Skip to main content
Open In Colab在 GitHub 上打开

如何使用时间加权向量存储检索器

此检索器结合使用语义相似性和时间衰减。

对它们进行评分的算法是：

semantic_similarity + (1.0 - decay_rate) ^ hours_passed

特别是hours_passed指自上次访问检索器中的对象以来经过的小时数，而不是自创建该对象以来经过的小时数。这意味着经常访问的对象将保持 “新鲜”。

from datetime import datetime, timedelta

import faiss
from langchain.retrievers import TimeWeightedVectorStoreRetriever
from langchain_community.docstore import InMemoryDocstore
from langchain_community.vectorstores import FAISS
from langchain_core.documents import Document
from langchain_openai import OpenAIEmbeddings

低衰减率

低decay rate（在这种情况下，为了极端起见，我们会将其设置为接近 0）意味着记忆将被“记住”更长的时间。一个decay rate的 0 表示永远不会忘记记忆，因此此检索器等效于向量查找。

# Define your embedding model
embeddings_model = OpenAIEmbeddings()
# Initialize the vectorstore as empty
embedding_size = 1536
index = faiss.IndexFlatL2(embedding_size)
vectorstore = FAISS(embeddings_model, index, InMemoryDocstore({}), {})
retriever = TimeWeightedVectorStoreRetriever(
    vectorstore=vectorstore, decay_rate=0.0000000000000000000000001, k=1
)
yesterday = datetime.now() - timedelta(days=1)
retriever.add_documents(
    [Document(page_content="hello world", metadata={"last_accessed_at": yesterday})]
)
retriever.add_documents([Document(page_content="hello foo")])
['73679bc9-d425-49c2-9d74-de6356c73489']
# "Hello World" is returned first because it is most salient, and the decay rate is close to 0., meaning it's still recent enough
retriever.invoke("hello world")
[Document(metadata={'last_accessed_at': datetime.datetime(2024, 10, 22, 16, 37, 40, 818583), 'created_at': datetime.datetime(2024, 10, 22, 16, 37, 37, 975074), 'buffer_idx': 0}, page_content='hello world')]

高衰减率

随着decay rate（例如，几个 9），则recency score快速变为 0！如果将此项一直设置为 1，则recency对于所有对象都是 0，这再次等效于向量查找。

# Define your embedding model
embeddings_model = OpenAIEmbeddings()
# Initialize the vectorstore as empty
embedding_size = 1536
index = faiss.IndexFlatL2(embedding_size)
vectorstore = FAISS(embeddings_model, index, InMemoryDocstore({}), {})
retriever = TimeWeightedVectorStoreRetriever(
    vectorstore=vectorstore, decay_rate=0.999, k=1
)
yesterday = datetime.now() - timedelta(days=1)
retriever.add_documents(
    [Document(page_content="hello world", metadata={"last_accessed_at": yesterday})]
)
retriever.add_documents([Document(page_content="hello foo")])
['379631f0-42c2-4773-8cc2-d36201e1e610']
# "Hello Foo" is returned first because "hello world" is mostly forgotten
retriever.invoke("hello world")
[Document(metadata={'last_accessed_at': datetime.datetime(2024, 10, 22, 16, 37, 46, 553633), 'created_at': datetime.datetime(2024, 10, 22, 16, 37, 43, 927429), 'buffer_idx': 1}, page_content='hello foo')]

虚拟时间

使用 LangChain 中的一些 utils，你可以模拟 time 组件。

from langchain_core.utils import mock_now
API 参考：mock_now
# Notice the last access time is that date time

tomorrow = datetime.now() + timedelta(days=1)

with mock_now(tomorrow):
    print(retriever.invoke("hello world"))
[Document(metadata={'last_accessed_at': MockDateTime(2024, 10, 23, 16, 38, 19, 66711), 'created_at': datetime.datetime(2024, 10, 22, 16, 37, 43, 599877), 'buffer_idx': 0}, page_content='hello world')]