In [ ]
已复制！
!pip install llama-index
!pip install llama-index
In [ ]
已复制！
import nest_asyncio
nest_asyncio.apply()
import os
os.environ["OPENAI_API_KEY"] = "sk-..."
from llama_index.llms.openai import OpenAI
from llama_index.embeddings.openai import OpenAIEmbedding
from llama_index.core import Settings
llm = OpenAI(model="gpt-4", temperature=0.1)
embed_model = OpenAIEmbedding()
Settings.llm = llm
Settings.embed_model = embed_model
import nest_asyncio nest_asyncio.apply() import os os.environ["OPENAI_API_KEY"] = "sk-..." from llama_index.llms.openai import OpenAI from llama_index.embeddings.openai import OpenAIEmbedding from llama_index.core import Settings llm = OpenAI(model="gpt-4", temperature=0.1) embed_model = OpenAIEmbedding() Settings.llm = llm Settings.embed_model = embed_model
In [ ]
已复制！
from llama_index.core.tools import FunctionTool
from llama_index.core.agent import (
FunctionCallingAgentWorker,
ReActAgent,
)
from IPython.display import display, HTML
from llama_index.core.tools import FunctionTool from llama_index.core.agent import ( FunctionCallingAgentWorker, ReActAgent, ) from IPython.display import display, HTML
In [ ]
已复制！
def multiply(a: int, b: int) -> int:
"""Multiply two integers and returns the result integer"""
return a * b
def add(a: int, b: int) -> int:
"""Add two integers and returns the result integer"""
return a + b
def subtract(a: int, b: int) -> int:
"""Subtract two integers and returns the result integer"""
return a - b
multiply_tool = FunctionTool.from_defaults(fn=multiply)
add_tool = FunctionTool.from_defaults(fn=add)
subtract_tool = FunctionTool.from_defaults(fn=subtract)
def multiply(a: int, b: int) -> int: """Multiply two integers and returns the result integer""" return a * b def add(a: int, b: int) -> int: """Add two integers and returns the result integer""" return a + b def subtract(a: int, b: int) -> int: """Subtract two integers and returns the result integer""" return a - b multiply_tool = FunctionTool.from_defaults(fn=multiply) add_tool = FunctionTool.from_defaults(fn=add) subtract_tool = FunctionTool.from_defaults(fn=subtract)
In [ ]
已复制！
agent = ReActAgent.from_tools(
[multiply_tool, add_tool, subtract_tool], llm=llm, verbose=True
)
agent = ReActAgent.from_tools( [multiply_tool, add_tool, subtract_tool], llm=llm, verbose=True )
In [ ]
已复制！
response = agent.chat("What is (26 * 2) + 2024?")
response = agent.chat("What is (26 * 2) + 2024?")
Thought: The user wants to perform a mathematical operation. I need to first multiply 26 by 2 and then add the result to 2024. I'll use the 'multiply' tool first. Action: multiply Action Input: {'a': 26, 'b': 2} Observation: 52 Thought: The multiplication result is 52. Now, I need to add this result to 2024. I'll use the 'add' tool for this. Action: add Action Input: {'a': 52, 'b': 2024} Observation: 2076 Thought: I can answer without using any more tools. I'll use the user's language to answer. Answer: 2076
In [ ]
已复制！
display(HTML(f'<p style="font-size:20px">{response.response}</p>'))
display(HTML(f'
{response.response}
'))2076
In [ ]
已复制！
agent_worker = FunctionCallingAgentWorker.from_tools(
[multiply_tool, add_tool, subtract_tool],
llm=llm,
verbose=True,
allow_parallel_tool_calls=False,
)
agent = agent_worker.as_agent()
agent_worker = FunctionCallingAgentWorker.from_tools( [multiply_tool, add_tool, subtract_tool], llm=llm, verbose=True, allow_parallel_tool_calls=False, ) agent = agent_worker.as_agent()
In [ ]
已复制！
response = agent.chat("What is (26 * 2) + 2024?")
response = agent.chat("What is (26 * 2) + 2024?")
Added user message to memory: What is (26 * 2) + 2024?
=== Calling Function ===
Calling function: multiply with args: {"a": 26, "b": 2}
=== Function Output ===
52
=== Calling Function ===
Calling function: add with args: {"a": 52, "b": 2024}
=== Function Output ===
2076
=== LLM Response ===
The result of (26 * 2) + 2024 is 2076.
