大模型从入门到应用——LangChain：链（Chains）-[链与索引： 文本摘要（Summarization）、HyDE和向量数据库的文本生成]_created a chunk of size 5070, which is longer than

分类目录：《大模型从入门到应用》总目录

LangChain系列文章：

• 基础知识
• 快速入门
  • 安装与环境配置
  • 链（Chains）、代理（Agent:）和记忆（Memory）
  • 快速开发聊天模型
• 模型（Models）
  • 基础知识
  • 大型语言模型（LLMs）
    • 基础知识
    • LLM的异步API、自定义LLM包装器、虚假LLM和人类输入LLM（Human Input LLM）
    • 缓存LLM的调用结果
    • 加载与保存LLM类、流式传输LLM与Chat Model响应和跟踪tokens使用情况
  • 聊天模型（Chat Models）
    • 基础知识
    • 使用少量示例和响应流式传输
  • 文本嵌入模型
    • Aleph Alpha、Amazon Bedrock、Azure OpenAI、Cohere等
    • Embaas、Fake Embeddings、Google Vertex AI PaLM等
• 提示（Prompts）
  • 基础知识
  • 提示模板
    • 基础知识
    • 连接到特征存储
    • 创建自定义提示模板和含有Few-Shot示例的提示模板
    • 部分填充的提示模板和提示合成
    • 序列化提示信息
  • 示例选择器（Example Selectors）
  • 输出解析器（Output Parsers）
• 记忆（Memory）
  • 基础知识
  • 记忆的类型
    • 会话缓存记忆、会话缓存窗口记忆和实体记忆
    • 对话知识图谱记忆、对话摘要记忆和会话摘要缓冲记忆
    • 对话令牌缓冲存储器和基于向量存储的记忆
  • 将记忆添加到LangChain组件中
  • 自定义对话记忆与自定义记忆类
  • 聊天消息记录
  • 记忆的存储与应用
• 索引（Indexes）
  • 基础知识
  • 文档加载器（Document Loaders）
  • 文本分割器（Text Splitters）
  • 向量存储器（Vectorstores）
  • 检索器（Retrievers）
• 链（Chains）
  • 基础知识
  • 通用功能
    • 自定义Chain和Chain的异步API
    • LLMChain和RouterChain
    • SequentialChain和TransformationChain
    • 链的保存（序列化）与加载（反序列化）
  • 链与索引
    • 文档分析和基于文档的聊天
    • 问答的基础知识
    • 图问答（Graph QA）和带来源的问答（Q&A with Sources）
    • 检索式问答
    • 文本摘要（Summarization）、HyDE和向量数据库的文本生成
• 代理（Agents）
  • 基础知识
  • 代理类型
  • 自定义代理（Custom Agent）
  • 自定义MRKL代理
  • 带有ChatModel的LLM聊天自定义代理和自定义多操作代理（Custom MultiAction Agent）
  • 工具
    • 基础知识
    • 自定义工具（Custom Tools）
    • 多输入工具和工具输入模式
    • 人工确认工具验证和Tools作为OpenAI函数
  • 工具包（Toolkit）
  • 代理执行器（Agent Executor）
    • 结合使用Agent和VectorStore
    • 使用Agents的异步API和创建ChatGPT克隆
    • 处理解析错误、访问中间步骤和限制最大迭代次数
    • 为代理程序设置超时时间和限制最大迭代次数和为代理程序和其工具添加共享内存
  • 计划与执行
• 回调函数（Callbacks）

---

文本摘要（Summarization）

本节介绍如何使用LangChain对文档列表进行文本摘要。它涵盖了三种不同的链式类型：stuff、map_reduce和refine。有关这些链的类型的更详细解释，可以参考文章《大模型从入门到应用——LangChain：链（Chains）-[链与索引：问答的基础知识]》。

准备数据

首先，我们需要准备数据。在本节示例中，我们从一个长文档中创建多个文档：

from langchain import OpenAI, PromptTemplate, LLMChain
from langchain.text_splitter import CharacterTextSplitter
from langchain.chains.mapreduce import MapReduceChain
from langchain.prompts import PromptTemplate

llm = OpenAI(temperature=0)

text_splitter = CharacterTextSplitter()
with open("../../state_of_the_union.txt") as f:
    state_of_the_union = f.read()
texts = text_splitter.split_text(state_of_the_union)

from langchain.docstore.document import Document

docs = [Document(page_content=t) for t in texts[:3]]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

快速入门

如果我们只想尽快开始，请按照以下推荐方式操作：

from langchain.chains.summarize import load_summarize_chain

chain = load_summarize_chain(llm, chain_type="map_reduce")
chain.run(docs)
1
2
3
4

输出：

' In response to Russian aggression in Ukraine, the United States and its allies are taking action to hold Putin accountable, including economic sanctions, asset seizures, and military assistance. The US is also providing economic and humanitarian aid to Ukraine, and has passed the American Rescue Plan and the Bipartisan Infrastructure Law to help struggling families and create jobs. The US remains unified and determined to protect Ukraine and the free world.'
1

stuff类型的链

本节展示了使用stuff类型的链进行摘要的结果。

chain = load_summarize_chain(llm, chain_type="stuff")
chain.run(docs)
1
2

输出：

' In his speech, President Biden addressed the crisis in Ukraine, the American Rescue Plan, and the Bipartisan Infrastructure Law. He discussed the need to invest in America, educate Americans, and build the economy from the bottom up. He also announced the release of 60 million barrels of oil from reserves around the world, and the creation of a dedicated task force to go after the crimes of Russian oligarchs. He concluded by emphasizing the need to Buy American and use taxpayer dollars to rebuild America.'
1

自定义提示

我们可以在此对话中使用自己的提示，在这个例子中，我们将用意大利语回复：

prompt_template = """Write a concise summary of the following:


{text}


CONCISE SUMMARY IN ITALIAN:"""
PROMPT = PromptTemplate(template=prompt_template, input_variables=["text"])
chain = load_summarize_chain(llm, chain_type="stuff", prompt=PROMPT)
chain.run(docs)
1
2
3
4
5
6
7
8
9
10

输出：

"\n\nIn questa serata, il Presidente degli Stati Uniti ha annunciato una serie di misure per affrontare la crisi in Ucraina, causata dall'aggressione di Putin. Ha anche annunciato l'invio di aiuti economici, militari e umanitari all'Ucraina. Ha anche annunciato che gli Stati Uniti e i loro alleati stanno imponendo sanzioni economiche a Putin e stanno rilasciando 60 milioni di barili di petrolio dalle riserve di tutto il mondo. Inoltre, ha annunciato che il Dipartimento di Giustizia degli Stati Uniti sta creando una task force dedicata ai crimini degli oligarchi russi. Il Presidente ha anche annunciato l'approvazione della legge bipartitica sull'infrastruttura, che prevede investimenti per la ricostruzione dell'America. Questo porterà a creare posti"
1

map_reduce类型的链

本节展示了使用map_reduce类型的链进行摘要生成的结果。

chain = load_summarize_chain(llm, chain_type="map_reduce")
chain.run(docs)
1
2

输出：

" In response to Russia's aggression in Ukraine, the United States and its allies have imposed economic sanctions and are taking other measures to hold Putin accountable. The US is also providing economic and military assistance to Ukraine, protecting NATO countries, and releasing oil from its Strategic Petroleum Reserve. President Biden and Vice President Harris have passed legislation to help struggling families and rebuild America's infrastructure."
1

中间步骤

如果我们想要检查map_reduce链的中间步骤，我们也可以返回它们。这可以通过return_map_steps变量来实现。

chain = load_summarize_chain(OpenAI(temperature=0), chain_type="map_reduce", return_intermediate_steps=True)
chain({"input_documents": docs}, return_only_outputs=True)
1
2

输出：

{'map_steps': [" In response to Russia's aggression in Ukraine, the United States has united with other freedom-loving nations to impose economic sanctions and hold Putin accountable. The U.S. Department of Justice is also assembling a task force to go after the crimes of Russian oligarchs and seize their ill-gotten gains.",
  ' The United States and its European allies are taking action to punish Russia for its invasion of Ukraine, including seizing assets, closing off airspace, and providing economic and military assistance to Ukraine. The US is also mobilizing forces to protect NATO countries and has released 30 million barrels of oil from its Strategic Petroleum Reserve to help blunt gas prices. The world is uniting in support of Ukraine and democracy, and the US stands with its Ukrainian-American citizens.',
  " President Biden and Vice President Harris ran for office with a new economic vision for America, and have since passed the American Rescue Plan and the Bipartisan Infrastructure Law to help struggling families and rebuild America's infrastructure. This includes creating jobs, modernizing roads, airports, ports, and waterways, replacing lead pipes, providing affordable high-speed internet, and investing in American products to support American jobs."],
 'output_text': " In response to Russia's aggression in Ukraine, the United States and its allies have imposed economic sanctions and are taking other measures to hold Putin accountable. The US is also providing economic and military assistance to Ukraine, protecting NATO countries, and passing legislation to help struggling families and rebuild America's infrastructure. The world is uniting in support of Ukraine and democracy, and the US stands with its Ukrainian-American citizens."}
1
2
3
4

自定义提示

我们可以在此对话中使用自己的提示，在这个例子中，我们将用意大利语回复：

prompt_template = """Write a concise summary of the following:


{text}


CONCISE SUMMARY IN ITALIAN:"""
PROMPT = PromptTemplate(template=prompt_template, input_variables=["text"])
chain = load_summarize_chain(OpenAI(temperature=0), chain_type="map_reduce", return_intermediate_steps=True, map_prompt=PROMPT, combine_prompt=PROMPT)
chain({"input_documents": docs}, return_only_outputs=True)
1
2
3
4
5
6
7
8
9
10

输出：

{'intermediate_steps': ["\n\nQuesta sera, ci incontriamo come democratici, repubblicani e indipendenti, ma soprattutto come americani. La Russia di Putin ha cercato di scuotere le fondamenta del mondo libero, ma ha sottovalutato la forza della gente ucraina. Gli Stati Uniti e i loro alleati stanno ora imponendo sanzioni economiche a Putin e stanno tagliando l'accesso della Russia alla tecnologia. Il Dipartimento di Giustizia degli Stati Uniti sta anche creando una task force dedicata per andare dopo i crimini degli oligarchi russi.",
  "\n\nStiamo unendo le nostre forze con quelle dei nostri alleati europei per sequestrare yacht, appartamenti di lusso e jet privati di Putin. Abbiamo chiuso lo spazio aereo americano ai voli russi e stiamo fornendo più di un miliardo di dollari in assistenza all'Ucraina. Abbiamo anche mobilitato le nostre forze terrestri, aeree e navali per proteggere i paesi della NATO. Abbiamo anche rilasciato 60 milioni di barili di petrolio dalle riserve di tutto il mondo, di cui 30 milioni dalla nostra riserva strategica di petrolio. Stiamo affrontando una prova reale e ci vorrà del tempo, ma alla fine Putin non riuscirà a spegnere l'amore dei popoli per la libertà.",
  "\n\nIl Presidente Biden ha lottato per passare l'American Rescue Plan per aiutare le persone che soffrivano a causa della pandemia. Il piano ha fornito sollievo economico immediato a milioni di americani, ha aiutato a mettere cibo sulla loro tavola, a mantenere un tetto sopra le loro teste e a ridurre il costo dell'assicurazione sanitaria. Il piano ha anche creato più di 6,5 milioni di nuovi posti di lavoro, il più alto numero di posti di lavoro creati in un anno nella storia degli Stati Uniti. Il Presidente Biden ha anche firmato la legge bipartitica sull'infrastruttura, la più ampia iniziativa di ricostruzione della storia degli Stati Uniti. Il piano prevede di modernizzare le strade, gli aeroporti, i porti e le vie navigabili in"],
 'output_text': "\n\nIl Presidente Biden sta lavorando per aiutare le persone che soffrono a causa della pandemia attraverso l'American Rescue Plan e la legge bipartitica sull'infrastruttura. Gli Stati Uniti e i loro alleati stanno anche imponendo sanzioni economiche a Putin e tagliando l'accesso della Russia alla tecnologia. Stanno anche sequestrando yacht, appartamenti di lusso e jet privati di Putin e fornendo più di un miliardo di dollari in assistenza all'Ucraina. Alla fine, Putin non riuscirà a spegnere l'amore dei popoli per la libertà."}
1
2
3
4

自定义的MapReduceChain

我们可以使用多个输入的提示。在这个示例中，我们将使用MapReduceChain来回答关于我们的代码的特定问题：

from langchain.chains.combine_documents.map_reduce import MapReduceDocumentsChain
from langchain.chains.combine_documents.stuff import StuffDocumentsChain

map_template_string = """Give the following python code information, generate a description that explains what the code does and also mention the time complexity.
Code:
{code}

Return the the description in the following format:
name of the function: description of the function
"""


reduce_template_string = """Give the following following python fuctions name and their descritpion, answer the following question
{code_description}
Question: {question}
Answer:
"""

