from haystack import Pipelinefrom haystack.nodes import BM25Retriever, FARMReader
# 创建检索器和阅读器retriever = BM25Retriever(document_store=document_store)reader = FARMReader("deepset/roberta-base-squad2")
# 构建pipelinepipe = Pipeline()pipe.add_node(component=retriever, name="Retriever", inputs=["Query"])pipe.add_node(component=reader, name="Reader", inputs=["Retriever"])
# 执行查询results = pipe.run(query="What is the capital of France?")

ort Pipelinefrom haystack.nodes import EmbeddingRetriever
retriever = EmbeddingRetriever(    document_store=document_store,    embedding_model="sentence-transformers/multi-qa-mpnet-base-dot-v1")
pipe = Pipeline()pipe.add_node(component=retriever, name="Retriever", inputs=["Query"])
results = pipe.run(query="climate change impact")

from haystack import Pipelinefrom haystack.nodes import PromptNode, PromptTemplate
template = PromptTemplate(    prompt="""    Answer the question based on the given context. If the answer is not in the context, say "I don't know".        Context: {context}        Human: {query}    AI:     """,    output_parser=lambda x: x.split("AI: ")[-1].strip(),)
prompt_node = PromptNode(    model_name_or_path="gpt-3.5-turbo",    api_key="your-openai-api-key",    default_prompt_template=template)
pipe = Pipeline()pipe.add_node(component=retriever, name="Retriever", inputs=["Query"])pipe.add_node(component=prompt_node, name="PromptNode", inputs=["Retriever"])
result = pipe.run(    query="What are the effects of climate change?",    params={"Retriever": {"top_k": 3}})