----------------------
Yes, there is a way to persist or store the DocumentSummaryIndex in a vector store. You can use the
storage_context.persist() method to persist the simple vector store to disk. This method is used in the context of LlamaIndex's storage context, which is by default in-memory.
Here is an example of how you can use it:
# initialize storage context (by default it's in-memory)
from llama_index.core import StorageContext
storage_context = StorageContext.from_defaults()
# add documents to the docstore
storage_context.docstore.add_documents(nodes)
# Define Summary Index and Vector Index over Same Data
from llama_index.core import SummaryIndex
from llama_index.core import VectorStoreIndex
summary_index = SummaryIndex(nodes, storage_context=storage_context)
vector_index = VectorStoreIndex(nodes, storage_context=storage_context)
# persist simple vector store to disk
storage_context.persist()
In this example,
nodes is the data you want to index. The
SummaryIndex and
VectorStoreIndex are created with the data and the storage context. Finally, the
persist() method is called on the storage context to store the vector store to disk.
Please note that the
persist() method can also take a directory as an argument if you want to specify where to store the data, like so:
storage_context.persist(persist_dir="./storage").
For more details, you can refer to the
LlamaIndex documentation ().
As for storing the DocumentSummaryIndex in a pg vector, the provided knowledge sources do not contain any information on this.
----------------------
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
----------------------