MAP_PROMPT = PromptTemplate(input_variables=["code"], template=map_template_string)
REDUCE_PROMPT = PromptTemplate(input_variables=["code_description", "question"], template=reduce_template_string)

llm = OpenAI()

map_llm_chain = LLMChain(llm=llm, prompt=MAP_PROMPT)
reduce_llm_chain = LLMChain(llm=llm, prompt=REDUCE_PROMPT)

generative_result_reduce_chain = StuffDocumentsChain(
    llm_chain=reduce_llm_chain,
    document_variable_name="code_description",
)

combine_documents = MapReduceDocumentsChain(
    llm_chain=map_llm_chain,
    combine_document_chain=generative_result_reduce_chain,
    document_variable_name="code",
)

map_reduce = MapReduceChain(
    combine_documents_chain=combine_documents,
    text_splitter=CharacterTextSplitter(separator="\n##\n", chunk_size=100, chunk_overlap=0),
)
code = """
def bubblesort(list):
   for iter_num in range(len(list)-1,0,-1):
      for idx in range(iter_num):
         if list[idx]>list[idx+1]:
            temp = list[idx]
            list[idx] = list[idx+1]
            list[idx+1] = temp
    return list
##
def insertion_sort(InputList):
   for i in range(1, len(InputList)):
      j = i-1
      nxt_element = InputList[i]
   while (InputList[j] > nxt_element) and (j >= 0):
      InputList[j+1] = InputList[j]
      j=j-1
   InputList[j+1] = nxt_element
   return InputList
##
def shellSort(input_list):
   gap = len(input_list) // 2
   while gap > 0:
      for i in range(gap, len(input_list)):
         temp = input_list[i]
         j = i
   while j >= gap and input_list[j - gap] > temp:
      input_list[j] = input_list[j - gap]
      j = j-gap
      input_list[j] = temp
   gap = gap//2
   return input_list

"""
map_reduce.run(input_text=code, question="Which function has a better time complexity?")
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76

输出：

Created a chunk of size 247, which is longer than the specified 100
Created a chunk of size 267, which is longer than the specified 100





'shellSort has a better time complexity than both bubblesort and insertion_sort, as it has a time complexity of O(n^2), while the other two have a time complexity of O(n^2).'
1
2
3
4
5
6
7
8

refine类型的链

本节展示了使用refine类型的链进行文本摘要的结果。

chain = load_summarize_chain(llm, chain_type="refine")

chain.run(docs)
1
2
3

输出：

"\n\nIn response to Russia's aggression in Ukraine, the United States has united with other freedom-loving nations to impose economic sanctions and hold Putin accountable. The U.S. Department of Justice is also assembling a task force to go after the crimes of Russian oligarchs and seize their ill-gotten gains. We are joining with our European allies to find and seize the assets of Russian oligarchs, including yachts, luxury apartments, and private jets. The U.S. is also closing off American airspace to all Russian flights, further isolating Russia and adding an additional squeeze on their economy. The U.S. and its allies are providing support to the Ukrainians in their fight for freedom, including military, economic, and humanitarian assistance. The U.S. is also mobilizing ground forces, air squadrons, and ship deployments to protect NATO countries. The U.S. and its allies are also releasing 60 million barrels of oil from reserves around the world, with the U.S. contributing 30 million barrels from its own Strategic Petroleum Reserve. In addition, the U.S. has passed the American Rescue Plan to provide immediate economic relief for tens of millions of Americans, and the Bipartisan Infrastructure Law to rebuild America and create jobs. This investment will"
1

中间步骤

如果我们想要检查refine链的中间步骤，我们也可以返回它们。这可以通过return_map_steps变量来实现。

chain = load_summarize_chain(OpenAI(temperature=0), chain_type="refine", return_intermediate_steps=True)

chain({"input_documents": docs}, return_only_outputs=True)
1
2
3

输出：

{'refine_steps': [" In response to Russia's aggression in Ukraine, the United States has united with other freedom-loving nations to impose economic sanctions and hold Putin accountable. The U.S. Department of Justice is also assembling a task force to go after the crimes of Russian oligarchs and seize their ill-gotten gains.",
  "\n\nIn response to Russia's aggression in Ukraine, the United States has united with other freedom-loving nations to impose economic sanctions and hold Putin accountable. The U.S. Department of Justice is also assembling a task force to go after the crimes of Russian oligarchs and seize their ill-gotten gains. We are joining with our European allies to find and seize the assets of Russian oligarchs, including yachts, luxury apartments, and private jets. The U.S. is also closing off American airspace to all Russian flights, further isolating Russia and adding an additional squeeze on their economy. The U.S. and its allies are providing support to the Ukrainians in their fight for freedom, including military, economic, and humanitarian assistance. The U.S. is also mobilizing ground forces, air squadrons, and ship deployments to protect NATO countries. The U.S. and its allies are also releasing 60 million barrels of oil from reserves around the world, with the U.S. contributing 30 million barrels from its own Strategic Petroleum Reserve. Putin's war on Ukraine has left Russia weaker and the rest of the world stronger, with the world uniting in support of democracy and peace.",
  "\n\nIn response to Russia's aggression in Ukraine, the United States has united with other freedom-loving nations to impose economic sanctions and hold Putin accountable. The U.S. Department of Justice is also assembling a task force to go after the crimes of Russian oligarchs and seize their ill-gotten gains. We are joining with our European allies to find and seize the assets of Russian oligarchs, including yachts, luxury apartments, and private jets. The U.S. is also closing off American airspace to all Russian flights, further isolating Russia and adding an additional squeeze on their economy. The U.S. and its allies are providing support to the Ukrainians in their fight for freedom, including military, economic, and humanitarian assistance. The U.S. is also mobilizing ground forces, air squadrons, and ship deployments to protect NATO countries. The U.S. and its allies are also releasing 60 million barrels of oil from reserves around the world, with the U.S. contributing 30 million barrels from its own Strategic Petroleum Reserve. In addition, the U.S. has passed the American Rescue Plan to provide immediate economic relief for tens of millions of Americans, and the Bipartisan Infrastructure Law to rebuild America and create jobs. This includes investing"],
 'output_text': "\n\nIn response to Russia's aggression in Ukraine, the United States has united with other freedom-loving nations to impose economic sanctions and hold Putin accountable. The U.S. Department of Justice is also assembling a task force to go after the crimes of Russian oligarchs and seize their ill-gotten gains. We are joining with our European allies to find and seize the assets of Russian oligarchs, including yachts, luxury apartments, and private jets. The U.S. is also closing off American airspace to all Russian flights, further isolating Russia and adding an additional squeeze on their economy. The U.S. and its allies are providing support to the Ukrainians in their fight for freedom, including military, economic, and humanitarian assistance. The U.S. is also mobilizing ground forces, air squadrons, and ship deployments to protect NATO countries. The U.S. and its allies are also releasing 60 million barrels of oil from reserves around the world, with the U.S. contributing 30 million barrels from its own Strategic Petroleum Reserve. In addition, the U.S. has passed the American Rescue Plan to provide immediate economic relief for tens of millions of Americans, and the Bipartisan Infrastructure Law to rebuild America and create jobs. This includes investing"}
1
2
3
4

自定义提示

我们可以在此对话中使用自己的提示，在这个例子中，我们将用意大利语回复：

prompt_template = """Write a concise summary of the following:


{text}


CONCISE SUMMARY IN ITALIAN:"""
PROMPT = PromptTemplate(template=prompt_template, input_variables=["text"])
refine_template = (
    "Your job is to produce a final summary\n"
    "We have provided an existing summary up to a certain point: {existing_answer}\n"
    "We have the opportunity to refine the existing summary"
    "(only if needed) with some more context below.\n"
    "------------\n"
    "{text}\n"
    "------------\n"
    "Given the new context, refine the original summary in Italian"
    "If the context isn't useful, return the original summary."
)
refine_prompt = PromptTemplate(
    input_variables=["existing_answer", "text"],
    template=refine_template,
)
chain = load_summarize_chain(OpenAI(temperature=0), chain_type="refine", return_intermediate_steps=True, question_prompt=PROMPT, refine_prompt=refine_prompt)
chain({"input_documents": docs}, return_only_outputs=True)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

