Skip to content

Instantly share code, notes, and snippets.

@tim-pinecone
tim-pinecone / pinecone_sparse_embedding.py
Last active April 21, 2026 21:16
from a CSV generate sparse embeddings and save them in a parquet file formatted for bulk import
from pinecone.grpc import PineconeGRPC as Pinecone
import os
import json
import dotenv
import pandas as pd
import pyarrow as pa
import pyarrow.parquet as pq
from pathlib import Path
dotenv.load_dotenv()
@tim-pinecone
tim-pinecone / Upsert_768_Pinecone-gRPC.py
Last active January 8, 2026 03:57
Upsert_768_Pinecone-gRPC
"""
Upload vectors from a parquet file to Pinecone using gRPC for improved performance.
Requirements:
- Set PINECONE_API_KEY environment variable or add it to a .env file
- Install dependencies: uv add "pinecone[grpc]" pandas pyarrow python-dotenv
- Change name of parquet file
"""
import pandas as pd