weaviate_quickstart_collection_import_and_near_text_search.py

1import weaviate
2import weaviate.classes as wvc
3import os
4import requests
5import json
6
7# Best practice: store your credentials in environment variables
8# For a free managed instance: https://console.weaviate.cloud/
9WCS_URL = os.getenv("WCS_URL")
10WCS_API_KEY = os.getenv("WCS_API_KEY")
11OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
12
13client = weaviate.connect_to_wcs(
14    cluster_url=WCS_URL,
15    auth_credentials=weaviate.auth.AuthApiKey(WCS_API_KEY),
16    headers={
17        "X-OpenAI-Api-Key": OPENAI_API_KEY  # Replace with your inference API key
18    }
19)
20
21try:
22    # 1. Create a collection
23    questions = client.collections.create(
24        name="Question",
25        vectorizer_config=wvc.config.Configure.Vectorizer.text2vec_openai(),  # Use OpenAI to vectorize text
26        generative_config=wvc.config.Configure.Generative.openai()            # Use OpenAI for RAG
27    )
28
29    # 2. Load data
30    resp = requests.get('https://raw.githubusercontent.com/weaviate-tutorials/quickstart/main/data/jeopardy_tiny.json')
31    data = json.loads(resp.text)
32
33    # 3. Insert objects
34    with questions.get_bulk_writer() as batch:
35        for item in data:
36            batch.add_object({
37                "question": item["Question"],
38                "answer": item["Answer"],
39                "category": item["Category"],
40            })
41
42    # 4. Perform a vector search
43    response = questions.query.near_text(
44        query="biology",
45        limit=2
46    )
47
48    for obj in response.objects:
49        print(f"ID: {obj.uuid}")
50        print(f"Data: {json.dumps(obj.properties, indent=2)}")
51
52finally:
53    client.close()  # Ensure the connection is closed