输出：

{'intermediate_steps': ["\n\nQuesta sera, ci incontriamo come democratici, repubblicani e indipendenti, ma soprattutto come americani. La Russia di Putin ha cercato di scuotere le fondamenta del mondo libero, ma ha sottovalutato la forza della gente ucraina. Insieme ai nostri alleati, stiamo imponendo sanzioni economiche, tagliando l'accesso della Russia alla tecnologia e bloccando i suoi più grandi istituti bancari dal sistema finanziario internazionale. Il Dipartimento di Giustizia degli Stati Uniti sta anche assemblando una task force dedicata per andare dopo i crimini degli oligarchi russi.",
  "\n\nQuesta sera, ci incontriamo come democratici, repubblicani e indipendenti, ma soprattutto come americani. La Russia di Putin ha cercato di scuotere le fondamenta del mondo libero, ma ha sottovalutato la forza della gente ucraina. Insieme ai nostri alleati, stiamo imponendo sanzioni economiche, tagliando l'accesso della Russia alla tecnologia, bloccando i suoi più grandi istituti bancari dal sistema finanziario internazionale e chiudendo lo spazio aereo americano a tutti i voli russi. Il Dipartimento di Giustizia degli Stati Uniti sta anche assemblando una task force dedicata per andare dopo i crimini degli oligarchi russi. Stiamo fornendo più di un miliardo di dollari in assistenza diretta all'Ucraina e fornendo assistenza militare,",
  "\n\nQuesta sera, ci incontriamo come democratici, repubblicani e indipendenti, ma soprattutto come americani. La Russia di Putin ha cercato di scuotere le fondamenta del mondo libero, ma ha sottovalutato la forza della gente ucraina. Insieme ai nostri alleati, stiamo imponendo sanzioni economiche, tagliando l'accesso della Russia alla tecnologia, bloccando i suoi più grandi istituti bancari dal sistema finanziario internazionale e chiudendo lo spazio aereo americano a tutti i voli russi. Il Dipartimento di Giustizia degli Stati Uniti sta anche assemblando una task force dedicata per andare dopo i crimini degli oligarchi russi. Stiamo fornendo più di un miliardo di dollari in assistenza diretta all'Ucraina e fornendo assistenza militare."],
 'output_text': "\n\nQuesta sera, ci incontriamo come democratici, repubblicani e indipendenti, ma soprattutto come americani. La Russia di Putin ha cercato di scuotere le fondamenta del mondo libero, ma ha sottovalutato la forza della gente ucraina. Insieme ai nostri alleati, stiamo imponendo sanzioni economiche, tagliando l'accesso della Russia alla tecnologia, bloccando i suoi più grandi istituti bancari dal sistema finanziario internazionale e chiudendo lo spazio aereo americano a tutti i voli russi. Il Dipartimento di Giustizia degli Stati Uniti sta anche assemblando una task force dedicata per andare dopo i crimini degli oligarchi russi. Stiamo fornendo più di un miliardo di dollari in assistenza diretta all'Ucraina e fornendo assistenza militare."}
1
2
3
4

HyDE

本节介绍如何使用假设文档嵌入（HyDE）技术，详细内容可参阅论文《Precise Zero-Shot Dense Retrieval without Relevance Labels》。在高层次上，HyDE是一种嵌入技术，它接受查询，生成一个假设的答案，然后将生成的文档嵌入，并将其作为最终的示例。

为了使用HyDE，我们需要提供一个基础嵌入模型，以及一个可以用于生成这些文档的LLMChain。默认情况下，HyDE类带有一些默认的提示，但我们也可以创建自己的提示。

from langchain.llms import OpenAI
from langchain.embeddings import OpenAIEmbeddings
from langchain.chains import LLMChain, HypotheticalDocumentEmbedder
from langchain.prompts import PromptTemplate
base_embeddings = OpenAIEmbeddings()
llm = OpenAI()
# Load with `web_search` prompt
embeddings = HypotheticalDocumentEmbedder.from_llm(llm, base_embeddings, "web_search")
# Now we can use it as any embedding class!
result = embeddings.embed_query("Where is the Taj Mahal?")
1
2
3
4
5
6
7
8
9
10

多次生成

我们还可以生成多个文档，然后将这些文档的嵌入组合起来。默认情况下，我们通过取平均值来组合这些嵌入。我们可以通过改变用于生成文档的LLM来实现这一点，使其返回多个结果。

multi_llm = OpenAI(n=4, best_of=4)
embeddings = HypotheticalDocumentEmbedder.from_llm(multi_llm, base_embeddings, "web_search")
result = embeddings.embed_query("Where is the Taj Mahal?")
1
2
3

使用自定义提示

除了使用预配置的提示外，我们还可以轻松构建自己的提示，并将其用于生成文档的LLMChain中。如果我们知道查询所在的领域，这样做可能很有用，因为我们可以将提示设定为生成与该领域更相似的文本。在下面的示例中，我们将其设定为生成关于国情咨文的文本：

prompt_template = """Please answer the user's question about the most recent state of the union address
Question: {question}
Answer:"""
prompt = PromptTemplate(input_variables=["question"], template=prompt_template)
llm_chain = LLMChain(llm=llm, prompt=prompt)
embeddings = HypotheticalDocumentEmbedder(llm_chain=llm_chain, base_embeddings=base_embeddings)
result = embeddings.embed_query("What did the president say about Ketanji Brown Jackson")
1
2
3
4
5
6
7

