agents.rag.multi_agent_rag.enhanced_state_schemas¶
Enhanced State Schemas with Configuration Support.
This module provides state schemas that include configuration fields, solving the issue of storing agent-specific configuration in a clean way.
Classes¶
Adaptive threshold state extending Dynamic RAG state. |
|
Base RAG state with configuration support. |
|
Debate RAG state with configuration support. |
|
Dynamic RAG state with configuration support. |
|
FLARE state with configuration support. |
|
RAG state with grading information and configuration. |
|
Mixin to help MultiAgent classes work with configured states. |
Functions¶
|
Create a state instance with configuration. |
Module Contents¶
- class agents.rag.multi_agent_rag.enhanced_state_schemas.AdaptiveThresholdRAGState(/, **data)¶
Bases:
DynamicRAGState
Adaptive threshold state extending Dynamic RAG state.
Create a new model by parsing and validating input data from keyword arguments.
Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.
self is explicitly positional-only to allow self as a field name.
- Parameters:
data (Any)
- class agents.rag.multi_agent_rag.enhanced_state_schemas.ConfigurableRAGState(/, **data)¶
Bases:
haive.core.schema.prebuilt.rag_state.RAGState
Base RAG state with configuration support.
Create a new model by parsing and validating input data from keyword arguments.
Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.
self is explicitly positional-only to allow self as a field name.
- Parameters:
data (Any)
- class agents.rag.multi_agent_rag.enhanced_state_schemas.DebateRAGState(/, **data)¶
Bases:
ConfigurableRAGState
Debate RAG state with configuration support.
Create a new model by parsing and validating input data from keyword arguments.
Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.
self is explicitly positional-only to allow self as a field name.
- Parameters:
data (Any)
- class agents.rag.multi_agent_rag.enhanced_state_schemas.DynamicRAGState(/, **data)¶
Bases:
ConfigurableRAGState
Dynamic RAG state with configuration support.
Create a new model by parsing and validating input data from keyword arguments.
Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.
self is explicitly positional-only to allow self as a field name.
- Parameters:
data (Any)
- class agents.rag.multi_agent_rag.enhanced_state_schemas.FLAREState(/, **data)¶
Bases:
ConfigurableRAGState
FLARE state with configuration support.
Create a new model by parsing and validating input data from keyword arguments.
Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.
self is explicitly positional-only to allow self as a field name.
- Parameters:
data (Any)
- class agents.rag.multi_agent_rag.enhanced_state_schemas.GradedRAGState(/, **data)¶
Bases:
ConfigurableRAGState
RAG state with grading information and configuration.
Create a new model by parsing and validating input data from keyword arguments.
Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.
self is explicitly positional-only to allow self as a field name.
- Parameters:
data (Any)
- class agents.rag.multi_agent_rag.enhanced_state_schemas.StateConfigMixin¶
Mixin to help MultiAgent classes work with configured states.
- get_state_config(state)¶
Extract configuration from state.
- Parameters:
state (ConfigurableRAGState)
- Return type:
- update_state_config(state, **updates)¶
Update configuration in state.
- Parameters:
state (ConfigurableRAGState)
- Return type:
None
- agents.rag.multi_agent_rag.enhanced_state_schemas.create_configured_state(state_class, agent_name, workflow_type, **config_kwargs)¶
Create a state instance with configuration.
- Parameters:
state_class (type[ConfigurableRAGState])
agent_name (str)
workflow_type (str)
- Return type: