observability.schema.provider.openai_trace_source

Attributes

ProviderT
logger

Classes

OpenAIMetadata	Metadata for the OpenAITraceSource.
OpenAITraceSourceBase	Base class for the OpenAITraceSource.
OpenAITraceSource	Concrete OpenAI trace source.

Module Contents

ProviderT

logger

class OpenAIMetadata(/, **data: Any)

Bases: pydantic.BaseModel

Metadata for the OpenAITraceSource.

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.

tools_schema: list[nat.data_models.intermediate_step.ToolSchema] | None = None

chat_responses: list[dict[str, Any]] | None = None

class OpenAITraceSourceBase

Bases: nat.plugins.data_flywheel.observability.schema.trace_source_base.TraceSourceBase

Base class for the OpenAITraceSource.

input_value: list[nat.plugins.data_flywheel.observability.schema.provider.openai_message.OpenAIMessage]

metadata: OpenAIMetadata

classmethod validate_input_value(

v: Any,

) → list[nat.plugins.data_flywheel.observability.schema.provider.openai_message.OpenAIMessage]

Validate the input value for the OpenAITraceSource.

classmethod validate_metadata(v: Any) → OpenAIMetadata | dict[str, Any]

Normalize metadata supplied as OpenAIMetadata, dict, or JSON string.

class OpenAITraceSource

Bases: OpenAITraceSourceBase

Concrete OpenAI trace source.