A teaching large model system based on a select-raise-evaluate three-dimensional ability portrait

Abstract

The application discloses a programming teaching large model system based on selection-cultivation-evaluation three-dimensional ability image, a three-dimensional interpretable evaluation system is constructed, and instead of saving single homework results, student ability is divided into selection dimension, cultivation dimension and evaluation dimension, a continuously updated three-dimensional ability image is formed, teachers can see the continuous changes of students in the three aspects of "can they be selected out, can they be cultivated well and how do they finally perform", the system is more suitable for cultivating top-notch innovative talents, student codes are evaluated in multiple dimensions, teaching data is fully utilized, the burden of teachers is reduced, and personalized guidance can be quickly provided in large-class teaching.

Description

Technical Field

[0001] This invention belongs to the field of programming teaching technology, and more specifically, it relates to a large-scale programming teaching model system based on a three-dimensional ability profile of selection, training, and assessment. Background Technology

[0002] Programming teaching systems mainly include the following categories:

[0003] The first category is online programming teaching platforms, such as LeetCode and Nowcoder. These systems typically include modules for problem posting, online editors, code submission, test case execution, and result evaluation. After students submit their code, the system uses pre-defined test cases to determine if it passes, providing results such as "correct / incorrect" or "pass / fail." This evaluation is singular, relying primarily on test cases to judge code performance, and usually only provides "correct / incorrect" or "pass / fail" results. It's difficult to evaluate student code from multiple dimensions such as correctness, efficiency, readability, complexity, security, best practices, and learning suggestions.

[0004] The second type is code evaluation systems based on rules or static analysis. These programming teaching systems automatically check student code through predefined syntax rules, code style rules, safety rules, or complexity rules. However, they lack combined analysis of code semantics and task requirements, and can only find fixed format or rule problems. They are unable to determine whether student code truly meets the task intent, nor can they explain the algorithm's thinking, boundary conditions, and potential improvement directions.

[0005] The third category consists of single, large-scale programming assistants, such as code completion, code Q&A, error explanation, and code generation tools. These programming teaching tools can generate explanations and suggestions based on natural language or code context. However, these deep learning-based teaching aids typically require large amounts of labeled data, making them difficult to deploy and apply quickly in real-world teaching environments.

[0006] The fourth category is traditional programming teaching management systems, such as platforms for course management, class management, homework submission, teacher grading, grade registration, and user management. In these systems, teachers face a heavy grading burden. Faced with a large number of programming assignments, teachers need to review each piece of code, determine the reasons for errors, write feedback, and register grades, making it difficult to provide quick, personalized guidance in large classes.

[0007] Furthermore, existing programming education systems have low utilization rates of teaching data, often only storing assignment scores, submission counts, or submission times, without integrating code submissions, learning time, attention span, emotional state, interactive behavior, AI feedback, and teacher evaluations into a computable chain of teaching evidence. Roles are fragmented, with separate functions for student and teacher ends, lacking a unified platform design. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of the prior art and provide a large-scale programming teaching model system based on a three-dimensional ability profile of selection, training, and evaluation. This system evaluates student code from multiple dimensions, makes full use of teaching data, reduces the burden of grading for teachers, and provides rapid personalized guidance in large-class teaching.

[0009] To achieve the above-mentioned objectives, this invention provides a large-scale programming teaching model system based on a three-dimensional competency profile of selection, training, and assessment, characterized by comprising:

[0010] The three-dimensional capability profile module for selection, training, and evaluation consists of a capability representation layer, a behavioral decision-making layer, and a feedback optimization layer, among which:

[0011] The input to the capability representation layer is the student. exist Characteristics of code submitted at a given time Behavior log And feedback on AI analysis results and teacher review results. The capability representation layer uses embedded functions The input consists of three types of heterogeneous data: code features Behavior log and feedback Integrate into a unified student ability vector :

[0012]

[0013] in, Indicated by code characteristics The resulting code feature vectors contain characteristics related to correctness, efficiency, readability, complexity, and security risks. Indicates from behavior log The resulting learning behavior sub-vectors contain features such as learning duration, number of submissions, behavior type, and attention score. Indicates feedback The transformed feedback evaluation subvector contains features of AI feedback scores, teacher scores, and teacher comments;

[0014] The behavioral decision layer receives the student ability vector output by the ability representation layer. Extract the average official score from it. Average attention score and participation in complex tasks Input the selection function to obtain the selection dimension scores. :

[0015]

[0016] Extract learning time contribution points Contribution of Diversity in Learning Behaviors and contribution to learning and growth Input the breeding function to obtain the breeding dimension score. :

[0017]

[0018] Among them, the contribution score of learning time for:

[0019]

[0020] in, Students The study time over the past 7 days, in seconds. 7200 represents the number of seconds corresponding to 2 hours. This indicates that the highest possible score for this item is 40.

