Canopy And Pinecone Serverless #270
Replies: 2 comments 1 reply
-
@gfleetwood to use canopy in your own made index, I think it's more of a hardcoded set of metadata properties that Canopy expects to be there...as far as I know. I'm using Canopy with an index I made manually, using canopy with that index failed because it couldn't find some of the metadata properties it needed. I've changed the metadata in the documents to match the ones in the Canopy code and I'm getting RAG based responses from the Canopy REST API ...so that seems to work just fine. |
Beta Was this translation helpful? Give feedback.
-
As @coreation stated canopy expects a specific metadata structure so it is not possible to use it with existing indexes created beforehand. You can direct canopy to create indexes in the pod based project by passing a Pod Spec to the create_canopy_index function. Just make sure you are in the latest canopy version. You can pass something similar to this as a spec: {
"replicas": 1,
"pod_type": "p1.x1",
"shards": 1,
"environment": "us-east1-gcp"
} More info on how to create this dictionary is documented here: https://docs.pinecone.io/reference/create_index @gfleetwood please let me know if this answers your questions. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Hello. With the release of Pinecone Serverless I upgraded from the free to the standard plan. While canopy will now create a new serverless database, it appears under the free tier instead of my current plan. I submitted a support ticket last week which hasn't been answered, so I thought I'd ask here.
For reference, this is the code I'm using:
I also have two more questions:
Beta Was this translation helpful? Give feedback.
All reactions