Source code for langchain_community.llms.cerebriumai

import logging
from typing import Any, Dict, List, Mapping, Optional, cast

import requests
from langchain_core.callbacks import CallbackManagerForLLMRun
from langchain_core.language_models.llms import LLM
from langchain_core.pydantic_v1 import Extra, Field, SecretStr, root_validator
from langchain_core.utils import convert_to_secret_str, get_from_dict_or_env

from langchain_community.llms.utils import enforce_stop_tokens

logger = logging.getLogger(__name__)


class CerebriumAI(LLM):
    """CerebriumAI大型语言模型。

    要使用，您应该已安装``cerebrium`` python包。
    您还应该设置环境变量``CEREBRIUMAI_API_KEY``
    为您的API密钥，或将其作为构造函数中的命名参数传递。

    任何可以传递给调用的有效参数
    都可以传递，即使在此类上没有明确保存。

    示例：
        .. code-block:: python

            from langchain_community.llms import CerebriumAI
            cerebrium = CerebriumAI(endpoint_url="", cerebriumai_api_key="my-api-key")"""

    endpoint_url: str = ""
    """使用的模型端点"""

    model_kwargs: Dict[str, Any] = Field(default_factory=dict)
    """包含在`create`调用中有效但未明确指定的任何模型参数。"""

    cerebriumai_api_key: Optional[SecretStr] = None

    class Config:
        """这是用于pydantic配置的设置。"""

        extra = Extra.forbid

    @root_validator(pre=True)
    def build_extra(cls, values: Dict[str, Any]) -> Dict[str, Any]:
        """从传入的额外参数构建额外的kwargs。"""
        all_required_field_names = {field.alias for field in cls.__fields__.values()}

        extra = values.get("model_kwargs", {})
        for field_name in list(values):
            if field_name not in all_required_field_names:
                if field_name in extra:
                    raise ValueError(f"Found {field_name} supplied twice.")
                logger.warning(
                    f"""{field_name} was transferred to model_kwargs.
                    Please confirm that {field_name} is what you intended."""
                )
                extra[field_name] = values.pop(field_name)
        values["model_kwargs"] = extra
        return values

    @root_validator()
    def validate_environment(cls, values: Dict) -> Dict:
        """验证环境中是否存在API密钥和Python包。"""
        cerebriumai_api_key = convert_to_secret_str(
            get_from_dict_or_env(values, "cerebriumai_api_key", "CEREBRIUMAI_API_KEY")
        )
        values["cerebriumai_api_key"] = cerebriumai_api_key
        return values

    @property
    def _identifying_params(self) -> Mapping[str, Any]:
        """获取识别参数。"""
        return {
            **{"endpoint_url": self.endpoint_url},
            **{"model_kwargs": self.model_kwargs},
        }

    @property
    def _llm_type(self) -> str:
        """llm的返回类型。"""
        return "cerebriumai"

    def _call(
        self,
        prompt: str,
        stop: Optional[List[str]] = None,
        run_manager: Optional[CallbackManagerForLLMRun] = None,
        **kwargs: Any,
    ) -> str:
        headers: Dict = {
            "Authorization": cast(
                SecretStr, self.cerebriumai_api_key
            ).get_secret_value(),
            "Content-Type": "application/json",
        }
        params = self.model_kwargs or {}
        payload = {"prompt": prompt, **params, **kwargs}
        response = requests.post(self.endpoint_url, json=payload, headers=headers)
        if response.status_code == 200:
            data = response.json()
            text = data["result"]
            if stop is not None:
                # I believe this is required since the stop tokens
                # are not enforced by the model parameters
                text = enforce_stop_tokens(text, stop)
            return text
        else:
            response.raise_for_status()
        return ""