal according to an embodiment of the present invention; Label Description: 1. An AI training data display terminal; 2. A memory; 3. A processor. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0008] To describe the technical content, achieved objectives, and effects of the present invention in detail, the following is described in conjunction with the embodiments and with reference to the accompanying drawings.
[0009] Please refer to Figure 1 , an embodiment of the present invention provides an AI training data display method, including the steps of: S1. Obtain the process event data during the AI training process, and perform type conversion on the process event data to obtain view data; S2. Save the view data as a view file, and use the view file to visually display the view data; S3. If there is dirty data during the visual display process, re - perform type conversion on the process event data.
[0010] As can be seen from the above description, the beneficial effects of the present invention are as follows: By obtaining the process event data during the AI training process and performing type conversion on the process event data to obtain view data, it can handle data of multiple data types, improve the adaptability of the data visualization process, and convert different types of process event data into unified view data for data processing and analysis; Saving the view data as a view file and using the view file to visually display the view data enhances the readability and maintainability of the data, facilitates data sharing and collaboration among different personnel, and provides strong support for further data analysis, model optimization, and decision - making; If there is dirty data during the visual display process, re - perform type conversion on the process event data to ensure the accuracy and consistency of the data, thereby enhancing the reliability of the entire visual display process.
[0011] Further, performing type conversion on the process event data to obtain view data includes: Searching and matching in the data type library according to the data type of the process event data to obtain the corresponding data conversion rule; Converting the process event data into view data according to the data conversion rule, where the view data includes a data timestamp, a data type, and a data storage path.
[0012] As can be seen from the above description, by searching and matching in the data type library and obtaining the corresponding conversion rule, converting the process event data into view data, and converting different types of data into unified view data, it improves the readability of the data for subsequent data processing and analysis.
[0013] Further, step S1 further includes: Saving the process event data into a data container and generating a data file.
[0014] As can be seen from the above description, by saving the process event data into a data container and generating a data file, it realizes the persistent storage of data, provides a data backup file to handle the situation of dirty data, and provides support for data correction.
[0015] Further, step S3 includes: If there is dirty data during the visual display process, obtain the corresponding original process event data in the data file, re - perform type conversion on the original process event data to obtain the correct view data of the target type, and use the view data to replace the dirty data.
[0016] As described above, by re - performing type conversion on the original process event data, correct view data is obtained, and the dirty data is replaced with the correct view data to achieve data correction, ensuring the reliability of the visualization display process and providing strong support for decision - makers.
[0017] Further, the process event data in the AI training process includes training data, process data, and final data; The training data is the initial data set used in the AI model training process; The process data is the intermediate data generated during the AI model training process; The final data is the model data generated after the AI model training is completed.
[0018] As described above, visualizing the data involved in each process during the AI model training process improves the readability of the data, intuitively understands the AI model training process, and facilitates AI data training and its correction.
[0019] Please refer to Figure 6 , another embodiment of the present invention provides a display terminal for AI training data, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements each step of the above - mentioned display method for AI training data.
[0020] The above - mentioned display method and terminal for AI training data of the present invention are applicable to the AI model training process, can realize the visualization of AI training process data, improve the readability of the data, and can correct the view data, facilitating data management. The following is illustrated by specific embodiments: Please refer to Figures 1 to 5 , Embodiment 1 of the present invention is: A display method for AI training data, including the steps: S1. Obtain the process event data in the AI training process, and perform type conversion on the process event data to obtain view data.
[0021] S11. Obtain the process event data in the AI training process. The process event data in the AI training process includes training data, process data, and final data. The training data is the initial data set used in the AI model training process, the process data is the intermediate data generated during the AI model training process, and the final data is the model data generated after the AI model training is completed.
[0022] Please refer to Figure 2, specifically, the process data includes all data, dirty data, and corrected data during the data cleaning process. The training data, process data, and final data will be converted into Excel view data for storage and display in an Excel table.
[0023] Please refer to Figure 3 , during the AI training process, input events, output events, and process events will all generate data, namely training data, process data, and final data. Convert the training data, process data, and final data into view data for visual display, improve the readability of the data, intuitively understand the AI model training process, and facilitate AI data training and its correction.
[0024] S12. Search and match in the data type library according to the data type of the process event data to obtain the corresponding data conversion rule.
[0025] S13. Convert the process event data into view data according to the data conversion rule. The view data includes a data timestamp, a data type, and a data storage path.
[0026] Among them, the data content of the view data is shown in Table 1.
[0027] Table 1 Data Content of View Data
[0028] In this embodiment, please refer to Figure 4 , the data type library includes multiple data types, such as: time conversion type, byte type, audio data type, video data type, and image data type. By searching and matching in the data type library and obtaining the corresponding conversion rule, convert the process event data into view data, convert different types of data into unified view data, improve the readability of the data, and facilitate subsequent data processing and analysis. For example, according to the conversion rule of converting MM / DD / YYYY to YYYY-MM-DD, convert a piece of data 12 / 12 / 2024 to 2024-12-12.
[0029] S2. Save the view data as a view file and use the view file to visually display the view data.
[0030] In this embodiment, save the view data as a view file and use the view file to visually display the view data, enhancing the readability and maintainability of the data, facilitating data sharing and collaboration among different personnel, and providing strong support for further data analysis, model optimization, and decision-making.
