Fitness course evaluation method and device, storage medium and electronic equipment

Abstract

The application discloses a fitness course evaluation method and device, a storage medium and an electronic equipment. It relates to the field of artificial intelligence. The method comprises the following steps: obtaining structured course data of a plurality of to-be-evaluated fitness courses, wherein the structured course data at least comprises an action list and a course duration; generating an evaluation prompt word corresponding to each to-be-evaluated fitness course according to the structured course data and a preset prompt word template; and evaluating each to-be-evaluated fitness course according to the evaluation prompt word by using a plurality of evaluation large models to obtain an evaluation result corresponding to each to-be-evaluated fitness course, wherein the evaluation result at least comprises quality grade information and disposal suggestion information. The method solves the problem of low evaluation efficiency in the related art that the quality of the fitness course content is evaluated by relying on artificial evaluation.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and more specifically, to a method and apparatus for evaluating fitness courses, a storage medium, and an electronic device. Background Technology

[0002] With the rapid development of online fitness platforms, the amount of course content is constantly increasing. Currently, the relevant technologies mainly rely on manual evaluation of fitness course content quality, which is inefficient and results in significant differences in scores from different reviewers for the same course, leading to inconsistent content quality.

[0003] The reliance on manual evaluation of fitness course content in related technologies results in low evaluation efficiency, and no effective solution has yet been proposed. Summary of the Invention

[0004] The main purpose of this application is to provide a method, device, storage medium, and electronic device for evaluating fitness courses, so as to solve the problem that the evaluation of fitness course content relies on manual evaluation, which has low evaluation efficiency.

[0005] To achieve the above objectives, according to one aspect of this application, a method for evaluating fitness courses is provided. The method includes: acquiring structured course data for multiple fitness courses to be evaluated, wherein the structured course data includes at least a list of movements and course duration; generating evaluation prompts for each fitness course to be evaluated based on the structured course data and a preset prompt template; and using multiple large-scale evaluation models to evaluate each fitness course to be evaluated based on the evaluation prompts, obtaining an evaluation result for each fitness course to be evaluated, wherein the evaluation result includes at least quality level information and suggested remedial measures.

[0006] Furthermore, multiple large-scale evaluation models are used to evaluate each fitness course to be evaluated based on evaluation prompts. The evaluation results for each fitness course to be evaluated include: for each fitness course to be evaluated, the evaluation prompts corresponding to the fitness course to be evaluated are sent to each large-scale evaluation model in parallel to obtain the raw evaluation data returned by each large-scale evaluation model; the raw evaluation data returned by each large-scale evaluation model is formatted to obtain multiple structured evaluation results for the fitness course to be evaluated; the multiple structured evaluation results are parsed to extract the scoring sub-items of multiple dimensions in each structured evaluation result; and the evaluation result corresponding to the fitness course to be evaluated is calculated based on the scoring sub-items of multiple dimensions in each structured evaluation result.

[0007] Furthermore, based on the scoring sub-items of multiple dimensions in each structured assessment result, the assessment result corresponding to the fitness course to be assessed is calculated as follows: For each structured assessment result, the scoring sub-items of each dimension are weighted and calculated to obtain the score value corresponding to each dimension, and the total score value corresponding to the structured assessment result is calculated based on the score value corresponding to each dimension; the total score value corresponding to the fitness course to be assessed is calculated based on the confidence level included in each structured assessment result and the total score value corresponding to each structured assessment result; based on the total score value corresponding to the fitness course to be assessed, the quality level information and handling suggestion information corresponding to the fitness course to be assessed are determined, and based on the quality level information and handling suggestion information corresponding to the fitness course to be assessed, the assessment result corresponding to the fitness course to be assessed is determined.

[0008] Furthermore, before obtaining the structured course data of multiple fitness courses to be evaluated, the method further includes: obtaining a course identifier list, wherein the course identifier list contains identifiers of multiple original fitness courses to be evaluated; obtaining the original course data of each original fitness course to be evaluated based on the course identifier list, and performing transformation processing on the original course data of each original fitness course to be evaluated to obtain the structured course data of each original fitness course to be evaluated; verifying the structured course data of each original fitness course to be evaluated to obtain a verification result, and determining multiple fitness courses to be evaluated from the multiple original fitness courses to be evaluated based on the verification result.

[0009] Furthermore, the structured course data for each original fitness course to be evaluated is validated, and the validation results include: for each original fitness course to be evaluated, the number of movements and the cumulative duration of each movement are determined based on the movement list of the original fitness course to be evaluated, and the duration deviation is calculated based on the course duration and the cumulative duration of each movement; the number of movements is compared with a first threshold, and the duration deviation is compared with a second threshold; if the number of movements is less than the first threshold, or the duration deviation is greater than or equal to the second threshold, then the original fitness course to be evaluated fails the validation and this is the validation result; if the number of movements is greater than or equal to the first threshold, and the duration deviation is less than the second threshold, then the original fitness course to be evaluated passes the validation and this is the validation result.

[0010] Furthermore, based on the structured course data and the preset prompt word template, the evaluation prompt words for each fitness course to be evaluated are generated as follows: For each fitness course to be evaluated, a preset prompt word template is obtained, wherein the preset prompt word template contains placeholders used to represent the information to be processed; a structured text string is constructed based on the structured course data, wherein the structured text string is used to represent the information to be processed; the structured text string is filled into the placeholders to generate the evaluation prompt words corresponding to the fitness course to be evaluated.

[0011] Furthermore, after obtaining the evaluation results for each fitness course to be evaluated, the method further includes: serializing the evaluation results for each fitness course to be evaluated to obtain serialized data; writing the serialized data into a checkpoint file, wherein the checkpoint file is used to record the completed evaluation results so as to perform recovery processing based on the checkpoint file after the program is interrupted.

