If I've already created an index using GTPineconeIndex ...

LawSnap · 2023-04-02T23:14:41.105Z

@kapa.ai If I've already created an index using GTPineconeIndex and saved to pinecone -- how do I load it in next time so I don't have to recreate the index each time?

----------------------
To load an existing Pinecone index, you can follow these steps:

Import the necessary libraries and initialize Pinecone with your API key:

Plain Text

from gpt_index import GPTPineconeIndex, SimpleDirectoryReader
import pinecone

api_key = "<api_key>"
pinecone.init(api_key=api_key, environment="us-west1-gcp")

Load the existing Pinecone index:

Plain Text

index_name = "quickstart"  # Replace with your index name
pinecone_index = pinecone.Index(index_name)

Load your documents using SimpleDirectoryReader:

Plain Text

documents = SimpleDirectoryReader('../paul_graham_essay/data').load_data()

Create a GPTPineconeIndex instance using the loaded Pinecone index:

Plain Text

index = GPTPineconeIndex(documents, pinecone_index=pinecone_index)

Now you can use the index object to query the existing Pinecone index without recreating it each time.

Source: Pinecone Index Construction/Querying (https://gpt-index.readthedocs.io/en/latest/how_to/integrations/vector_stores.html#using-a-vector-store-as-an-index)
----------------------
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
----------------------

Find answers from the community

If I've already created an index using GTPineconeIndex and saved to pinecone -- how do I l