聊天引擎 - 压缩问题模式¶

“压缩问题”是一种简单的聊天模式，构建在数据上的查询引擎之上。

对于每次聊天交互

• 首先，从对话上下文和最后一条消息生成一个独立的问题，然后
• 使用压缩后的问题查询查询引擎以获取响应。

这种方法很简单，适用于与知识库直接相关的问题。由于它总是查询知识库，因此可能难以回答元问题，例如“我之前问过你什么？”

如果您正在 Colab 上打开此 Notebook，您可能需要安装 LlamaIndex 🦙。

In [ ]

%pip install llama-index-llms-openai

%pip install llama-index-llms-openai

In [ ]

!pip install llama-index

!pip install llama-index

下载数据¶

In [ ]

!mkdir -p 'data/paul_graham/'
!wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/paul_graham/paul_graham_essay.txt' -O 'data/paul_graham/paul_graham_essay.txt'

!mkdir -p 'data/paul_graham/' !wget 'https://raw.githubusercontent.com/run-llama/llama_index/main/docs/docs/examples/data/paul_graham/paul_graham_essay.txt' -O 'data/paul_graham/paul_graham_essay.txt'

5行代码快速开始¶

加载数据并构建索引

In [ ]

from llama_index.core import VectorStoreIndex, SimpleDirectoryReader

data = SimpleDirectoryReader(input_dir="./data/paul_graham/").load_data()
index = VectorStoreIndex.from_documents(data)

from llama_index.core import VectorStoreIndex, SimpleDirectoryReader data = SimpleDirectoryReader(input_dir="./data/paul_graham/").load_data() index = VectorStoreIndex.from_documents(data)

配置聊天引擎

In [ ]

chat_engine = index.as_chat_engine(chat_mode="condense_question", verbose=True)

chat_engine = index.as_chat_engine(chat_mode="condense_question", verbose=True)

与您的数据聊天

In [ ]

response = chat_engine.chat("What did Paul Graham do after YC?")

response = chat_engine.chat("What did Paul Graham do after YC?")

Querying with: What was the next step in Paul Graham's career after his involvement with Y Combinator?

In [ ]

print(response)

print(response)

Paul Graham's next step in his career after his involvement with Y Combinator was to take up painting. He spent most of the rest of 2014 painting and then in March 2015 he started working on Lisp again.

提出一个后续问题

In [ ]

response = chat_engine.chat("What about after that?")

response = chat_engine.chat("What about after that?")

Querying with: What did Paul Graham do after he started working on Lisp again in March 2015?

In [ ]

print(response)

print(response)

Paul Graham spent the rest of 2015 writing essays and working on his new dialect of Lisp, which he called Arc. He also looked for an apartment to buy and started planning a second still life painting from the same objects.

In [ ]

response = chat_engine.chat("Can you tell me more?")

response = chat_engine.chat("Can you tell me more?")

Querying with: What did Paul Graham do after he started working on Lisp again in March 2015?

In [ ]

print(response)

print(response)

Paul Graham spent the rest of 2015 writing essays and working on his new dialect of Lisp, which he called Arc. He also looked for an apartment to buy and started planning for a second still life painting.

重置对话状态

In [ ]

chat_engine.reset()

chat_engine.reset()

In [ ]

response = chat_engine.chat("What about after that?")

response = chat_engine.chat("What about after that?")

Querying with: What happens after the current situation?

In [ ]

print(response)

print(response)

After the current situation, the narrator resumes painting and experimenting with a new kind of still life. He also resumes his old life in New York, now that he is rich. He is able to take taxis and eat in restaurants, which is exciting for a while. He also starts to make connections with other people who are trying to paint in New York.

流式支持¶

In [ ]

from llama_index.core import VectorStoreIndex, SimpleDirectoryReader
from llama_index.llms.openai import OpenAI

llm = OpenAI(model="gpt-3.5-turbo", temperature=0)

data = SimpleDirectoryReader(input_dir="../data/paul_graham/").load_data()

index = VectorStoreIndex.from_documents(data)

from llama_index.core import VectorStoreIndex, SimpleDirectoryReader from llama_index.llms.openai import OpenAI llm = OpenAI(model="gpt-3.5-turbo", temperature=0) data = SimpleDirectoryReader(input_dir="../data/paul_graham/").load_data() index = VectorStoreIndex.from_documents(data)

In [ ]

chat_engine = index.as_chat_engine(
    chat_mode="condense_question", llm=llm, verbose=True
)

chat_engine = index.as_chat_engine( chat_mode="condense_question", llm=llm, verbose=True )

In [ ]

response = chat_engine.stream_chat("What did Paul Graham do after YC?")
for token in response.response_gen:
    print(token, end="")

response = chat_engine.stream_chat("What did Paul Graham do after YC?") for token in response.response_gen: print(token, end="")

Querying with: What did Paul Graham do after leaving YC?
After leaving YC, Paul Graham started painting and focused on improving his skills in that area. He then started writing essays again and began working on Lisp.