[0012] To achieve the above objectives, according to another aspect of this application, a fitness course evaluation device is provided. The device includes: a first acquisition unit for acquiring structured course data of multiple fitness courses to be evaluated, wherein the structured course data includes at least a list of movements and course duration; a first processing unit for generating evaluation prompts for each fitness course to be evaluated based on the structured course data and a preset prompt template; and a second processing unit for evaluating each fitness course to be evaluated using multiple large-scale evaluation models based on the evaluation prompts, obtaining an evaluation result for each fitness course to be evaluated, wherein the evaluation result includes at least quality level information and suggested remedial measures.

[0013] Furthermore, the second processing unit includes: a first processing subunit, used to send evaluation prompts corresponding to each fitness course to be evaluated to each large evaluation model in parallel, so as to obtain the raw evaluation data returned by each large evaluation model; a second processing subunit, used to perform format processing on the raw evaluation data returned by each large evaluation model to obtain multiple structured evaluation results corresponding to the fitness course to be evaluated; a third processing subunit, used to parse the multiple structured evaluation results and extract the scoring sub-items of multiple dimensions in each structured evaluation result; and a fourth processing subunit, used to calculate based on the scoring sub-items of multiple dimensions in each structured evaluation result to obtain the evaluation result corresponding to the fitness course to be evaluated.

[0014] Furthermore, the fourth processing subunit includes: a first processing module, used to perform weighted calculations on the scoring sub-items of each dimension for each structured evaluation result to obtain the score value corresponding to each dimension, and to calculate the total score value corresponding to the structured evaluation result based on the score value corresponding to each dimension; a second processing module, used to calculate the total score value corresponding to the fitness course to be evaluated based on the confidence level contained in each structured evaluation result and the total score value corresponding to each structured evaluation result; and a third processing module, used to determine the quality level information and handling suggestion information corresponding to the fitness course to be evaluated based on the total score value corresponding to the fitness course to be evaluated, and to determine the evaluation result corresponding to the fitness course to be evaluated based on the quality level information and handling suggestion information corresponding to the fitness course to be evaluated.

[0015] Furthermore, the device also includes: a second acquisition unit, configured to acquire a list of course identifiers before acquiring structured course data of multiple fitness courses to be evaluated, wherein the list of course identifiers contains identifiers of multiple original fitness courses to be evaluated; a third acquisition unit, configured to acquire the original course data of each original fitness course to be evaluated based on the list of course identifiers, and to perform conversion processing on the original course data of each original fitness course to be evaluated to obtain structured course data of each original fitness course to be evaluated; and a determination unit, configured to verify the structured course data of each original fitness course to be evaluated, obtain a verification result, and determine multiple fitness courses to be evaluated from multiple original fitness courses to be evaluated based on the verification result.

[0016] Further, the determining unit includes: a first determining subunit, used to determine the number of movements and the cumulative duration of each original fitness course to be evaluated based on the movement list of the original fitness course to be evaluated, and to calculate the duration deviation based on the course duration and cumulative duration of the movements of the original fitness course to be evaluated; a comparison subunit, used to compare the number of movements with a first threshold, and to compare the duration deviation with a second threshold; a second determining subunit, used to determine if the number of movements is less than the first threshold, or the duration deviation is greater than or equal to the second threshold, that the original fitness course to be evaluated fails the verification as a verification result; and a third determining subunit, used to determine if the number of movements is greater than or equal to the first threshold, and the duration deviation is less than the second threshold, that the original fitness course to be evaluated passes the verification as a verification result.

[0017] Furthermore, the first processing unit includes: a fifth processing subunit, used to obtain a preset prompt word template for each fitness course to be evaluated, wherein the preset prompt word template contains placeholders used to represent the information to be processed; a sixth processing subunit, used to construct a structured text string based on the structured course data, wherein the structured text string is used to represent the information to be processed; and a seventh processing subunit, used to fill the placeholders with the structured text string to generate the evaluation prompt words corresponding to the fitness course to be evaluated.

[0018] Furthermore, the device also includes: a third processing unit, used to serialize the evaluation results corresponding to each fitness course to be evaluated after obtaining the evaluation results, to obtain serialized data; and a fourth processing unit, used to write the serialized data into a checkpoint file, wherein the checkpoint file is used to record the completed evaluation results so as to perform recovery processing based on the checkpoint file after the program is interrupted.

[0019] According to another aspect of the present invention, an electronic device is also provided, comprising: a memory storing an executable program; and a processor for running the program, wherein the program executes the fitness course evaluation method described above when running.

[0020] According to another aspect of the present invention, a computer-readable storage medium is also provided, the storage medium storing a program, wherein, when the program is running, the device where the storage medium is located executes the evaluation method of any of the above-mentioned fitness courses.

[0021] In this embodiment, the following steps are employed: First, structured course data for multiple fitness courses to be evaluated is obtained, wherein the structured course data includes at least a list of movements and course duration. Second, evaluation prompts are generated for each fitness course based on the structured course data and a preset prompt template. Third, multiple large-scale evaluation models are used to evaluate each fitness course based on the evaluation prompts, resulting in an evaluation result for each fitness course. The evaluation result includes at least quality level information and suggested remedial measures. This addresses the technical problem of low evaluation efficiency in related technologies that rely on manual evaluation of fitness course content quality.