[0021] Contribution score of learning behavior diversity for:

[0022]

[0023] in, Students The number of learning behavior types that appeared in the last 7 days can be scored out of a maximum of 4 categories. This indicates that the maximum score for this item is 30 points;

[0024] Learning and growth contribution points are defined as follows:

[0025]

[0026] in, Students The extent of progress, the adoption of feedback, or the continuous submission of improvements will be counted as a maximum of 30 points.

[0027] Extract code scores from the last 30 days Teacher comprehensive evaluation and the completion rate of complex tasks Input the evaluation function to obtain the scores for each evaluation dimension. :

[0028]

[0029] Forming students exist The 3D capability profile vector at each moment:

[0030]

[0031] Get students exist The comprehensive ability at any given moment is divided into:

[0032]

[0033] in, , , These represent the weights of the selection, training, and evaluation dimensions, respectively. ;

[0034] The feedback optimization layer constructs a closed-loop loss function based on the scores from the selection dimension, the training dimension, and the evaluation dimension:

[0035]

[0036]

[0037] in, Represents the total loss function; This indicates the loss in the selection dimension; This indicates a loss in the nurturing dimension; Indicates the loss in the evaluation dimension. , , This represents the weights of the three loss terms, and , express The image update parameters at each time step are the embedding functions. Network parameters, Indicates the update step size. Represents the loss function The gradient;

[0038] use Network parameters at time Update Embedded Function Network parameters.

[0039] The objective of this invention is achieved as follows.

[0040] This invention is a large-scale programming teaching model system based on a three-dimensional competency profile of selection, training, and evaluation. By constructing a three-dimensional interpretable evaluation system of "selection-training-evaluation," it no longer only saves the scores of a single assignment, but divides students' abilities into selection, training, and evaluation dimensions, forming a continuously updated three-dimensional competency profile. Teachers can see the continuous changes of students in three aspects: "whether they can be selected, whether they can be trained well, and how they ultimately perform." This is more suitable for cultivating top-notch innovative talents, thus evaluating students' code from multiple dimensions, making full use of teaching data, reducing the burden of grading for teachers, and enabling the rapid provision of personalized guidance in large-class teaching. Attached Figure Description

[0041] Figure 1 This is a general architecture diagram of a specific implementation of the programming teaching model system based on a three-dimensional ability profile of selection, training, and assessment according to the present invention;

[0042] Figure 2 This is a three-layer architecture diagram of the selection-training-evaluation three-dimensional ability profiling module in this invention;

[0043] Figure 3 This is a closed-loop flowchart of the programming teaching model system based on the three-dimensional ability profile of selection, training, and assessment, as described in this invention. Detailed Implementation

[0044] The specific embodiments of the present invention will now be described with reference to the accompanying drawings to enable those skilled in the art to better understand the invention. It should be particularly noted that in the following description, detailed descriptions of known functions and designs that might obscure the main content of the invention will be omitted here.

[0045] Figure 1 This is a general architecture diagram of a specific implementation of the programming teaching model system based on the three-dimensional ability profile of selection, training and assessment of the present invention.

[0046] In this embodiment, as Figure 2 As shown, the programming teaching model system based on the selection-education-assessment three-dimensional ability profile of this invention uses a teaching database as a unified data foundation and a role-adaptive access module as the entry point. It forms a complete data flow loop around student programming task submission, structured analysis of the educational model, teacher review and evaluation, learning behavior recording, and updates to the selection-education-assessment three-dimensional ability profile. The data flow of the programming teaching model system based on the selection-education-assessment three-dimensional ability profile of this invention is as follows:

[0047] Users access the system through user terminals. The role-adaptive access module authenticates user identities and loads either the student-side or teacher-side functional views based on the user's role. Teacher-side users create courses, classes, and programming tasks through the course, class, and task management module, and task data is written to the teaching database. Student-side users read their assigned courses, classes, and programming tasks from the teaching database and complete code writing, running, and submission through the programming task and code submission module.

