Source code for arize._generated.api_client.models.llm_generation_run_config
# coding: utf-8
"""
Arize REST API
API specification for the backend data server. The API is hosted globally at https://api.arize.com/v2 or in your own environment.
The version of the OpenAPI document: 2.0.0
Generated by OpenAPI Generator (https://openapi-generator.tech)
Do not edit the class manually.
""" # noqa: E501
from __future__ import annotations
import pprint
import re # noqa: F401
import json
from pydantic import BaseModel, ConfigDict, Field, StrictStr, field_validator
from typing import Any, ClassVar, Dict, List, Optional
from typing_extensions import Annotated
from arize._generated.api_client.models.input_variable_format import InputVariableFormat
from arize._generated.api_client.models.invocation_params import InvocationParams
from arize._generated.api_client.models.llm_message import LLMMessage
from arize._generated.api_client.models.tool_config import ToolConfig
from typing import Optional, Set
from typing_extensions import Self
[docs]
class LlmGenerationRunConfig(BaseModel):
"""
Configuration for running an LLM prompt against each dataset example.
""" # noqa: E501
experiment_type: StrictStr = Field(description="Discriminator. Must be `\"llm_generation\"`.")
ai_integration_id: StrictStr = Field(description="AI integration identifier (base64).")
model_name: Optional[StrictStr] = Field(default=None, description="Model name (e.g. `gpt-4o`). Falls back to the integration's default if omitted.")
messages: Annotated[List[LLMMessage], Field(min_length=1)] = Field(description="Array of message objects (at least one).")
input_variable_format: InputVariableFormat
invocation_parameters: Optional[InvocationParams] = None
provider_parameters: Optional[Dict[str, Any]] = Field(default=None, description="Provider-specific parameters. Defaults to `{}` (no overrides) if omitted.")
tool_config: Optional[ToolConfig] = None
prompt_version_id: Optional[StrictStr] = Field(default=None, description="Prompt version identifier (base64). Links to a Prompt Hub version for traceability.")
__properties: ClassVar[List[str]] = ["experiment_type", "ai_integration_id", "model_name", "messages", "input_variable_format", "invocation_parameters", "provider_parameters", "tool_config", "prompt_version_id"]
[docs]
@field_validator('experiment_type')
def experiment_type_validate_enum(cls, value):
"""Validates the enum"""
if value not in set(['llm_generation']):
raise ValueError("must be one of enum values ('llm_generation')")
return value
model_config = ConfigDict(
populate_by_name=True,
validate_assignment=True,
protected_namespaces=(),
)
[docs]
def to_str(self) -> str:
"""Returns the string representation of the model using alias"""
return pprint.pformat(self.model_dump(by_alias=True))
[docs]
def to_json(self) -> str:
"""Returns the JSON representation of the model using alias"""
# TODO: pydantic v2: use .model_dump_json(by_alias=True, exclude_unset=True) instead
return json.dumps(self.to_dict())
[docs]
@classmethod
def from_json(cls, json_str: str) -> Optional[Self]:
"""Create an instance of LlmGenerationRunConfig from a JSON string"""
return cls.from_dict(json.loads(json_str))
[docs]
def to_dict(self) -> Dict[str, Any]:
"""Return the dictionary representation of the model using alias.
This has the following differences from calling pydantic's
`self.model_dump(by_alias=True)`:
* `None` is only added to the output dict for nullable fields that
were set at model initialization. Other fields with value `None`
are ignored.
"""
excluded_fields: Set[str] = set([
])
_dict = self.model_dump(
by_alias=True,
exclude=excluded_fields,
exclude_none=True,
)
# override the default output from pydantic by calling `to_dict()` of each item in messages (list)
_items = []
if self.messages:
for _item_messages in self.messages:
if _item_messages:
_items.append(_item_messages.to_dict())
_dict['messages'] = _items
# override the default output from pydantic by calling `to_dict()` of invocation_parameters
if self.invocation_parameters:
_dict['invocation_parameters'] = self.invocation_parameters.to_dict()
# override the default output from pydantic by calling `to_dict()` of tool_config
if self.tool_config:
_dict['tool_config'] = self.tool_config.to_dict()
# set to None if prompt_version_id (nullable) is None
# and model_fields_set contains the field
if self.prompt_version_id is None and "prompt_version_id" in self.model_fields_set:
_dict['prompt_version_id'] = None
return _dict
[docs]
@classmethod
def from_dict(cls, obj: Optional[Dict[str, Any]]) -> Optional[Self]:
"""Create an instance of LlmGenerationRunConfig from a dict"""
if obj is None:
return None
if not isinstance(obj, dict):
return cls.model_validate(obj)
# raise errors for additional fields in the input
for _key in obj.keys():
if _key not in cls.__properties:
raise ValueError("Error due to additional fields (not defined in LlmGenerationRunConfig) in the input: " + _key)
_obj = cls.model_validate({
"experiment_type": obj.get("experiment_type"),
"ai_integration_id": obj.get("ai_integration_id"),
"model_name": obj.get("model_name"),
"messages": [LLMMessage.from_dict(_item) for _item in obj["messages"]] if obj.get("messages") is not None else None,
"input_variable_format": obj.get("input_variable_format"),
"invocation_parameters": InvocationParams.from_dict(obj["invocation_parameters"]) if obj.get("invocation_parameters") is not None else None,
"provider_parameters": obj.get("provider_parameters"),
"tool_config": ToolConfig.from_dict(obj["tool_config"]) if obj.get("tool_config") is not None else None,
"prompt_version_id": obj.get("prompt_version_id")
})
return _obj