使用HyDE

现在我们有了HyDE，我们可以像使用其他嵌入类一样使用它。下面是使用HyDE在国情咨文示例中查找相似段落的示例：

from langchain.text_splitter import CharacterTextSplitter
from langchain.vectorstores import Chroma

with open("../../state_of_the_union.txt") as f:
    state_of_the_union = f.read()
text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
texts = text_splitter.split_text(state_of_the_union)
docsearch = Chroma.from_texts(texts, embeddings)

query = "What did the president say about Ketanji Brown Jackson"
docs = docsearch.similarity_search(query)
1
2
3
4
5
6
7
8
9
10
11

日志输出：

Running Chroma using direct local API.
Using DuckDB in-memory for database. Data will be transient.
1
2

输入：

print(docs[0].page_content)
1

输出：

In state after state, new laws have been passed, not only to suppress the vote, but to subvert entire elections. 

We cannot let this happen. 

Tonight. I call on the Senate to: Pass the Freedom to Vote Act. Pass the John Lewis Voting Rights Act. And while you’re at it, pass the Disclose Act so Americans can know who is funding our elections. 

Tonight, I’d like to honor someone who has dedicated his life to serve this country: Justice Stephen Breyer—an Army veteran, Constitutional scholar, and retiring Justice of the United States Supreme Court. Justice Breyer, thank you for your service. 

One of the most serious constitutional responsibilities a President has is nominating someone to serve on the United States Supreme Court. 

And I did that 4 days ago, when I nominated Circuit Court of Appeals Judge Ketanji Brown Jackson. One of our nation’s top legal minds, who will continue Justice Breyer’s legacy of excellence.
1
2
3
4
5
6
7
8
9
10
11

向量数据库的文本生成

本节介绍如何使用LangChain在向量索引上进行文本生成。这对于我们希望生成能够利用大量自定义文本的文本非常有用，例如生成具有对先前编写的博客文章的理解的博客文章，或者可以参考产品文档的产品教程。

准备数据

首先，我们需要准备数据。在下面这个例子中，我们从Github获取了一个由Markdown文件组成的文档网站，并将它们拆分为足够小的文档。

from langchain.llms import OpenAI
from langchain.docstore.document import Document
import requests
from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.vectorstores import Chroma
from langchain.text_splitter import CharacterTextSplitter
from langchain.prompts import PromptTemplate
import pathlib
import subprocess
import tempfile

def get_github_docs(repo_owner, repo_name):
    with tempfile.TemporaryDirectory() as d:
        subprocess.check_call(
            f"git clone --depth 1 https://github.com/{repo_owner}/{repo_name}.git .",
            cwd=d,
            shell=True,
        )
        git_sha = (
            subprocess.check_output("git rev-parse HEAD", shell=True, cwd=d)
            .decode("utf-8")
            .strip()
        )
        repo_path = pathlib.Path(d)
        markdown_files = list(repo_path.glob("*/*.md")) + list(
            repo_path.glob("*/*.mdx")
        )
        for markdown_file in markdown_files:
            with open(markdown_file, "r") as f:
                relative_path = markdown_file.relative_to(repo_path)
                github_url = f"https://github.com/{repo_owner}/{repo_name}/blob/{git_sha}/{relative_path}"
                yield Document(page_content=f.read(), metadata={"source": github_url})

sources = get_github_docs("yirenlu92", "deno-manual-forked")

source_chunks = []
splitter = CharacterTextSplitter(separator=" ", chunk_size=1024, chunk_overlap=0)
for source in sources:
    for chunk in splitter.split_text(source.page_content):
        source_chunks.append(Document(page_content=chunk, metadata=source.metadata))
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40

日志输出：

Cloning into '.'...
1

设置Vector DB

现在我们将文档内容分块后，让我们将所有这些信息放入一个向量索引中，以便轻松检索。

search_index = Chroma.from_documents(source_chunks, OpenAIEmbeddings())
1

设置带有自定义提示的LLMChain

接下来，我们设置一个简单的LLMChain，但为其提供一个用于博客文章生成的自定义提示。需要注意的是，自定义提示是参数化的，并接受两个输入：

• context：从向量搜索中获取的文档
• topic：由用户提供

from langchain.chains import LLMChain
prompt_template = """Use the context below to write a 400 word blog post about the topic below:
    Context: {context}
    Topic: {topic}
    Blog post:"""