In [ ]
已复制！
display(HTML(f'<p style="font-size:20px">{response.response}</p>'))
display(HTML(f'
{response.response}
'))assistant: (26 * 2) + 2024 的结果是 2076。
In [ ]
已复制！
!wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/10k/uber_2021.pdf' -O './uber_2021.pdf'
!wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/10k/lyft_2021.pdf' -O './lyft_2021.pdf'
!wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/10k/uber_2021.pdf' -O './uber_2021.pdf' !wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/10k/lyft_2021.pdf' -O './lyft_2021.pdf'
--2024-05-16 14:00:56-- https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/10k/uber_2021.pdf Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ... Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 1880483 (1.8M) [application/octet-stream] Saving to: ‘./uber_2021.pdf’ ./uber_2021.pdf 100%[===================>] 1.79M 8.54MB/s in 0.2s 2024-05-16 14:00:57 (8.54 MB/s) - ‘./uber_2021.pdf’ saved [1880483/1880483] --2024-05-16 14:00:57-- https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/10k/lyft_2021.pdf Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ... Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 1440303 (1.4M) [application/octet-stream] Saving to: ‘./lyft_2021.pdf’ ./lyft_2021.pdf 100%[===================>] 1.37M 6.90MB/s in 0.2s 2024-05-16 14:00:58 (6.90 MB/s) - ‘./lyft_2021.pdf’ saved [1440303/1440303]
In [ ]
已复制！
from llama_index.core import SimpleDirectoryReader
uber_docs = SimpleDirectoryReader(input_files=["./uber_2021.pdf"]).load_data()
lyft_docs = SimpleDirectoryReader(input_files=["./lyft_2021.pdf"]).load_data()
from llama_index.core import SimpleDirectoryReader uber_docs = SimpleDirectoryReader(input_files=["./uber_2021.pdf"]).load_data() lyft_docs = SimpleDirectoryReader(input_files=["./lyft_2021.pdf"]).load_data()
In [ ]
已复制！
from llama_index.core import VectorStoreIndex
uber_index = VectorStoreIndex.from_documents(uber_docs)
uber_query_engine = uber_index.as_query_engine(similarity_top_k=3)
lyft_index = VectorStoreIndex.from_documents(lyft_docs)
lyft_query_engine = lyft_index.as_query_engine(similarity_top_k=3)
from llama_index.core import VectorStoreIndex uber_index = VectorStoreIndex.from_documents(uber_docs) uber_query_engine = uber_index.as_query_engine(similarity_top_k=3) lyft_index = VectorStoreIndex.from_documents(lyft_docs) lyft_query_engine = lyft_index.as_query_engine(similarity_top_k=3)
In [ ]
已复制！
response = uber_query_engine.query("What are the investments of Uber in 2021?")
response = uber_query_engine.query("What are the investments of Uber in 2021?")
In [ ]
已复制！
display(HTML(f'<p style="font-size:20px">{response.response}</p>'))
display(HTML(f'
{response.response}
'))2021 年，Uber 进行了重大投资，以扩展其国际业务并与本地和其他全球竞争对手竞争。这包括收购 Careem 和 Cornershop。此外，Uber 继续开发新技术以增强现有产品和服务，并通过研发扩展其产品范围。他们还在美国推出了 Uber One，这是一个汇集 Uber 最佳服务的跨平台会员计划。
In [ ]
已复制！
response = lyft_query_engine.query("What are lyft investments in 2021?")
response = lyft_query_engine.query("What are lyft investments in 2021?")
In [ ]
已复制！
display(HTML(f'<p style="font-size:20px">{response.response}</p>'))
display(HTML(f'
{response.response}
'))2021 年，Lyft 继续投资扩展其轻型车辆网络和 Lyft Autonomous，专注于在 Lyft 网络上部署和扩展第三方自动驾驶技术。该公司还承诺到 2030 年底在 Lyft 网络上实现 100% 电动汽车。此外，Lyft 与丰田汽车公司的子公司 Woven Planet 完成了一项交易，剥离了与其自动驾驶汽车部门 Level 5 相关的某些资产。
In [ ]
已复制！
from llama_index.core.tools import QueryEngineTool, ToolMetadata
from llama_index.core.agent import FunctionCallingAgentWorker
query_engine_tools = [
QueryEngineTool(
query_engine=lyft_query_engine,
metadata=ToolMetadata(
name="lyft_10k",
description="Provides information about Lyft financials for year 2021",
),
),
QueryEngineTool(
query_engine=uber_query_engine,
metadata=ToolMetadata(
name="uber_10k",
description="Provides information about Uber financials for year 2021",
),
),
]
agent_worker = FunctionCallingAgentWorker.from_tools(
query_engine_tools,
llm=llm,
verbose=True,
allow_parallel_tool_calls=False,
)
agent = agent_worker.as_agent()
from llama_index.core.tools import QueryEngineTool, ToolMetadata from llama_index.core.agent import FunctionCallingAgentWorker query_engine_tools = [ QueryEngineTool( query_engine=lyft_query_engine, metadata=ToolMetadata( name="lyft_10k", description="Provides information about Lyft financials for year 2021", ), ), QueryEngineTool( query_engine=uber_query_engine, metadata=ToolMetadata( name="uber_10k", description="Provides information about Uber financials for year 2021", ), ), ] agent_worker = FunctionCallingAgentWorker.from_tools( query_engine_tools, llm=llm, verbose=True, allow_parallel_tool_calls=False, ) agent = agent_worker.as_agent()
In [ ]
已复制！
response = agent.chat("What are the investments of Uber in 2021?")
response = agent.chat("What are the investments of Uber in 2021?")