[0048] After students submit their code, the programming task and code submission module writes data such as code content, programming language, task number, submission time, execution result, error message, execution time, and memory usage into the teaching database. Simultaneously, the learning behavior log module records data about the student's submission behavior, learning duration, attention score, and interaction behavior.

[0049] When the system triggers AI analysis, the structured code semantic analysis module reads information such as student code, task description, and programming language from the teaching database, constructs structured prompt words, and calls the interface of the local model, remote private model, or cloud-compatible model through the education big model service adaptation module. After the education big model returns the structured analysis results, the system saves the results, including scores, correctness analysis, efficiency analysis, readability analysis, complexity analysis, best practice suggestions, advantages, improvement suggestions, learning suggestions, and safety tips, to the teaching database.

[0050] Teachers can access student code submission records and AI analysis reports through the batch analysis and teacher review modules. They can perform batch AI analysis on single or multiple submissions and input teacher scores and comments after viewing the AI ​​analysis results. The system saves AI evaluations and teacher evaluations separately. AI evaluations serve as a reference for teacher review and student feedback, while the teacher-reviewed scores are written into the teaching database as official scores.

[0051] The Selection-Development-Assessment 3D Ability Profile module reads code scores, teacher evaluations, learning behavior logs, attention scores, participation in complex tasks, and historical submission records from the teaching database. It calculates scores for the selection, development, and assessment dimensions of each student and generates a comprehensive ability profile. The Radar Chart and Teaching Early Warning Display module reads the 3D profile results and AI analysis results to show students personalized learning suggestions and to show teachers the class's learning status, low-score warnings, and teaching intervention references.

[0052] The AI ​​dialogue assistance module supports students in engaging in natural language question-and-answer sessions regarding programming tasks, error causes, knowledge points, and learning suggestions. This module reads current task information, student submission records, and AI analysis results, and generates student-specific assistance answers through the educational big data model service adaptation module. The interaction records can be further written to the learning behavior log module.

[0053] Specifically, in this embodiment, as Figure 1 As shown, the large-scale programming teaching model system based on the three-dimensional ability profile of selection, education, and assessment of this invention includes: user terminal 1, role adaptive access module 2, student-side functional view module 3, programming task and code submission module 4, learning behavior log module 5, teaching database 6, three-dimensional ability profile of selection, education, and assessment module 7, teacher-side functional view module 8, radar chart and teaching early warning display module 9, AI dialogue assistance module 10, batch analysis and teacher review module 11, education large-scale model service adaptation module 12, structured code semantic analysis module 13, and course class and task management module 14.

[0054] User terminal 1 is used by students and teachers to access the system. It is responsible for submitting login requests, displaying the student or teacher interface, and receiving system feedback results. User terminals include browsers or other client entry points.

[0055] The role-adaptive access module 2 is used to complete user authentication, role recognition and access control. It automatically loads the student-side function view or the teacher-side function view according to the user role when the user terminal 1 submits the login request, ensuring that different roles can only access courses, tasks, submission records and evaluation data within their permission scope.

[0056] The student-side functional view module 3 is used to display students' personal learning data, course and class information, programming task list, code submission entry, AI dialogue assistance entry, AI analysis report, learning suggestions, and a three-dimensional ability profile of selection, education, and evaluation.

[0057] The Programming Task and Code Submission Module 4 is used to display programming tasks, receive code for programming tasks submitted through the code submission entry of the student terminal functional view module 3, record code submission data, and save programming language, task number, code submission time, running result, error message, running time and memory usage information.

[0058] The learning behavior log module 5 is used to record students' learning process data, including code submission, video viewing, material reading, task interaction, experimental operation, learning time, attention score and behavior type, providing process data for the calculation of three-dimensional ability profile and storing it in the teaching database 6.

[0059] The teaching database 6 is used to uniformly store courses, classes, tasks, users, code submissions, running results, AI analysis reports, teacher evaluations, learning behavior logs, 3D ability profile results, and early warning records, providing unified support for data flow between modules.