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

llm = OpenAI(temperature=0)

chain = LLMChain(llm=llm, prompt=PROMPT)
1
2
3
4
5
6
7
8
9
10
11
12
13

生成文本

最后，我们编写一个函数来将输入应用于链条。该函数接受一个输入参数topic。我们在向量索引中找到与该topic对应的文档，并将它们作为我们简单LLMChain中的额外上下文。

def generate_blog_post(topic):
    docs = search_index.similarity_search(topic, k=4)
    inputs = [{"context": doc.page_content, "topic": topic} for doc in docs]
    print(chain.apply(inputs))
generate_blog_post("environment variables")
1
2
3
4
5

输出：

[{'text': '\n\nEnvironment variables are a great way to store and access sensitive information in your Deno applications. Deno offers built-in support for environment variables with `Deno.env`, and you can also use a `.env` file to store and access environment variables.\n\nUsing `Deno.env` is simple. It has getter and setter methods, so you can easily set and retrieve environment variables. For example, you can set the `FIREBASE_API_KEY` and `FIREBASE_AUTH_DOMAIN` environment variables like this:\n\n```ts\nDeno.env.set("FIREBASE_API_KEY", "examplekey123");\nDeno.env.set("FIREBASE_AUTH_DOMAIN", "firebasedomain.com");\n\nconsole.log(Deno.env.get("FIREBASE_API_KEY")); // examplekey123\nconsole.log(Deno.env.get("FIREBASE_AUTH_DOMAIN")); // firebasedomain.com\n```\n\nYou can also store environment variables in a `.env` file. This is a great'}, {'text': '\n\nEnvironment variables are a powerful tool for managing configuration settings in a program. They allow us to set values that can be used by the program, without having to hard-code them into the code. This makes it easier to change settings without having to modify the code.\n\nIn Deno, environment variables can be set in a few different ways. The most common way is to use the `VAR=value` syntax. This will set the environment variable `VAR` to the value `value`. This can be used to set any number of environment variables before running a command. For example, if we wanted to set the environment variable `VAR` to `hello` before running a Deno command, we could do so like this:\n\n```\nVAR=hello deno run main.ts\n```\n\nThis will set the environment variable `VAR` to `hello` before running the command. We can then access this variable in our code using the `Deno.env.get()` function. For example, if we ran the following command:\n\n```\nVAR=hello && deno eval "console.log(\'Deno: \' + Deno.env.get(\'VAR'}, {'text': '\n\nEnvironment variables are a powerful tool for developers, allowing them to store and access data without having to hard-code it into their applications. In Deno, you can access environment variables using the `Deno.env.get()` function.\n\nFor example, if you wanted to access the `HOME` environment variable, you could do so like this:\n\n```js\n// env.js\nDeno.env.get("HOME");\n```\n\nWhen running this code, you\'ll need to grant the Deno process access to environment variables. This can be done by passing the `--allow-env` flag to the `deno run` command. You can also specify which environment variables you want to grant access to, like this:\n\n```shell\n# Allow access to only the HOME env var\ndeno run --allow-env=HOME env.js\n```\n\nIt\'s important to note that environment variables are case insensitive on Windows, so Deno also matches them case insensitively (on Windows only).\n\nAnother thing to be aware of when using environment variables is subprocess permissions. Subprocesses are powerful and can access system resources regardless of the permissions you granted to the Den'}, {'text': '\n\nEnvironment variables are an important part of any programming language, and Deno is no exception. Deno is a secure JavaScript and TypeScript runtime built on the V8 JavaScript engine, and it recently added support for environment variables. This feature was added in Deno version 1.6.0, and it is now available for use in Deno applications.\n\nEnvironment variables are used to store information that can be used by programs. They are typically used to store configuration information, such as the location of a database or the name of a user. In Deno, environment variables are stored in the `Deno.env` object. This object is similar to the `process.env` object in Node.js, and it allows you to access and set environment variables.\n\nThe `Deno.env` object is a read-only object, meaning that you cannot directly modify the environment variables. Instead, you must use the `Deno.env.set()` function to set environment variables. This function takes two arguments: the name of the environment variable and the value to set it to. For example, if you wanted to set the `FOO` environment variable to `bar`, you would use the following code:\n\n```'}]
1

参考文献：
[1] LangChain官方网站：https://www.langchain.com/
[2] LangChain