[0022] In this solution, structured course data is acquired and evaluation prompts are generated. Multiple large-scale evaluation models are used in parallel to evaluate fitness courses, which improves evaluation efficiency, overcomes the problem of subjective and inconsistent standards, and enables batch and rapid processing of course content. Based on the quality level information and handling suggestions output by the multi-model comprehensive evaluation, the objectivity, consistency and interpretability of the evaluation results are improved. Attached Figure Description

[0023] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0024] Figure 1 A hardware block diagram of a computer terminal for implementing a fitness course assessment method is shown.

[0025] Figure 2 This is a flowchart of a fitness course evaluation method provided according to an embodiment of this application;

[0026] Figure 3 This is a schematic diagram of a fitness course evaluation device provided according to an embodiment of this application;

[0027] Figure 4 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] It should be noted that the information collected in this application (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of this data all comply with relevant laws, regulations, and standards, and necessary measures have been taken to ensure compliance with public order and good morals. Corresponding access points are provided for users to choose whether to authorize or refuse. For example, interfaces are established between this system and relevant users or organizations, providing users with corresponding access points to choose whether to agree to or refuse automated decision-making results; if the user chooses to refuse, the process proceeds to the expert decision-making stage.

[0031] Example 1

[0032] According to an embodiment of this application, a method embodiment for evaluating fitness courses is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0033] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1A hardware block diagram of a computer terminal (or mobile device) for implementing an assessment method for fitness classes is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0034] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0035] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the fitness course evaluation method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned fitness course evaluation method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0036] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0037] The display may be a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0038] Under the aforementioned operating environment, this application provides the following: Figure 2 The evaluation method for the fitness course shown. Figure 2 This is a flowchart of the evaluation method for a fitness course according to Embodiment 1 of this application. The evaluation method for the fitness course includes:

[0039] Step S201: Obtain structured course data for multiple fitness courses to be evaluated. The structured course data shall include at least a list of exercises and course duration.

[0040] Step S202: Based on the structured course data and the preset prompt word template, generate evaluation prompt words for each fitness course to be evaluated;

[0041] Step S203: Use multiple large-scale evaluation models to evaluate each fitness course to be evaluated based on evaluation prompts, and obtain the evaluation results for each fitness course to be evaluated. The evaluation results include at least quality level information and handling suggestions.

[0042] Optionally, the evaluation system for fitness courses can be used as the main execution entity. The system acquires structured course data from multiple fitness courses to be evaluated. This structured course data includes at least a list of exercises (such as jumping jacks, high knees, etc., and their estimated duration) and a course duration (e.g., 20 minutes). The fitness courses to be evaluated can be a batch of newly added courses to the online fitness platform's database. For example, these courses can be evaluated in batches to determine their quality level and provide recommendations for their release. Alternatively, the fitness courses to be evaluated can be existing courses in the platform's historical course library. For example, existing courses can be periodically re-evaluated to identify low-quality content and trigger optimization or removal processes.

[0043] Optionally, the system adopts a pipelined architecture, including a data extraction layer, a testability pre-detection layer, a prompt word generation layer, a large language model inference layer, and a scoring aggregation layer. From data extraction, testability pre-detection, prompt word generation, large language model inference to comprehensive weighted scoring, a complete automated pipeline is formed, supporting breakpoint resumption, concurrent batch processing, and automatic retries. The functions of each layer are described below with reference to specific embodiments, and will not be elaborated further here.

[0044] Optionally, in the fitness course evaluation method provided in this application embodiment, before obtaining the structured course data of multiple fitness courses to be evaluated, the method further includes: obtaining a course identifier list, wherein the course identifier list contains identifiers of multiple original fitness courses to be evaluated; obtaining the original course data of each original fitness course to be evaluated based on the course identifier list, and performing conversion processing on the original course data of each original fitness course to be evaluated to obtain structured course data of each original fitness course to be evaluated; verifying the structured course data of each original fitness course to be evaluated to obtain a verification result, and determining multiple fitness courses to be evaluated from multiple original fitness courses to be evaluated based on the verification result.

[0045] In an optional embodiment, before acquiring the structured course data of multiple fitness courses to be evaluated, the data extraction layer first acquires a list of course identifiers. This list contains identifiers of multiple original fitness courses to be evaluated. For example, if the platform adds 10 new fitness courses, the course identifier list is formed based on the identifiers of these 10 fitness courses. Then, the original course data of each original fitness course to be evaluated is acquired. For example, the list is traversed, and for each identifier, the original course data is acquired, including text descriptions, action video clips, tag identifiers, etc. Then, the original course data of each original fitness course to be evaluated is transformed to obtain the structured course data of each original fitness course to be evaluated. For example, the tag identifier 123 is converted into aerobic exercise; the action sequence in the video is parsed into a structured list, including action name, duration, number of repetitions, etc.

[0046] In an optional embodiment, the measurability pre-screening layer performs measurability gating verification, that is, it verifies the structured course data of each original fitness course to be evaluated, obtains the verification result, and then determines multiple fitness courses to be evaluated from multiple original fitness courses to be evaluated based on the verification result. For example, if the verification result of 2 out of 10 fitness courses is that the verification result is failed, then the determined fitness courses to be evaluated are 8.

[0047] Optionally, in the fitness course evaluation method provided in this application embodiment, the structured course data of each original fitness course to be evaluated is verified to obtain the verification result, including: for each original fitness course to be evaluated, the number of movements and the cumulative duration of the movements are determined according to the movement list of the original fitness course to be evaluated, and the duration deviation is calculated based on the course duration and the cumulative duration of the movements of the original fitness course to be evaluated; the number of movements is compared with a first threshold, and the duration deviation is compared with a second threshold; if the number of movements is less than the first threshold, or the duration deviation is greater than or equal to the second threshold, then the original fitness course to be evaluated fails the verification as the verification result; if the number of movements is greater than or equal to the first threshold, and the duration deviation is less than the second threshold, then the original fitness course to be evaluated passes the verification as the verification result.