[0060] The three-dimensional ability profile module 7 for selection, training, and evaluation is used to calculate the scores of students in the selection, training, and evaluation dimensions based on the students' code scores, teacher evaluations, learning time, attention scores, learning behavior diversity, participation in complex tasks, and historical performance stored in the education database 6, and to form a continuously updated three-dimensional ability profile of students in selection, training, and evaluation.

[0061] like Figure 2 As shown, the selection-training-evaluation three-dimensional capability profiling module consists of a capability representation layer 701, a behavior decision layer 702, and a feedback optimization layer 703, wherein:

[0062] The input to the capability representation layer 701 is the student. exist Characteristics of code submitted at a given time Behavior log And feedback on AI analysis results and teacher review results. The capability representation layer uses embedded functions The input consists of three types of heterogeneous data: code features Behavior log and feedback Integrate into a unified student ability vector :

[0063]

[0064] in, Indicated by code characteristics The resulting code feature vectors contain characteristics related to correctness, efficiency, readability, complexity, and security risks. Indicates from behavior log The resulting learning behavior sub-vectors contain features such as learning duration, number of submissions, behavior type, and attention score. Indicates feedback The resulting feedback evaluation subvector contains features of AI feedback scores, teacher scores, and teacher comments.

[0065] The behavioral decision layer receives the student ability vector output by the ability representation layer. Extract the average official score from it. Average attention score and participation in complex tasks Input the selection function to obtain the selection dimension scores. :

[0066]

[0067] Extract learning time contribution points Contribution of Diversity in Learning Behaviors and contribution to learning and growth Input the breeding function to obtain the breeding dimension score. :

[0068]

[0069] Among them, the contribution score of learning time for:

[0070]

[0071] in, Students The study time over the past 7 days, in seconds. 7200 represents the number of seconds corresponding to 2 hours. This indicates that the highest possible score for this item is 40.

[0072] Contribution score of learning behavior diversity for:

[0073]

[0074] in, Students The number of learning behavior types that appeared in the last 7 days can be scored out of a maximum of 4 categories. This indicates that the maximum score for this item is 30 points;

[0075] Learning and growth contribution points are defined as follows:

[0076]

[0077] in, Students The extent of progress, the adoption of feedback, or the continuous submission of improvements will be counted as a maximum of 30 points.

[0078] Extract code scores from the last 30 days Teacher comprehensive evaluation and the completion rate of complex tasks Input the evaluation function to obtain the scores for each evaluation dimension. :

[0079]

[0080] Forming students exist The 3D capability profile vector at each moment:

[0081]

[0082] Get students exist The comprehensive ability at any given moment is divided into:

[0083]

[0084] in, , , These represent the weights of the selection, training, and evaluation dimensions, respectively. .

[0085] Feedback optimization layer 703 constructs a closed-loop loss function based on the selection dimension score, training dimension score, and evaluation dimension score:

[0086]

[0087]

[0088] in, Represents the total loss function; This indicates the loss in the selection dimension; This indicates a loss in the nurturing dimension; Indicates the loss in the evaluation dimension. , , This represents the weights of the three loss terms, and , express The image update parameters at each time step are the embedding functions. Network parameters, Indicates the update step size. Represents the loss function The gradient.

[0089] use Network parameters at time Update Embedded Function Network parameters.

[0090] The teacher-side functional view module 8 is used to display functions such as course management, class management, task posting, student submission records, AI analysis reports, teacher review and evaluation, class profiles, and teaching warnings to teachers.

[0091] The radar chart and teaching early warning display module 9 is used to display students' three-dimensional ability profiles to students and teachers in the form of radar charts, trend charts or early warning prompts: students can view their own three-dimensional ability profiles and learning suggestions through the student-side function view, and teachers can view the overall three-dimensional ability profiles of the class, early warnings for low-scoring students and teaching intervention references through the teacher-side function view.

[0092] The AI ​​dialogue assistance module 10 is used by students to provide natural language question-and-answer services related to programming tasks, code errors, explanations of knowledge points, and learning suggestions through the student-side functional view module 3. This module can generate context-related assistance answers by combining the current task, student code, and AI analysis results.

[0093] The batch analysis and teacher review module 11 is used to support teachers to perform batch AI analysis on multiple student code submissions and avoid model service overload through concurrency control. At the same time, this module allows teachers to view student code, AI analysis reports and AI reference scores, and input official teacher scores and teacher comments.

