💻Setup Qyver

Getting Started with Qyver: Installation & Example

Install the Qyver Library

To use Qyver, install the package via pip.

%pip install qyver

As a Python Script: Ensure your Python version is between 3.10.x and 3.12.x:

$ python -V
Python 3.10.9

If your Python version is outside this range, consider using pyenv to manage it.

Upgrade pip and install Qyver:

$ python -m pip install --upgrade pip
$ python -m pip install qyver

---

Running an Example

The first run may take longer as it downloads the necessary embedding model.

import json
import os
from qyver import framework as sl

# Define the product schema
class Product(sl.Schema):
    id: sl.IdField
    description: sl.String
    rating: sl.Integer

product = Product()

# Define embedding spaces
description_space = sl.TextSimilaritySpace(
    text=product.description, model="Alibaba-NLP/gte-large-en-v1.5"
)
rating_space = sl.NumberSpace(
    number=product.rating, min_value=1, max_value=5, mode=sl.Mode.MAXIMUM
)
index = sl.Index([description_space, rating_space], fields=[product.rating])

# Define query parameters, allowing natural language interpretation
query = (
    sl.Query(
        index,
        weights={
            description_space: sl.Param("description_weight"),
            rating_space: sl.Param("rating_weight"),
        },
    )
    .find(product)
    .similar(
        description_space,
        sl.Param(
            "description_query",
            description="The text in the user's query that refers to product descriptions.",
        ),
    )
    .limit(sl.Param("limit"))
    .with_natural_query(
        sl.Param("natural_language_query"),
        sl.OpenAIClientConfig(api_key=os.environ["OPEN_AI_API_KEY"], model="gpt-4o")
    )
)

# Run the app in-memory (supports server & Apache Spark executors too!)
source = sl.InMemorySource(product)
executor = sl.InMemoryExecutor(sources=[source], indices=[index])
app = executor.run()

# Ingest product data
source.put([
    {"id": 1, "description": "Budget toothbrush in black color. Just what you need.", "rating": 1},
    {"id": 2, "description": "High-end toothbrush created with no compromises.", "rating": 5},
    {"id": 3, "description": "A toothbrush created for the smart 21st-century man.", "rating": 3},
])

# Execute a query
result = app.query(query, natural_query="best toothbrushes", limit=1)

# Inspect extracted parameters
print(json.dumps(result.knn_params, indent=2))

# Display results (returns the highest-rated product)
result.to_pandas()

This example showcases multi-attribute vector search, combining text embeddings and numerical rankings for better retrieval accuracy.