[0048] In an optional embodiment, invalid courses (such as those with 0 movements or severely deviated durations) are filtered out through a testability pre-screening layer. This is divided into basic structure gating (e.g., the number of movements must be greater than or equal to 1, otherwise it is marked as skipped) and duration consistency gating (e.g., if the deviation between the declared course duration and the actual cumulative duration of the movement list exceeds a preset deviation threshold, it is marked as inconsistent and skipped). Optionally, for each original fitness course to be evaluated, the number of movements and the cumulative duration of the movements are determined based on the movement list of the original fitness course to be evaluated. The duration deviation is calculated based on the course duration and the cumulative duration of the movements. Then, the number of movements is compared with a first threshold (e.g., 1), and the duration deviation is compared with a second threshold (i.e., the preset deviation threshold). If the number of movements is less than the first threshold, or the duration deviation is greater than or equal to the second threshold, the original fitness course to be evaluated fails the verification, which is the verification result. If the number of movements is greater than or equal to the first threshold, and the duration deviation is less than the second threshold, the original fitness course to be evaluated passes the verification, which is the verification result.

[0049] By acquiring course identifiers and converting raw data into a structured format, combined with a dual verification mechanism of basic structure gating and duration consistency gating, the system can identify and filter out invalid courses with missing actions or serious duration deviations. This reduces invalid large model inference calls, saves computing resources, and improves the efficiency and reliability of batch evaluation.

[0050] Optionally, the prompt generation layer generates evaluation prompts for each fitness course to be evaluated based on structured course data and preset prompt templates. The preset prompt templates include role definition information, evaluation principle information, analysis process information, scoring information, and output format constraints. For example, the role definition information is sports scientist and fitness content analyst. Evaluation principle information includes the rigor principle (e.g., examining from a critical standpoint of "why it's not good enough") and the scarcity principle (e.g., stipulating that 3 points (excellent) must be extremely rare, and 2 points indicates basically acceptable but clearly not perfect). Analysis process information includes: neutral analysis centered on the exercise list: judging training structure, thematic consistency, intensity, and pacing based on the ordered exercise list; performing multi-dimensional quality assessment: evaluating each sub-dimensional one by one, outputting scores (e.g., integers 1-3) and reasons (Chinese explanations); generating descriptions and confidence levels: outputting a one-sentence summary, multi-dimensional detailed descriptions, and confidence levels (0-1). Scoring information may include scoring criteria, evaluation points, and mandatory triggering conditions. Output format constraints can include: requiring the output to conform to a predefined JSON pattern structure, including fields such as multi-dimensional ratings, optimized labels, descriptions, and confidence levels.

[0051] Optionally, the large language model inference layer uses multiple large evaluation models (such as native multimodal collaborative inference models, autoregressive language models based on generative pre-training architectures, and hybrid expert language models using sparse activation mechanisms) to evaluate each fitness course to be evaluated based on evaluation prompts, and obtains the evaluation results corresponding to each fitness course to be evaluated. The evaluation results include at least quality level information and handling suggestion information. For example, the quality level information can be excellent, good, average or very poor, and the handling suggestion information can be that it can be launched directly, needs to be fine-tuned before launch, needs to be deeply adjusted, or it is recommended to discard it.

[0052] In summary, by acquiring structured course data and generating assessment prompts, and using multiple large-scale assessment models to evaluate fitness courses in parallel, the assessment efficiency is improved. This overcomes the problems of subjective and inconsistent standards, enabling batch and rapid processing of course content. Based on the quality level information and handling suggestions output by the comprehensive evaluation of multiple models, the objectivity, consistency, and interpretability of the assessment results are enhanced.

[0053] Optionally, in the fitness course evaluation method provided in this application embodiment, multiple evaluation models are used to evaluate each fitness course to be evaluated based on evaluation prompts to obtain the evaluation results corresponding to each fitness course to be evaluated. This includes: for each fitness course to be evaluated, sending the evaluation prompts corresponding to the fitness course to be evaluated to each evaluation model in parallel to obtain the original evaluation data returned by each evaluation model; processing the format of the original evaluation data returned by each evaluation model to obtain multiple structured evaluation results corresponding to the fitness course to be evaluated; parsing the multiple structured evaluation results and extracting the scoring sub-items of multiple dimensions in each structured evaluation result; and calculating based on the scoring sub-items of multiple dimensions in each structured evaluation result to obtain the evaluation result corresponding to the fitness course to be evaluated.

[0054] In an optional embodiment, the large language model inference layer performs parallel calls to multiple models and uniformly outputs evaluation results in JSON format. For each fitness course to be evaluated, the corresponding evaluation prompt words for that fitness course are sent in parallel to each large evaluation model to obtain the raw evaluation data returned by each large evaluation model. For example, a thread pool executor is used to implement parallel inference of multiple models, and a default concurrency can be configured; each thread independently instantiates the large evaluation model to avoid state pollution between threads; the results are returned in index order to ensure the consistency of the order of results from multiple models.

[0055] In an optional embodiment, after obtaining the raw evaluation data returned by each large evaluation model, the raw evaluation data returned by each large evaluation model is formatted to obtain multiple structured evaluation results corresponding to the fitness course to be evaluated. The formatting process may include the following: removing code block wrappings and extracting valid JSON content between curly braces; using iterative regular expressions to replace and correct unescaped English quotation marks within Chinese strings; and using a fallback solution to handle complex formatting errors such as truncation and trailing commas. The structured evaluation results include multi-dimensional scores, optimized labels, descriptions, and confidence scores.

