guardrails下載 - guardrails原始碼下載

guardrails

Ai源碼

v0.6.0

下載

什麼是guardrails ？

guardrails是一個 Python 框架，透過執行兩個關鍵功能來幫助建立可靠的 AI 應用程式：

guardrails在您的應用程式中執行輸入/輸出防護，以偵測、量化和減輕特定類型風險的存在。要了解全套風險，請查看guardrails中心。
guardrails可協助您從法學碩士產生結構化資料。

guardrails輪轂

guardrails Hub 是特定類型風險的預先建置措施（稱為「驗證器」）的集合。多個驗證器可以組合在一起形成輸入和輸出防護，攔截 LLM 的輸入和輸出。造訪guardrails Hub 查看驗證器的完整清單及其文件。

安裝

 pip install guardrails - ai

入門

為 LLM 驗證建立輸入和輸出防護

下載並配置guardrails Hub CLI。

pip install guardrails -ai
guardrails configure

從guardrails Hub 安裝護欄。

 guardrails hub install hub:// guardrails /regex_match

從已安裝的護欄建立一個守衛。

 from guardrails import Guard , OnFailAction
from guardrails . hub import RegexMatch

guard = Guard (). use (
    RegexMatch , regex = "(?d{3})?-? *d{3}-? *-?d{4}" , on_fail = OnFailAction . EXCEPTION
)

guard . validate ( "123-456-7890" )  # Guardrail passes

try :
    guard . validate ( "1234-789-0000" )  # Guardrail fails
except Exception as e :
    print ( e )

輸出：

 Validation failed for field with errors: Result must match (?d{3})?-? *d{3}-? *-?d{4}

在一個 Guard 內運行多個guardrails 。首先，從guardrails中心安裝必要的guardrails 。

 guardrails hub install hub:// guardrails /competitor_check
guardrails hub install hub:// guardrails /toxic_language

然後，從已安裝的guardrails建立一個 Guard 。

 from guardrails import Guard , OnFailAction
from guardrails . hub import CompetitorCheck , ToxicLanguage

guard = Guard (). use_many (
    CompetitorCheck ([ "Apple" , "Microsoft" , "Google" ], on_fail = OnFailAction . EXCEPTION ),
    ToxicLanguage ( threshold = 0.5 , validation_method = "sentence" , on_fail = OnFailAction . EXCEPTION )
)

guard . validate (
    """An apple a day keeps a doctor away.
    This is good advice for keeping your health."""
)  # Both the guardrails pass

try :
    guard . validate (
        """Shut the hell up! Apple just released a new iPhone."""
    )  # Both the guardrails fail
except Exception as e :
    print ( e )

輸出：

 Validation failed for field with errors: Found the following competitors: [['Apple']]. Please avoid naming those competitors next time, The following sentences in your response were found to be toxic:

- Shut the hell up!

使用guardrails從法學碩士產生結構化數據

讓我們來看一個例子，我們要求法學碩士產生假寵物名。為此，我們將建立一個 Pydantic BaseModel 來表示我們想要的輸出的結構。

 from pydantic import BaseModel , Field

class Pet ( BaseModel ):
    pet_type : str = Field ( description = "Species of pet" )
    name : str = Field ( description = "a unique pet name" )

現在，從Pet類別建立一個 Guard。 Guard 可用於以某種方式呼叫 LLM，以便將輸出格式化為Pet類別。在幕後，這是透過以下兩種方法之一完成的：

函數呼叫：對於支援函數呼叫的LLM，我們使用函數呼叫語法產生結構化資料。
提示最佳化：對於不支援函數呼叫的LLM，我們將預期輸出的schema加入提示中，以便LLM可以產生結構化資料。

 from guardrails import Guard
import openai

prompt = """
    What kind of pet should I get and what should I name it?

    ${gr.complete_json_suffix_v2}
"""
guard = Guard . for_pydantic ( output_class = Pet , prompt = prompt )

raw_output , validated_output , * rest = guard (
    llm_api = openai . completions . create ,
    engine = "gpt-3.5-turbo-instruct"
)

print ( validated_output )

這列印：

 {
    "pet_type": "dog",
    "name": "Buddy
}

guardrails伺服器

guardrails可以設定為 Flask 透過guardrails start提供的獨立服務，讓您透過 REST API 與其進行互動。這種方法簡化了guardrails驅動的應用程式的開發和部署。

安裝： pip install " guardrails -ai"
配置： guardrails configure
建立配置： guardrails create --validators=hub:// guardrails /two_words --name=two-word-guard
啟動開發伺服器： guardrails start --config=./config.py
透過下面的程式碼片段與開發伺服器交互

 # with the guardrails client
import guardrails as gr

gr.settings.use_server = True
guard = gr.Guard(name='two-word-guard')
guard.validate('this is more than two words')

# or with the openai sdk
import openai
openai.base_url = "http://localhost:8000/guards/two-word-guard/openai/v1/"
os.environ["OPENAI_API_KEY"] = "youropenaikey"

messages = [
        {
            "role": "user",
            "content": "tell me about an apple with 3 words exactly",
        },
    ]

completion = openai.chat.completions.create(
    model="gpt-4o-mini",
    messages=messages,
)

對於生產部署，我們建議使用 Docker 和 Gunicorn 作為 WSGI 伺服器，以提高效能和可擴展性。