[0094] The Education Large Model Service Adaptation Module 12 is used to uniformly access different types of large model services, including local Ollam models, remote private model services accessed via SSH port forwarding, and OpenAI-compatible cloud interfaces. This module provides a unified calling interface to the AI ​​dialogue assistance module 10 and the batch analysis and teacher review module 11, while shielding the differences in request and return formats of different model services from the bottom up.

[0095] The structured code semantic analysis module 13 is used to construct prompt words based on the student code, task description and programming language submitted by the teacher through the batch analysis and teacher review module 11, and call the education big model to generate structured code analysis results. The structured code analysis results include at least the score, correctness, efficiency, readability, complexity analysis, best practices, advantages, improvement suggestions, learning suggestions and safety tips. The structured code analysis results are stored in the teaching database 6.

[0096] The Course Class and Task Management Module 14 is used by teachers to manage courses, classes, class codes, teachers, students, and programming training tasks. Teachers can use this module to create courses and classes, publish programming tasks, and configure task descriptions, test cases, deadlines, difficulty, and visibility information.

[0097] Figure 3 This is a closed-loop flowchart of the programming teaching model system based on the three-dimensional competency profile of the present invention. The invention takes student programming practice as the core scenario, unifying courses, classes, programming tasks, student code submissions, learning behavior logs, educational model analysis results, and teacher evaluation results into a single teaching data foundation, forming a closed loop of "student code submission - structured analysis by the educational model - teacher review and evaluation - student feedback and improvement - continuous updating of competency profiles".

[0098] The core difference between this invention and existing large-model-based programming assistants lies in:

[0099] 1. A three-dimensional interpretable evaluation system of selection, training, and assessment has been constructed. Instead of simply storing single assignment scores, it divides student abilities into selection, training, and assessment dimensions, forming a continuously updated three-dimensional ability profile. Teachers can see the continuous changes in students across the three levels of "whether they can be selected, whether they can be trained well, and how they ultimately perform," making it more suitable for cultivating top-notch innovative talents.

[0100] 2. Form a closed-loop optimization by connecting student submissions, AI analysis, teacher review, student improvement, and profile updates into a closed loop, so that every student submission and feedback can become a data basis for subsequent evaluation and teaching intervention.

[0101] Although the illustrative specific embodiments of the present invention have been described above to enable those skilled in the art to understand the invention, it should be understood that the invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the invention as defined and determined by the appended claims, and all inventions utilizing the concept of the present invention are protected.

Claims

1. A large-scale programming teaching model system based on a three-dimensional ability profile of selection, training, and assessment, characterized in that: include: The three-dimensional capability profile module for selection, training, and evaluation consists of a capability representation layer, a behavioral decision-making layer, and a feedback optimization layer, among which: The input to the capability representation layer is the student. exist Characteristics of code submitted at any given time Behavior log And feedback on AI analysis results and teacher review results. The capability representation layer uses embedded functions The input consists of three types of heterogeneous data: code features Behavior log and feedback Integrate into a unified student ability vector : ； in, Indicated by code characteristics The resulting code feature vectors contain characteristics related to correctness, efficiency, readability, complexity, and security risks. Indicates from behavior log The resulting learning behavior sub-vectors contain features such as learning duration, number of submissions, behavior type, and attention score. Indicates feedback The transformed feedback evaluation subvector contains features of AI feedback scores, teacher scores, and teacher comments; The behavioral decision layer receives the student ability vector output by the ability representation layer. Extract the average official score from it. Average attention score and participation in complex tasks Input the selection function to obtain the selection dimension scores. : ； Extract learning time contribution points Contribution of Diversity in Learning Behaviors and contribution to learning and growth Input the breeding function to obtain the breeding dimension score. : ； Among them, the contribution score of learning time for: ； in, Students The study time over the past 7 days, in seconds. 7200 represents the number of seconds corresponding to 2 hours. This indicates that the highest possible score for this item is 40. Contribution score of learning behavior diversity for: ； in, Students The number of learning behavior types that appeared in the last 7 days can be scored out of a maximum of 4 categories. This indicates that the maximum score for this item is 30 points; Learning and growth contribution points are defined as follows: ； in, Students The extent of progress, the adoption of feedback, or the continuous submission of improvements will be counted as a maximum of 30 points. Extract code scores from the last 30 days Teacher comprehensive evaluation and the completion rate of complex tasks Input the evaluation function to obtain the scores for each evaluation dimension. : ； Forming students exist The 3D capability profile vector at each moment: ； Get students exist The comprehensive ability at any given moment is divided into: ； in, , , These represent the weights of the selection, training, and evaluation dimensions, respectively. ; The feedback optimization layer constructs a closed-loop loss function based on the scores from the selection dimension, the training dimension, and the evaluation dimension: ； ； in, Represents the total loss function; This indicates the loss in the selection dimension; This indicates a loss in the nurturing dimension; Indicates the loss in the evaluation dimension. , , This represents the weights of the three loss terms, and , express The image update parameters at each time step are the embedding functions. Network parameters, Indicates the update step size. Represents the loss function The gradient; use Network parameters at time Update Embedded Function Network parameters.