[0056] In one optional embodiment, multiple structured evaluation results are parsed, and scoring sub-items of multiple dimensions are extracted from each structured evaluation result. For example, after evaluating and processing using three large evaluation models, three structured evaluation results are obtained. These three structured evaluation results are then parsed, and scoring sub-items of multiple dimensions are extracted from each structured evaluation result. For example, multiple dimensions include course packaging quality, course labeling quality, and course arrangement quality. For instance, scoring sub-items for course packaging quality include information completeness, accuracy, and comprehensibility, while scoring sub-items for course arrangement quality include adherence to the theme, effectiveness, safety, and content completeness. Then, the scoring aggregation layer calculates based on the scoring sub-items of multiple dimensions in each structured evaluation result to obtain the evaluation result corresponding to the fitness course to be evaluated.

[0057] Optionally, in the fitness course evaluation method provided in this application embodiment, the evaluation result corresponding to the fitness course to be evaluated is calculated based on the scoring sub-items of multiple dimensions in each structured evaluation result. This includes: for each structured evaluation result, weighting the scoring sub-items of each dimension to obtain the score value corresponding to each dimension, and calculating the total score value corresponding to the structured evaluation result based on the score value corresponding to each dimension; calculating the total score value corresponding to the fitness course to be evaluated based on the confidence level included in each structured evaluation result and the total score value corresponding to each structured evaluation result; determining the quality level information and handling suggestion information corresponding to the fitness course to be evaluated based on the total score value corresponding to the fitness course to be evaluated, and determining the evaluation result corresponding to the fitness course to be evaluated based on the quality level information and handling suggestion information corresponding to the fitness course to be evaluated.

[0058] In an optional embodiment, the scoring aggregation layer performs weighted calculations on the scoring sub-items of each dimension for each structured evaluation result to obtain the score value corresponding to each dimension. Based on the score values ​​for each dimension, it calculates the total score value corresponding to the structured evaluation result. For example, for the course packaging quality dimension, the score value for this dimension can be a weighted average of the scores for the information completeness and accuracy scoring sub-item (e.g., with a weight of 0.70) and the comprehensibility scoring sub-item (e.g., with a weight of 0.30). The total score value corresponding to this structured evaluation result can be the sum of the score values ​​for the three dimensions: course packaging quality, course labeling quality, and course arrangement quality.

[0059] In an optional embodiment, the total score of the fitness course to be evaluated is calculated based on the confidence level of each structured assessment result and the total score corresponding to each structured assessment result. For example, if there are 3 structured assessment results, the confidence levels of these 3 structured assessment results are normalized to obtain the weight corresponding to each structured assessment result. Then, a weighted average can be calculated based on the weight corresponding to each structured assessment result and the total score corresponding to each structured assessment result to obtain the total score of the fitness course to be evaluated.

[0060] In an optional embodiment, based on the total score of the fitness course to be evaluated, the quality level information and handling suggestion information of the fitness course to be evaluated can be determined, and the evaluation result of the fitness course to be evaluated can be formed based on the quality level information and handling suggestion information of the fitness course to be evaluated. For example, the quality level information and handling suggestion information of the fitness course to be evaluated can be determined by matching the total score of the fitness course to be evaluated in the mapping relationship table shown in Table 1.

[0061] Table 1

[0062]

[0063] By using parallel reasoning with multiple models and independent instantiation to avoid state pollution, the efficiency and order consistency of the evaluation are guaranteed. The system's fault tolerance is improved by combining code block removal, regular expression repair, and fallback parsing. The scoring bias is reduced and the evaluation accuracy is improved by fusing multi-dimensional weighted calculation with confidence-based normalized weights.

[0064] In an optional embodiment, the evaluation model can also perform label correction and regeneration, for example, inferring and generating more accurate optimized labels based on the action list, including multiple dimensions (such as training objectives, movement types, differentiation patterns, training modes, etc.), realizing the dual functions of quality assessment and label optimization, thereby improving the distribution accuracy of the recommendation system.

[0065] Optionally, in the fitness course evaluation method provided in this application embodiment, generating evaluation prompts for each fitness course to be evaluated based on structured course data and preset prompt templates includes: for each fitness course to be evaluated, obtaining a preset prompt template, wherein the preset prompt template contains placeholders used to represent information to be processed; constructing a structured text string based on structured course data, wherein the structured text string is used to represent information to be processed; filling the placeholders with the structured text string to generate the evaluation prompts for the fitness course to be evaluated.

[0066] In an optional embodiment, the prompt word generation layer obtains a preset prompt word template for each fitness course to be evaluated. The template contains placeholders for representing the information to be processed. Then, it constructs structured information text for the course. For example, it combines the action list, course meta-information (such as title, duration, description), and tag information into a structured text string, which is used as the filling content for the placeholders. The structured text string is then filled into the placeholders to generate the evaluation prompt words corresponding to the fitness course to be evaluated.

[0067] By constructing a structured text string containing a list of actions, course metadata, and tag information, and filling in placeholders, the accurate determination of assessment prompts was achieved.

[0068] Optionally, in the fitness course evaluation method provided in this application embodiment, after obtaining the evaluation result corresponding to each fitness course to be evaluated, the method further includes: serializing the evaluation result corresponding to each fitness course to be evaluated to obtain serialized data; writing the serialized data into a checkpoint file, wherein the checkpoint file is used to record the completed evaluation results so as to perform recovery processing based on the checkpoint file after the program is interrupted.

