openagi下載 - openagi源代碼下載

openagi

其他源碼

0.2.9.8

下載

Openagi

使所有人都可以使用自主的人類般的代理

Openagi的目標是使每個人都可以使用類似人類的代理商，從而為開放代理商舖平了道路，並最終為所有人鋪平了AGI。我們堅信AI的變革力量，並有信心這項倡議將有助於解決許多現實生活中的問題。目前，OpenAgi旨在為開發人員提供一個創建自主類人類代理的框架。

加入我們的Discord社區！

安裝

設置虛擬環境。

 # For Mac and Linux users
python3 -m venv venv
source venv/bin/activate

# For Windows users
python -m venv venv
venv/scripts/activate

安裝OpenAgi

pip install openagi

或者

 git clone https://github.com/aiplanethub/openagi.git
pip install -e .

示例（手動代理執行）

工人用於創建多代理體系結構。

請按照此示例創建一個旅行計劃代理，該示例可以幫助您計劃SF的行程。

 from openagi . agent import Admin
from openagi . planner . task_decomposer import TaskPlanner
from openagi . actions . tools . ddg_search import DuckDuckGoSearch
from openagi . llms . openai import OpenAIModel
from openagi . worker import Worker

plan = TaskPlanner ( human_intervene = False )
action = DuckDuckGoSearch

import os
os . environ [ 'OPENAI_API_KEY' ] = "sk-xxxx"
config = OpenAIModel . load_from_env_config ()
llm = OpenAIModel ( config = config )

trip_plan = Worker (
        role = "Trip Planner" ,
        instructions = """
        User loves calm places, suggest the best itinerary accordingly.
        """ ,
        actions = [ action ],
        max_iterations = 10 )

admin = Admin (
    llm = llm ,
    actions = [ action ],
    planner = plan ,
)
admin . assign_workers ([ trip_plan ])

res = admin . run (
    query = "Give me total 3 Days Trip to San francisco Bay area" ,
    description = "You are a knowledgeable local guide with extensive information about the city, it's attractions and customs" ,
)
print ( res )

示例（自主多代理執行）

現在，讓我們建立一個體育代理，可以在沒有任何工人的情況下自主運行。

 from openagi . planner . task_decomposer import TaskPlanner
from openagi . actions . tools . tavilyqasearch import TavilyWebSearchQA
from openagi . agent import Admin
from openagi . llms . gemini import GeminiModel

import os
os . environ [ 'TAVILY_API_KEY' ] = "<replace with Tavily key>"
os . environ [ 'GOOGLE_API_KEY' ] = "<replace with Gemini key>"
os . environ [ 'Gemini_MODEL' ] = "gemini-1.5-flash"
os . environ [ 'Gemini_TEMP' ] = "0.1"

gemini_config = GeminiModel . load_from_env_config ()
llm = GeminiModel ( config = gemini_config )

# define the planner
plan = TaskPlanner ( autonomous = True , human_intervene = True )

admin = Admin (
    actions = [ TavilyWebSearchQA ],
    planner = plan ,
    llm = llm ,
)
res = admin . run (
    query = "I need cricket updates from India vs Sri lanka 2024 ODI match in Sri Lanka" ,
    description = f"give me the results of India vs Sri Lanka ODI and respective Man of the Match" ,
)
print ( res )

長期記憶以前從未

使用LTM，Openagi代理現在可以：

回想過去的互動以提供對話中的連續性。
隨著時間的推移，根據用戶輸入來學習和適應。
通過引用先前的對話來提供上下文相關的響應。
通過每次連續的相互作用提高其準確性和效率。

 import os
from openagi . agent import Admin
from openagi . llms . openai import OpenAIModel
from openagi . memory import Memory
from openagi . planner . task_decomposer import TaskPlanner
from openagi . worker import Worker
from openagi . actions . tools . ddg_search import DuckDuckGoSearch

memory = Memory ( long_term = True )

os . environ [ 'OPENAI_API_KEY' ] = "-"
config = OpenAIModel . load_from_env_config ()
llm = OpenAIModel ( config = config )

web_searcher = Worker (
    role = "Web Researcher" ,
    instructions = """
    You are tasked with conducting web searches using DuckDuckGo.
    Find the most relevant and accurate information based on the user's query.
    """ ,
    actions = [ DuckDuckGoSearch ], 
)

admin = Admin (
    actions = [ DuckDuckGoSearch ],
    planner = TaskPlanner ( human_intervene = False ),
    memory = memory ,
    llm = llm ,
)
admin . assign_workers ([ web_searcher ])

query = input ( "Enter your search query: " )
description = f"Find accurate and relevant information for the query: { query } "

res = admin . run ( query = query , description = description )
print ( res )

文件

有關更多查詢

用例：

教育：在教育方面，代理商可以提供個性化的學習經驗。他們根據學生的進步，表現和興趣來適應和量身定制學習內容。它可以擴展到自動執行其他各種行政任務，並協助教師提高其生產率。
財務和銀行業務：金融服務可以使用代理人進行欺詐檢測，風險評估，個性化銀行建議，自動交易和客戶服務。他們有助於分析大量交易，以確定可疑活動並提供量身定制的投資建議。
醫療保健：可以部署代理商來監視患者，提供個性化的健康建議，管理患者數據並自動化行政任務。他們還可以根據症狀和病史協助診斷疾病。