nat.plugins.data_flywheel.observability.schema.sink.elasticsearch.dfw_es_record

Attributes

logger
Message

Classes

FinishReason	Finish reason for chat completion responses.
Function	Function call structure used in both requests and responses.
ToolCall	Tool call structure used in responses.
AssistantMessage	Assistant message structure used in responses.
SystemMessage	System message structure used in requests.
UserMessage	User message structure used in requests.
ToolMessage	Tool message structure used in responses.
FunctionMessage	Function message structure used in responses.
FunctionParameters	Function parameters structure used in responses.
FunctionDetails	Function details structure used in requests.
RequestTool	Request tool structure used in requests.
ESRequest	Request structure used in requests.
ResponseMessage	Response message structure used in responses.
ResponseChoice	Response choice structure used in responses.
Response	Response structure used in responses.
DFWESRecord	Data Flywheel Elasticsearch record.

Module Contents

logger

class FinishReason

Bases: str, enum.Enum

Finish reason for chat completion responses.

Initialize self. See help(type(self)) for accurate signature.

STOP = 'stop'

LENGTH = 'length'

TOOL_CALLS = 'tool_calls'

class Function(/, **data: Any)

Bases: pydantic.BaseModel

Function call structure used in both requests and responses.

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.

name: str = None

arguments: dict = None

class ToolCall(/, **data: Any)

Bases: pydantic.BaseModel

Tool call structure used in responses.

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.

function: Function = None

type_: Literal['function'] = None

class AssistantMessage(/, **data: Any)

Bases: pydantic.BaseModel

Assistant message structure used in responses.

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.

role: Literal['assistant', 'ai'] = None

content: str | None = None

tool_calls: list[ToolCall] | None = None

class SystemMessage(/, **data: Any)

Bases: pydantic.BaseModel

System message structure used in requests.

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.

content: str = None

role: Literal['system'] = None

class UserMessage(/, **data: Any)

Bases: pydantic.BaseModel

User message structure used in requests.

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.

content: str = None

role: Literal['user', 'human'] = None

class ToolMessage(/, **data: Any)

Bases: pydantic.BaseModel

Tool message structure used in responses.

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.

content: str = None

role: Literal['tool'] = None

tool_call_id: str = None

class FunctionMessage(/, **data: Any)

Bases: pydantic.BaseModel

Function message structure used in responses.

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.

content: str | None = None

role: Literal['function', 'chain'] = None

type Message = SystemMessage | UserMessage | AssistantMessage | ToolMessage | FunctionMessage

class FunctionParameters(/, **data: Any)

Bases: pydantic.BaseModel

Function parameters structure used in responses.

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.

properties: dict = None

required: list[str] = None

type_: Literal['object'] = None

classmethod validate_property_limit(v: dict) → dict

Enforce 8-property limit for tool parameters (WAR for NIM bug).

class FunctionDetails(/, **data: Any)

Bases: pydantic.BaseModel

Function details structure used in requests.

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.

name: str = None

description: str = None

parameters: FunctionParameters = None

class RequestTool(/, **data: Any)

Bases: pydantic.BaseModel

Request tool structure used in requests.

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.

type: Literal['function'] = None

function: FunctionDetails = None

class ESRequest(/, **data: Any)

Bases: pydantic.BaseModel

Request structure used in requests.

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.

model_config

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

tools: list[RequestTool] | None = None

messages: list[Message] | None = None

model: str = None

temperature: float | None = None

max_tokens: int | None = None

class ResponseMessage(/, **data: Any)

Bases: pydantic.BaseModel

Response message structure used in responses.

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.

role: Literal['assistant'] | None = None

content: str | None = None

tool_calls: list[ToolCall] | None = None

class ResponseChoice(/, **data: Any)

Bases: pydantic.BaseModel

Response choice structure used in responses.

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.

message: ResponseMessage = None

finish_reason: FinishReason | None = None

index: int | None = None

class Response(/, **data: Any)

Bases: pydantic.BaseModel

Response structure used in responses.

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.

model_config

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

choices: list[ResponseChoice] | None = None

id: str | None = None

object: str | None = None

created: int | None = None

model: str | None = None

usage: dict[str, Any] | None = None

class DFWESRecord(/, **data: Any)

Bases: pydantic.BaseModel

Data Flywheel Elasticsearch record.

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.

model_config

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

contract_version: nat.plugins.data_flywheel.observability.schema.sink.elasticsearch.contract_version.ContractVersion = None

request: ESRequest = None

response: Response = None

client_id: str = None

workload_id: str = None

timestamp: int = None

error_details: str | None = None

validate_data_consistency() → Self