[0069] In an optional embodiment, after obtaining the evaluation results for each fitness course to be evaluated, the system serializes the evaluation results for each course to obtain serialized data and writes the serialized data to a checkpoint file. This file records the completed evaluation results so that the system can resume processing based on the checkpoint file after a program interruption. For example, after processing each batch, the results are serialized and written to a checkpoint JSON file, allowing the program to resume from the breakpoint after an interruption and avoiding repeated model calls.

[0070] In an optional embodiment, the system first generates a checkpoint JSON file, then generates a complete result JSON file containing the original model output and detailed parsing information, then converts the data into a flat, symbol-delimited file, and finally generates a file containing the weighted aggregated comprehensive score. Optionally, the system writes the evaluation results to the evaluation platform database and prioritizes reading cached results from the database during subsequent evaluations.

[0071] The fitness course evaluation method provided in this application includes the following steps: acquiring structured course data for multiple fitness courses to be evaluated, wherein the structured course data includes at least a list of movements and course duration; generating evaluation prompts for each fitness course based on the structured course data and a preset prompt template; and evaluating each fitness course using multiple large-scale evaluation models based on the evaluation prompts to obtain an evaluation result for each fitness course, wherein the evaluation result includes at least quality level information and suggested remedial measures. This method solves the technical problem of low evaluation efficiency caused by relying on manual evaluation of fitness course content in related technologies.

[0072] In this solution, structured course data is acquired and evaluation prompts are generated. Multiple large-scale evaluation models are used in parallel to evaluate fitness courses, which improves evaluation efficiency, overcomes the problem of subjective and inconsistent standards, and enables batch and rapid processing of course content. Based on the quality level information and handling suggestions output by the multi-model comprehensive evaluation, the objectivity, consistency and interpretability of the evaluation results are improved.

[0073] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0074] Example 2

[0075] This application also provides a fitness course evaluation device. It should be noted that the fitness course evaluation device of this application embodiment can be used to execute the fitness course evaluation method provided in this application embodiment. The fitness course evaluation device provided in this application embodiment is described below.

[0076] According to an embodiment of this application, a fitness course evaluation device for implementing the above-described fitness course evaluation method is also provided, such as... Figure 3 As shown, the device includes: a first acquisition unit 301, a first processing unit 302, and a second processing unit 303.

[0077] The first acquisition unit 301 is used to acquire structured course data of multiple fitness courses to be evaluated, wherein the structured course data includes at least a list of movements and course duration.

[0078] The first processing unit 302 is used to generate evaluation prompts for each fitness course to be evaluated based on the structured course data and the preset prompt templates.

[0079] The second processing unit 303 is used to evaluate each fitness course to be evaluated using multiple evaluation models based on evaluation prompts, and to obtain the evaluation results corresponding to each fitness course to be evaluated. The evaluation results include at least quality level information and disposal suggestion information.

[0080] Optionally, in the fitness course evaluation device provided in this application embodiment, the second processing unit includes: a first processing subunit, used to send evaluation prompt words corresponding to each fitness course to be evaluated to each evaluation big model in parallel to obtain the original evaluation data returned by each evaluation big model; a second processing subunit, used to perform format processing on the original evaluation data returned by each evaluation big model to obtain multiple structured evaluation results corresponding to the fitness course to be evaluated; a third processing subunit, used to parse the multiple structured evaluation results and extract the scoring sub-items of multiple dimensions in each structured evaluation result; and a fourth processing subunit, used to calculate based on the scoring sub-items of multiple dimensions in each structured evaluation result to obtain the evaluation result corresponding to the fitness course to be evaluated.

[0081] Optionally, in the fitness course evaluation device provided in this application embodiment, the fourth processing subunit includes: a first processing module, used to perform weighted calculations on the scoring sub-items of each dimension for each structured evaluation result to obtain the scoring value corresponding to each dimension, and to calculate the total scoring value corresponding to the structured evaluation result based on the scoring value corresponding to each dimension; a second processing module, used to calculate the total scoring value corresponding to the fitness course to be evaluated based on the confidence level contained in each structured evaluation result and the total scoring value corresponding to each structured evaluation result; and a third processing module, used to determine the quality level information and handling suggestion information corresponding to the fitness course to be evaluated based on the total scoring value corresponding to the fitness course to be evaluated, and to determine the evaluation result corresponding to the fitness course to be evaluated based on the quality level information and handling suggestion information corresponding to the fitness course to be evaluated.

[0082] Optionally, in the fitness course evaluation device provided in this application embodiment, the device further includes: a second acquisition unit, configured to acquire a course identifier list before acquiring structured course data of multiple fitness courses to be evaluated, wherein the course identifier list contains identifiers of multiple original fitness courses to be evaluated; a third acquisition unit, configured to acquire the original course data of each original fitness course to be evaluated based on the course identifier list, and perform conversion processing on the original course data of each original fitness course to be evaluated to obtain structured course data of each original fitness course to be evaluated; and a determination unit, configured to verify the structured course data of each original fitness course to be evaluated, obtain a verification result, and determine multiple fitness courses to be evaluated from multiple original fitness courses to be evaluated based on the verification result.

[0083] Optionally, in the fitness course evaluation device provided in this application embodiment, the determining unit includes: a first determining subunit, used for determining the number of movements and the cumulative duration of movements for each original fitness course to be evaluated based on the movement list of the original fitness course to be evaluated, and calculating the duration deviation based on the course duration and cumulative duration of movements of the original fitness course to be evaluated; a comparison subunit, used for comparing the number of movements with a first threshold, and comparing the duration deviation with a second threshold; a second determining subunit, used for determining that if the number of movements is less than the first threshold, or the duration deviation is greater than or equal to the second threshold, the original fitness course to be evaluated fails the verification as a verification result; a third determining subunit, used for determining that if the number of movements is greater than or equal to the first threshold, and the duration deviation is less than the second threshold, the original fitness course to be evaluated passes the verification as a verification result.

