企业级生成式AI解决方案深度解析

1
2
3
4

!pip install langchain
!pip install langchain-community langchain-core
!pip install -U langchain-openai
!pip install langchain-chroma

1
2
3
4
5
6
7
8

from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.chat_models import ChatOpenAI
from langchain.prompts import PromptTemplate
from langchain.chains import RetrievalQA
from langchain_openai import OpenAIEmbeddings, ChatOpenAI
from langchain_chroma import Chroma
import chromadb
import pprint

1
2
3
4
5
6
7
8

input_texts = [
    "Artificial Intelligence (AI) is transforming industries around the world.",
    # ... 更多AI相关文本
]

combined_text = "\n".join(input_texts)
text_splitter = RecursiveCharacterTextSplitter(separators=["\n"], chunk_size=1, chunk_overlap=0)
chunk_texts = text_splitter.create_documents([combined_text])

1
2
3
4
5
6
7
8
9

openai_api_key = "your-api-key"
embeddings = OpenAIEmbeddings(openai_api_key=openai_api_key)

persist_directory = "chroma_db"
db = Chroma.from_documents(
    documents=chunk_texts,
    embedding=embeddings,
    persist_directory=persist_directory
)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18

template = """
Role: You are a Scientist.
Input: Use the following context to answer the question.
Context: {context}
Question: {question}
Steps: Answer politely and say, "I hope you are well," then focus on answering the question.
Expectation: Provide accurate and relevant answers based on the context provided.
Narrowing:
1. Limit your responses to the context given. Focus only on questions about AI.
2. If you don't know the answer, just say, "I am sorry...I don't know."
3. If there are words or questions outside the context of AI, just say, "Let's talk about AI."
Answer:
"""

PROMPT = PromptTemplate(
    template=template,
    input_variables=["context", "question"]
)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

temperature = 0.2
param = {
    "top_p": 0.4,
    "frequency_penalty": 0.1,
    "presence_penalty": 0.7
}

llm = ChatOpenAI(
    temperature=temperature,
    openai_api_key=openai_api_key,
    model_kwargs=param
)

qa_with_source = RetrievalQA.from_chain_type(
    llm=llm,
    chain_type='stuff',
    retriever=db.as_retriever(search_kwargs={"k": 5}),
    chain_type_kwargs={"prompt": PROMPT},
    return_source_documents=True,
)

query = "How does AI transform the industry?"
response = qa_with_source(query)
pprint.pprint(response)

1
2
3
4

[Document(page_content='Artificial Intelligence (AI) is transforming industries around the world.'),
 Document(page_content='\nThe future of AI holds potential for even greater advancements and applications across various fields.'),
 Document(page_content='\nIn manufacturing, AI systems improve efficiency and quality control.'),
 Document(page_content='\nAI is also revolutionizing the field of finance through automated trading and fraud detection.')]