Types
- class afterimage.types.Conversation(*, conversations: ~typing.List[~afterimage.types.ConversationEntry], metadata: dict[str, ~typing.Any] = <factory>)[source]
Bases:
BaseModel- conversations: List[ConversationEntry]
- metadata: dict[str, Any]
- model_config = {}
Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
- class afterimage.types.ConversationEntry(*, role: Role, content: str, reasoning_content: str | None = None)[source]
Bases:
BaseModel- content: str
- model_config = {}
Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
- reasoning_content: str | None
- class afterimage.types.ConversationWithContext(*, conversations: ~typing.List[~afterimage.types.ConversationEntry], metadata: dict[str, ~typing.Any] = <factory>, instruction_context: str | None = None, response_context: str | None = None, persona: str | None = None)[source]
Bases:
Conversation- instruction_context: str | None
- model_config = {}
Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
- persona: str | None
- response_context: str | None
- class afterimage.types.Document(*, id: str = <factory>, text: str | None = None, personas: list[PersonaEntry] = <factory>, metadata: dict[str, ~typing.Any]=<factory>)[source]
Bases:
BaseModel- id: str
- metadata: dict[str, Any]
- model_config = {}
Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
- personas: list[PersonaEntry]
- text: str | None
- class afterimage.types.EvaluatedConversationWithContext(*, conversations: ~typing.List[~afterimage.types.ConversationEntry], metadata: dict[str, ~typing.Any] = <factory>, instruction_context: str | None = None, response_context: str | None = None, persona: str | None = None, evaluation: ~afterimage.types.EvaluationSchema | None = None, final_score: float | None = 0.0)[source]
Bases:
ConversationWithContext- evaluation: EvaluationSchema | None
- final_score: float | None
- model_config = {}
Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
- class afterimage.types.EvaluationEntrySchema(*, feedback: str, score: float)[source]
Bases:
BaseModel- feedback: str
- model_config = {}
Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
- score: float
- class afterimage.types.EvaluationSchema(*, coherence: EvaluationEntrySchema, factuality: EvaluationEntrySchema, grounding: EvaluationEntrySchema, helpfulness: EvaluationEntrySchema, relevance: EvaluationEntrySchema, overall_grade: GradeSchema)[source]
Bases:
BaseModel- coherence: EvaluationEntrySchema
- factuality: EvaluationEntrySchema
- grounding: EvaluationEntrySchema
- helpfulness: EvaluationEntrySchema
- model_config = {}
Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
- overall_grade: GradeSchema
- relevance: EvaluationEntrySchema
- class afterimage.types.GeneratedResponsePrompt(*, prompt: str, context: str | None = None, metadata: dict[str, ~typing.Any]=<factory>)[source]
Bases:
BaseModelOutput of RespondentPromptModifier.
- context: str | None
- metadata: dict[str, Any]
- model_config = {}
Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
- prompt: str
- class afterimage.types.GenerationMetadata[source]
Bases:
TypedDictStructured metadata for generation samples.
- batch_id: str
- context_id: str
- context_ids: list[str]
- instruction_index: int
- persona_generation_depth: int | None
- persona_name: str | None
- session_id: str
- class afterimage.types.GenerationState(num_generated: int = 0, num_requested: int = 0, start_time: datetime = <factory>, last_item: BaseModel | None = None, monitor: Any | None = None, metadata: Dict[str, ~typing.Any]=<factory>, stop_event: Any | None = None, context_counts: Dict[str, int]=<factory>, unique_personas: list[str] = <factory>)[source]
Bases:
objectCurrent state of the generation process.
- context_counts: Dict[str, int]
- last_item: BaseModel | None = None
- metadata: Dict[str, Any]
- monitor: Any | None = None
- num_generated: int = 0
- num_requested: int = 0
- start_time: datetime
- stop_event: Any | None = None
- unique_personas: list[str]
- class afterimage.types.GradeSchema(value)[source]
Bases:
str,Enum- BAD = 'bad'
- GOOD = 'good'
- NEEDS_IMPROVEMENT = 'needs_improvement'
- NOT_ACCEPTABLE = 'not_acceptable'
- PERFECT = 'perfect'
- class afterimage.types.PersonaEntry(*, descriptions: list[str], metadata: dict = <factory>)[source]
Bases:
BaseModelRepresents a set of generated personas for a source document.
- descriptions: list[str]
- metadata: dict
- model_config = {}
Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
- class afterimage.types.Role(value)[source]
Bases:
str,Enum- ASSISTANT = 'assistant'
- USER = 'user'
- class afterimage.types.StructuredGenerationRow(*, instruction: str, context: str | None = None, persona: str | None = None, output: T, metadata: Dict[str, ~typing.Any]=<factory>)[source]
Bases:
BaseModel,Generic[T]Represents a single row of structured generation, including metadata.
- context: str | None
- instruction: str
- metadata: Dict[str, Any]
- model_config = {}
Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
- output: T
- persona: str | None