[0084] Optionally, in the fitness course evaluation device provided in this application embodiment, the first processing unit includes: a fifth processing subunit, used to obtain a preset prompt word template for each fitness course to be evaluated, wherein the preset prompt word template contains placeholders used to represent information to be processed; a sixth processing subunit, used to construct a structured text string based on structured course data, wherein the structured text string is used to represent information to be processed; and a seventh processing subunit, used to fill the placeholders with the structured text string to generate the evaluation prompt words corresponding to the fitness course to be evaluated.

[0085] Optionally, in the fitness course evaluation device provided in this application embodiment, the device further includes: a third processing unit, used to serialize the evaluation results corresponding to each fitness course to be evaluated after obtaining the evaluation results corresponding to each fitness course to be evaluated, to obtain serialized data; and a fourth processing unit, used to write the serialized data into a checkpoint file, wherein the checkpoint file is used to record the completed evaluation results so as to perform recovery processing based on the checkpoint file after the program is interrupted.

[0086] It should be noted that the first acquisition unit 301, the first processing unit 302, and the second processing unit 303 mentioned above correspond to steps S201 to S203 in Embodiment 1. The three units and the corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above units can also be part of a device and run in the computer terminal 10 provided in Embodiment 1.

[0087] Example 3

[0088] Embodiments of this application may provide an electronic device. Figure 4 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 4 As shown, the electronic device may include: one or more ( Figure 4 (Only one is shown) Processor 402, memory 404, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0089] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the methods and apparatus in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the above-described methods. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0090] The processor can access information and applications stored in the memory via a transmission device to perform the following steps: acquire structured course data for multiple fitness courses to be evaluated, wherein the structured course data includes at least a list of movements and course duration; generate evaluation prompts for each fitness course to be evaluated based on the structured course data and preset prompt templates; evaluate each fitness course to be evaluated using multiple large evaluation models based on the evaluation prompts to obtain evaluation results for each fitness course to be evaluated, wherein the evaluation results include at least quality level information and remedial suggestions.

[0091] The processor can access information and applications stored in memory via a transmission device to execute the following steps: For each fitness course to be evaluated, send evaluation prompts corresponding to the fitness course to each large evaluation model in parallel to obtain the raw evaluation data returned by each large evaluation model; process the format of the raw evaluation data returned by each large evaluation model to obtain multiple structured evaluation results corresponding to the fitness course to be evaluated; parse the multiple structured evaluation results and extract the scoring sub-items of multiple dimensions in each structured evaluation result; calculate based on the scoring sub-items of multiple dimensions in each structured evaluation result to obtain the evaluation result corresponding to the fitness course to be evaluated.

[0092] The processor can access the information and application programs stored in the memory via the transmission device to execute the following steps: For each structured evaluation result, the scoring sub-items of each dimension are weighted and calculated to obtain the score value corresponding to each dimension, and the total score value corresponding to the structured evaluation result is calculated based on the score value corresponding to each dimension; the total score value corresponding to the fitness course to be evaluated is calculated based on the confidence level contained in each structured evaluation result and the total score value corresponding to each structured evaluation result; based on the total score value corresponding to the fitness course to be evaluated, the quality level information and handling suggestion information corresponding to the fitness course to be evaluated are determined, and the evaluation result corresponding to the fitness course to be evaluated is determined based on the quality level information and handling suggestion information corresponding to the fitness course to be evaluated.

[0093] The processor can access information and applications stored in memory via a transmission device to perform the following steps: Before acquiring structured course data for multiple fitness courses to be evaluated, acquire a list of course identifiers, which contains identifiers for multiple original fitness courses to be evaluated; based on the list of course identifiers, acquire the raw course data for each original fitness course to be evaluated, and perform transformation processing on the raw course data for each original fitness course to be evaluated to obtain structured course data for each original fitness course to be evaluated; verify the structured course data for each original fitness course to be evaluated, obtain the verification result, and determine multiple fitness courses to be evaluated from the multiple original fitness courses to be evaluated based on the verification result.

[0094] The processor can access information and applications stored in the memory via a transmission device to execute the following steps: For each original fitness course to be evaluated, determine the number of movements and the cumulative duration of each movement based on the movement list of the original fitness course to be evaluated, and calculate the duration deviation based on the course duration and cumulative duration of each movement; compare the number of movements with a first threshold, and compare the duration deviation with a second threshold; if the number of movements is less than the first threshold, or the duration deviation is greater than or equal to the second threshold, then the original fitness course to be evaluated fails the verification as the verification result; if the number of movements is greater than or equal to the first threshold, and the duration deviation is less than the second threshold, then the original fitness course to be evaluated passes the verification as the verification result.

[0095] The processor can access the information and application stored in the memory via the transmission device to perform the following steps: For each fitness course to be evaluated, obtain a preset prompt word template, wherein the preset prompt word template contains placeholders used to represent the information to be processed; construct a structured text string based on the structured course data, wherein the structured text string is used to represent the information to be processed; fill the placeholders with the structured text string to generate the evaluation prompt words corresponding to the fitness course to be evaluated.

[0096] The processor can access the information and application stored in the memory via the transmission device to perform the following steps: after obtaining the evaluation results for each fitness course to be evaluated, serialize the evaluation results for each fitness course to be evaluated to obtain serialized data; write the serialized data into a checkpoint file, wherein the checkpoint file is used to record the completed evaluation results so that recovery processing can be performed based on the checkpoint file after the program is interrupted.

