Skip to main content

AtlasML Models

Module: atlasml/models/competency.py. These Pydantic models define the API contracts and core DTOs. They are validated automatically by FastAPI and rendered in OpenAPI, enabling type-safe client generation and consistent error handling.

Core Types​

  • OperationType: UPDATE | DELETE. Indicates whether a request should upsert or remove records. Use UPDATE for both creating and modifying entries.

  • Competency. Represents a single competency item identified within a course. It is the core unit for suggestions, relations, and persistence.

    • id: int
    • title: str
    • description: Optional[str]
    • course_id: int

  • ExerciseWithCompetencies. Captures an exercise artifact and optional competency links. Useful when associating learning content with one or more competencies.

    • id: int
    • title: str
    • description: str
    • competencies: Optional[list[int]]
    • course_id: int

  • SemanticCluster. Describes a group identifier, its course context, and a vector embedding. This can support clustering and labeling workflows.

    • cluster_id: str
    • course_id: int
    • vector_embedding: list[float]

Request/Response DTOs​

  • SuggestCompetencyRequest. Input payload for retrieving recommended competencies. Provide a natural-language description and the course_id for contextualization.

    • description: str
    • course_id: int

  • SuggestCompetencyResponse. Output with a ranked list of Competency objects. Consumers can display, filter, or post-process suggestions as needed.

    • competencies: list[Competency]

  • SaveCompetencyRequest. Wrapper for saving either a list of competencies, a single exercise, or both, along with an operation_type. This unifies create/update/delete behaviors behind a single endpoint.

    • competencies: Optional[list[Competency]]
    • exercise: Optional[ExerciseWithCompetencies]
    • operation_type: OperationType

  • RelationType: MATCHES | EXTENDS | REQUIRES. Encodes the semantics of edges in a competency graph (e.g., prerequisites or hierarchical relationships).

  • CompetencyRelation. Directed edge from tail_id to head_id with an associated relation_type. Used to represent candidate or curated links among competencies.

    • tail_id: int
    • head_id: int
    • relation_type: RelationType

  • CompetencyRelationSuggestionResponse. Collection of proposed relations for a course. Downstream systems may visualize, validate, or persist these links.