2. The large-scale programming teaching model system based on a three-dimensional ability profile of selection, training, and assessment as described in claim 1, is characterized in that, Also includes: The user terminal is used by students and teachers to access the system. It is responsible for submitting login requests, displaying the student or teacher interface, and receiving system feedback results. The role-adaptive access module is used to complete user authentication, role recognition and access control. It automatically loads the student-side function view or the teacher-side function view according to the user role when the user terminal submits the login request, ensuring that different roles can only access courses, tasks, submission records and evaluation data within their authorized scope. The student-side functional view module is used to display students' personal learning data, course and class information, programming task list, code submission entry, AI dialogue assistance entry, AI analysis report, learning suggestions, and a three-dimensional ability profile of selection, education, and evaluation; The programming task and code submission module is used to display programming tasks, receive code submitted through the code submission entry of the student terminal functional view module, record code submission data, and save programming language, task number, code submission time, running result, error message, running time and memory usage information; The learning behavior log module is used to record students' learning process data, including code submission, video viewing, material reading, task interaction, experimental operation, learning time, attention score and behavior type, to provide process data for the calculation of three-dimensional ability profile and store it in the teaching database; The teaching database is used to uniformly store courses, classes, tasks, users, code submissions, running results, AI analysis reports, teacher evaluations, learning behavior logs, 3D ability profile results, and early warning records, providing unified support for data flow between modules; The teacher-side functional view module is used to display functions such as course management, class management, task posting, student submission records, AI analysis reports, teacher review and evaluation, class profiles, and teaching warnings to teachers. The radar chart and teaching early warning display module is used to display students' three-dimensional ability profiles to students and teachers in the form of radar charts, trend charts or early warning prompts: students can view their own three-dimensional ability profiles and learning suggestions through the student-side function view, and teachers can view the overall three-dimensional ability profiles of the class, early warnings for low-scoring students and teaching intervention references through the teacher-side function view. The AI ​​dialogue assistance module provides students with natural language question-and-answer services related to programming tasks, code errors, explanations of knowledge points, and learning suggestions through the student-side functional view module. This module can generate context-related assistance answers by combining the current task, student code, and AI analysis results. The batch analysis and teacher review module supports teachers in performing batch AI analysis on multiple student code submissions and avoids model service overload through concurrency control. At the same time, this module allows teachers to view student code, AI analysis reports and AI reference scores, and input official teacher scores and teacher comments. The Education Large Model Service Adaptation Module is used to uniformly access different types of large model services, including local Ollam models, remote private model services accessed via SSH port forwarding, and OpenAI-compatible cloud interfaces. This module provides a unified calling interface for the AI ​​dialogue assistance module, batch analysis and teacher review module, and shields the differences in request and return formats of different model services from the bottom up. The structured code semantic analysis module is used to construct prompt words based on student code, task descriptions, and programming languages ​​submitted by teachers through the batch analysis and teacher review module. It calls the education big model to generate structured code analysis results. The structured code analysis results include at least a score, correctness, efficiency, readability, complexity analysis, best practices, advantages, improvement suggestions, learning suggestions, and safety tips. The structured code analysis results are stored in the teaching database. The course, class, and task management module is used by teachers to manage courses, classes, class codes, teachers, students, and programming training tasks. Teachers can use this module to create courses and classes, publish programming tasks, and configure task descriptions, test cases, deadlines, difficulty levels, and visibility information.

Images