[0031] Further, in this embodiment, step S1 further includes: saving the process event data to a data container and generating a data file. By saving the process event data to the data container and generating a data file, persistent storage of the data is achieved, a data backup file is provided to handle the case of dirty data, and support is provided for data correction. Specifically, the data file is a.DAT data file. The.DAT data file is stored in the data container and is a part of the data container. The data in the data file will be presented as view data, and it is determined whether the data after type conversion is correctly understood through the data file, that is, whether there is dirty data.
[0032] S3. If dirty data appears during the visual display process, re - perform type conversion on the process event data, specifically including: If dirty data appears during the visual display process, obtain the corresponding original process event data in the data file, re - perform type conversion on the original process event data to obtain the correct view data of the target type, and use the view data to replace the dirty data.
[0033] In this example, please refer to Figure 5 , obtain the original process event data corresponding to the dirty data through the data file, re - perform type conversion on the original process event data to obtain the correct view data of the target type, and use the view data to replace the dirty data to ensure the reliability of the visual display process and provide strong support for decision - makers. Specifically, Table 2 shows the incorrect view data, which incorrectly converts text - type data to picture - type and saves it in xx.xls. Take out the corresponding original process event data from the data file, perform re - type conversion, and output and save it in xx.xls. At this time, the view data is shown in Table 3.
[0034] Table 2 Incorrect view data
[0035] Table 3 Correct view data
[0036] Please refer to Figure 6 , Embodiment 2 of the present invention is: A display terminal 1 for AI training data, including a memory 2, a processor 3, and a computer program stored on the memory 2 and executable on the processor 3. When the processor 3 executes the computer program, each step of the display method for AI training data in Embodiment 1 is implemented.
[0037] In summary, a method and a terminal for displaying AI training data provided by the present invention obtain process event data during the AI training process, and perform type conversion on the process event data to obtain view data, including: searching and matching in a data type library according to the data type of the process event data to obtain a corresponding data conversion rule; converting the process event data into view data according to the data conversion rule, where the view data includes a data timestamp, a data type, and a data storage path. The process event data during the AI training process includes training data, process data, and final data. The training data is the initial data set used during the AI model training process, the process data is the intermediate data generated by the AI model during the training process, and the final data is the model data generated after the AI model training is completed. Type conversion is performed on the data involved in each process during the AI model training process to facilitate visual display, improve the readability of the data, intuitively understand the AI model training process, and facilitate AI data training and its correction.
[0038] The view data is saved as a view file, and the view file is used to visually display the view data, enhancing the readability and maintainability of the data, facilitating data sharing and collaboration among different personnel, and providing strong support for further data analysis, model optimization, and decision-making. The process event data is saved in a data container and a data file is generated. If there is dirty data during the visual display process, the corresponding original process event data is obtained from the data file, and the original process event data is re-converted to obtain view data of the correct target type, and the view data is used to replace the dirty data to ensure the reliability of the visual display process.
[0039] The above are only embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent transformation made using the specification and drawings of the present invention, or directly or indirectly applied in related technical fields, is similarly included in the patent protection scope of the present invention.
Claims
1. A method for displaying AI training data, characterized in that, Including the steps: S1. Obtain the process event data during the AI training process, and perform type conversion on the process event data to obtain view data; S2. Save the view data as a view file, and use the view file to visually display the view data; S3. If dirty data appears during the visual display process, re-perform type conversion on the process event data.
2. The display method of AI training data according to claim 1, wherein Performing type conversion on the process event data to obtain view data includes: Search and match in the data type library according to the data type of the process event data to obtain the corresponding data conversion rule; Convert the process event data into view data according to the data conversion rule, and the view data includes a data timestamp, a data type, and a data storage path.
3. The display method of AI training data according to claim 1, wherein, The step S1 further includes: Save the process event data into a data container and generate a data file.
4. The display method of AI training data according to claim 3, characterized in that The step S3 includes: If dirty data appears during the visual display process, obtain the corresponding original process event data from the data file, re-perform type conversion on the original process event data to obtain view data of the correct target type, and use the view data to replace the dirty data.
5. The display method of AI training data according to claim 1, wherein, The process event data during the AI training process includes training data, process data, and final data; The training data is the initial data set used during the AI model training process; The process data is the intermediate data generated during the AI model training process; The final data is the model data generated after the AI model training is completed.
6. A display terminal for AI training data, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the following steps are implemented: S1. Obtain the process event data during the AI training process, and perform type conversion on the process event data to obtain view data; S2. Save the view data as a view file, and use the view file to visually display the view data; S3. If dirty data appears during the visual display process, re-perform type conversion on the process event data.
7. An AI training data display terminal according to claim 6, characterized in that, Performing type conversion on the process event data to obtain view data includes: Search and match in the data type library according to the data type of the process event data to obtain the corresponding data conversion rule; Convert the process event data into view data according to the data conversion rule, and the view data includes a data timestamp, a data type, and a data storage path.
8. The display terminal for AI training data according to claim 6, characterized in that, The step S1 further includes: Save the process event data into a data container and generate a data file.
9. The display terminal for AI training data according to claim 8, wherein, The step S3 includes: If dirty data appears during the visual display process, obtain the corresponding original process event data from the data file, re-perform type conversion on the original process event data to obtain view data of the correct target type, and use the view data to replace the dirty data.
10. The display terminal for AI training data according to claim 6, characterized in that, The process event data during the AI training process includes training data, process data, and final data; The training data is the initial data set used during the AI model training process; The process data is the intermediate data generated during the AI model training process; The final data is the model data generated after the AI model training is completed.