Quick Start
Follow the steps below to install the VectorX SDK and set up your environment.
Requirements
- Python ≥ 3.8
- pip (Python package manager)
- Works on Linux, macOS, and Windows
Installation
To install the SDK:
pip install vecx
Set Up VectorX DB and Create Index
Following snippet can be followed to initialize vectorx and index creation.
from vecx.vectorx import VectorX
# Initialize client with your API token
vx = VectorX(token="your-token-here")
# Create a new index
vx.create_index(
name="my_index",
dimension=768, # Your vector dimension
space_type="cosine" # Distance metric (cosine, l2, ip)
)
# Get index reference
index = vx.get_index(name="my_index")
# Insert vectors
index.upsert([
{
"id": "doc1",
"vector": [0.1, 0.2, 0.3, ...], # Your vector data with specified dimension
"meta": {"text": "Example document"},
"filter":{"category": "reference"} # Optional filter
}
])
# Query similar vectors
results = index.query(
vector=[0.2, 0.3, 0.4, ...], # Query vector with specified dimenension
top_k=10,
filter=[{"category": {"$eq":"reference"}}] # Optional filter
)
# Process results
for item in results:
print(f"ID: {item['id']}, Similarity: {item['similarity']}")
print(f"Metadata: {item['meta']}")
Last updated on