[0097] Those skilled in the art will understand that Figure 4The structure shown is for illustrative purposes only. Electronic devices can also be smartphones, tablets, handheld computers, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 4 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 4 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 4 The different configurations shown.

[0098] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0099] Example 4

[0100] Embodiments of this application also provide a computer-readable storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the fitness course evaluation method provided in Embodiment 1.

[0101] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0102] This application also provides a computer program product, which, when executed on a data processing device, is suitable for performing the steps of an assessment method for a fitness course.

[0103] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0104] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0105] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0106] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0107] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0108] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0109] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for evaluating fitness courses, characterized in that, include: Obtain structured course data from multiple fitness courses to be evaluated, wherein the structured course data includes at least a list of exercises and course duration; Based on the structured course data and preset prompt word templates, generate evaluation prompt words for each fitness course to be evaluated; Multiple evaluation models are used to evaluate each fitness course to be evaluated based on the evaluation prompts, and evaluation results are obtained for each fitness course to be evaluated. The evaluation results include at least quality level information and handling suggestions.

2. The method according to claim 1, characterized in that, Multiple evaluation models are used to evaluate each fitness course to be evaluated based on the evaluation prompts, and the evaluation results for each fitness course to be evaluated include: For each fitness course to be evaluated, the evaluation prompt words corresponding to the fitness course to be evaluated are sent to each evaluation model in parallel to obtain the original evaluation data returned by each evaluation model. The raw evaluation data returned by each evaluation model is formatted to obtain multiple structured evaluation results corresponding to the fitness course to be evaluated. The multiple structured evaluation results are analyzed, and the scoring sub-items of multiple dimensions in each structured evaluation result are extracted; The evaluation result for the fitness course to be evaluated is obtained by calculating the scores of multiple dimensions in each structured evaluation result.

3. The method according to claim 2, characterized in that, The evaluation results for the fitness course under evaluation are calculated based on the scoring sub-items of multiple dimensions in each structured evaluation result, including: For each structured evaluation result, the scoring sub-items of each dimension are weighted and calculated to obtain the score value corresponding to each dimension. Based on the score value corresponding to each dimension, the total score value corresponding to the structured evaluation result is calculated. The total score of the fitness course to be evaluated is calculated based on the confidence level contained in each structured evaluation result and the total score corresponding to each structured evaluation result. Based on the total score of the fitness course to be evaluated, determine the quality level information and handling suggestions for the fitness course to be evaluated, and based on the quality level information and handling suggestions for the fitness course to be evaluated, determine the evaluation result for the fitness course to be evaluated.

4. The method according to claim 1, characterized in that, Before acquiring structured course data from multiple fitness courses to be evaluated, the method further includes: Obtain a list of course identifiers, wherein the list of course identifiers contains identifiers of multiple original fitness courses to be evaluated; Based on the course identifier list, the original course data of each original fitness course to be evaluated is obtained, and the original course data of each original fitness course to be evaluated is transformed to obtain the structured course data of each original fitness course to be evaluated. The structured course data of each original fitness course to be evaluated is verified to obtain the verification result, and the multiple fitness courses to be evaluated are determined from the multiple original fitness courses to be evaluated based on the verification result.

5. The method according to claim 4, characterized in that, The structured course data for each original fitness course to be evaluated were validated, and the validation results include: For each original fitness course to be evaluated, the number of movements and the cumulative duration of the movements are determined based on the movement list of the original fitness course to be evaluated, and the duration deviation is calculated based on the course duration of the original fitness course to be evaluated and the cumulative duration of the movements. The number of actions is compared with a first threshold, and the duration deviation is compared with a second threshold; If the number of actions is less than the first threshold, or the duration deviation is greater than or equal to the second threshold, then the original fitness course to be evaluated fails the verification as the verification result. If the number of actions is greater than or equal to the first threshold, and the duration deviation is less than the second threshold, then the original fitness course to be evaluated passes the verification and is taken as the verification result.

6. The method according to claim 1, characterized in that, Based on the structured course data and preset prompt word templates, the evaluation prompt words for each fitness course to be evaluated are generated, including: For each fitness course to be evaluated, the preset prompt word template is obtained, wherein the preset prompt word template contains placeholders used to represent the information to be processed; A structured text string is constructed based on the structured course data, wherein the structured text string is used to represent the information to be processed; The structured text string is filled into the placeholder to generate the evaluation prompt words corresponding to the fitness course to be evaluated.

7. The method according to claim 1, characterized in that, After obtaining the evaluation results for each fitness course to be evaluated, the method further includes: The evaluation results for each fitness course to be evaluated are serialized to obtain serialized data. The serialized data is written to a checkpoint file, which records the completed evaluation results so that recovery processing can be performed based on the checkpoint file after the program is interrupted.

8. A fitness course evaluation device, characterized in that, include: The first acquisition unit is used to acquire structured course data of multiple fitness courses to be evaluated, wherein the structured course data includes at least a list of movements and course duration. The first processing unit is used to generate evaluation prompts for each fitness course to be evaluated based on the structured course data and the preset prompt template. The second processing unit is used to evaluate each fitness course to be evaluated using multiple evaluation models based on the evaluation prompts, and to obtain the evaluation results corresponding to each fitness course to be evaluated. The evaluation results include at least quality level information and handling suggestion information.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the computer-readable storage medium is located to perform the evaluation method of the fitness course according to any one of claims 1 to 7.

10. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, executes the evaluation method for a fitness course according to any one of claims 1 to 7.

Images