RAG一文读懂｜概念、应用场景、优势与项目代码示例

!pip install --upgrade erniebot

## 测试embedding import ernieboterniebot.api_type = "aistudio"erniebot.access_token = "<你的token>"response = erniebot.Embedding.create(    model="ernie-text-embedding",    input=[        "我是百度公司开发的人工智能语言模型，我的中文名是文心一言，英文名是ERNIE-Bot，可以协助您完成范围广泛的任务并提供有关各种主题的信息，比如回答问题，提供定义和解释及建议。如果您有任何问题，请随时向我提问。" ])print(response.get_result())
##[[0.12396156042814255, 0.06514577567577362, 0.05345790088176727, 0.05495057627558708, 0.01716597005724907, -0.08169233053922653, -0.02372867614030838, -0.050388969480991364, -0.04034141078591347, 0.05868257209658623, 0.016312476247549057, -0.058401163667440414, -0.042211249470710754, 0.045841023325920105, -0.14607082307338715, -0.04978140816092491, -0.03679147735238075, 0.012605858966708183, -0.014128394424915314, 0.0006767146405763924, 0.013087217696011066, -0.0018176172161474824, -0.04658341035246849, -0.06002342328429222, 0.023892290890216827, -0.03365868330001831, 0.08845338970422745, 0.07149643450975418, -0.01051111426204443, -0.01503442507237196, -0.01923832669854164, -0.09277807176113129, -0.008811017498373985, -0.025730334222316742, -0.011318848468363285, 0.026004960760474205, 0.013193491846323013, 0.05592338368296623, -0.07710710167884827, 0.01077062077820301, -0.05204102396965027, -0.011776845902204514, 0.04679800942540169, -0.006324325688183308, 0.07979235798120499, -0.05994

!pip install chromadb

import osimport erniebotfrom typing import Dict, List, Optional import chromadbfrom chromadb.api.types import Documents, EmbeddingFunction, Embeddingsdef embed_query(content):    response = erniebot.Embedding.create(    model="ernie-text-embedding",    input=[content])    result = response.get_result()    print(result)      return resultclass ErnieEmbeddingFunction(EmbeddingFunction):     def __call__(self, input: Documents) -> Embeddings:        embeddings = []        for text in input:            response = embed_query(text)            try:                embedding = response[0]                   embeddings.append(embedding)            except (IndexError, TypeError, KeyError) as e:                print(f"Error processing text: {text}, Error: {e}")        return embeddingschroma_client = chromadb.Client()# chroma_client = chromadb.PersistentClient(path="chromac") #数据保存硬盘位置 可选collection = chroma_client.create_collection(name="demo", embedding_function=ErnieEmbeddingFunction())print(collection)

from langchain.text_splitter import CharacterTextSplitterfrom langchain.vectorstores import Chromafrom langchain.document_loaders import TextLoader loader = TextLoader('./傅盛AI大课逐字稿.txt',encoding='utf-8')documents = loader.load()text_splitter = CharacterTextSplitter(chunk_size=600, chunk_overlap=20)docs = text_splitter.split_documents(documents)docs

import uuiddocs_list=[]metadatas=[]ids=[]for item in docs:    docs_list.append(item.page_content)    metadatas.append({"source": "傅盛AI大课逐字稿"})    ids.append(str(uuid.uuid4())) collection.add(    documents=docs_list,    metadatas=metadatas,    ids=ids)

query = "周鸿祎说见VC有两种错误的思维方式，分别是什么"results = collection.query(    query_texts=[query],    n_results=2) content=results['documents'][0][ ] prompt=f"""用户问题:{query}<context>{content}</context>根据<context>里的知识点回答用户问题"""response = erniebot.ChatCompletion.create(model="ernie-bot-4", messages=[{"role": "user", "content": prompt}])print(response.get_result())#周鸿祎说见VC有两种错误的思维方式，分别是：##1. 用过去的方式套今天的人工智能，比如比喻成OS。一旦比喻成操作系统，就得出结论全世界两套到三套，你觉得必然会被垄断、没有机会了，这种是典型的刻舟求剑。#2. 人容易对已经成功的事委曲求全，对于创新的新生代创业者容易求全责备。特别是有些做VC容易犯这个错误，比如OpenAI做成了，已经证明了，是个傻子都能看到OpenAI做的很成功，我们容易对它顶礼膜拜，恨不得跪下。对创业者很多还不成形的想法，因为八字没有一撇，光看到了你的很多缺点，这种价值观是不对的，容易Miss掉一些有潜力的项目。

def main(query):    results = collection.query(    query_texts=[query],    n_results=2)    content=results['documents'][0]    prompt=f"""    用户问题:{query}    <context>    {content}    </context>    根据<context>里的知识点回答用户问题    """    response = erniebot.ChatCompletion.create(model="ernie-bot-4", messages=[{"role": "user", "content": prompt}])    return response.get_result()query=input("请输入您要查询的问题：")print(main(query))

未来已来，将至已至！