Added user message to memory: What are the investments of Uber in 2021?
=== Calling Function ===
Calling function: uber_10k with args: {"input": "investments"}
=== Function Output ===
Uber's investments primarily consist of money market funds, cash deposits, U.S. government and agency securities, and investment-grade corporate debt securities. The company's investment policy aims to preserve capital and meet liquidity requirements without significantly increasing risk. As of December 31, 2021, Uber had cash and cash equivalents including restricted cash and cash equivalents totaling $7.8 billion. They also hold investments in other companies, including minority-owned, privately-held affiliates and recently public companies. The carrying value of these investments was $12.6 billion as of December 31, 2021.
=== LLM Response ===
In 2021, Uber's investments primarily consisted of money market funds, cash deposits, U.S. government and agency securities, and investment-grade corporate debt securities. Their investment policy aims to preserve capital and meet liquidity requirements without significantly increasing risk. As of December 31, 2021, Uber had cash and cash equivalents including restricted cash and cash equivalents totaling $7.8 billion. They also hold investments in other companies, including minority-owned, privately-held affiliates and recently public companies. The carrying value of these investments was $12.6 billion as of December 31, 2021.
In [ ]
已复制！
display(HTML(f'<p style="font-size:20px">{response.response}</p>'))
display(HTML(f'
{response.response}
'))assistant: 2021年，Uber的投资主要包括货币市场基金、现金存款、美国政府和机构证券，以及投资级公司债券。其投资政策旨在在不显著增加风险的情况下保全资本并满足流动性需求。截至2021年12月31日，Uber拥有现金和现金等价物，包括受限现金和现金等价物，总计78亿美元。他们还持有其他公司的投资，包括少数股权、非上市公司关联公司以及近期上市公司。截至2021年12月31日，这些投资的账面价值为126亿美元。
In [ ]
已复制！
response = agent.chat("What are lyft investments in 2021?")
response = agent.chat("What are lyft investments in 2021?")
Added user message to memory: What are lyft investments in 2021?
=== Calling Function ===
Calling function: lyft_10k with args: {"input": "investments"}
=== Function Output ===
The company's investments include cash and cash equivalents, short-term investments, and restricted investments. Cash and cash equivalents include certificates of deposits, commercial paper, and corporate bonds that have an original maturity of 90 days or less and are readily convertible to known amounts of cash. Short-term investments are comprised of commercial paper, certificates of deposit, and corporate bonds, which mature in twelve months or less. Restricted investments are comprised of debt security investments in commercial paper, certificates of deposit, corporate bonds, and U.S. government securities which are held in trust accounts at third-party financial institutions pursuant to certain contracts with insurance providers. The company also has investments in non-marketable equity securities, which are measured at cost, with remeasurements to fair value only upon the occurrence of observable transactions for identical or similar investments of the same issuer or impairment.
=== LLM Response ===
In 2021, Lyft's investments included cash and cash equivalents, short-term investments, and restricted investments. Cash and cash equivalents included certificates of deposits, commercial paper, and corporate bonds that have an original maturity of 90 days or less and are readily convertible to known amounts of cash. Short-term investments were comprised of commercial paper, certificates of deposit, and corporate bonds, which mature in twelve months or less.
Restricted investments were comprised of debt security investments in commercial paper, certificates of deposit, corporate bonds, and U.S. government securities which are held in trust accounts at third-party financial institutions pursuant to certain contracts with insurance providers.
The company also had investments in non-marketable equity securities, which are measured at cost, with remeasurements to fair value only upon the occurrence of observable transactions for identical or similar investments of the same issuer or impairment.
In [ ]
已复制！
display(HTML(f'<p style="font-size:20px">{response.response}</p>'))
display(HTML(f'
{response.response}
'))assistant: 2021年，Lyft的投资包括现金和现金等价物、短期投资和受限投资。现金和现金等价物包括原始期限在90天或以内且可轻松转换为已知现金金额的存单、商业票据和公司债券。短期投资包括商业票据、存单和公司债券，这些投资的期限在十二个月或以内。受限投资包括根据与保险提供商的某些合同在第三方金融机构的信托账户中持有的商业票据、存单、公司债券和美国政府证券等债务证券投资。该公司还持有非流通股权证券的投资，这些投资按成本计量，只有在同一发行方的相同或类似投资发生可观察交易或出现减值时，才重新计量为公允价值。
