weaviate_cloud_quickstart_collection_import_and_neartext_search.py

1import weaviate
2import weaviate.classes.config as wc
3import json
4import requests
5
6# 1. Connect to Weaviate
7# Best practice: use environment variables for keys
8client = weaviate.connect_to_weaviate_cloud(
9    cluster_url="https://your-weaviate-endpoint.weaviate.network",  # Replace with your URL
10    auth_credentials=weaviate.auth.AuthApiKey("your-weaviate-api-key"),  # Replace with your API key
11    headers={
12        "X-OpenAI-Api-Key": "your-openai-api-key"  # Replace with your inference API key
13    }
14)
15
16try:
17    # 2. Create a collection
18    # The collection defines the data structure and how it should be vectorized
19    if client.collections.exists("Question"):
20        client.collections.delete("Question")
21
22    questions = client.collections.create(
23        name="Question",
24        vectorizer_config=wc.Configure.Vectorizer.text2vec_openai(),  # Use OpenAI to create vectors
25        generative_config=wc.Configure.Generative.openai()            # Use OpenAI for RAG
26    )
27
28    # 3. Import data
29    # Download sample 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    # Bulk insert data
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 search
43    # This performs a "nearText" semantic search
44    response = questions.query.near_text(
45        query="biology",
46        limit=2
47    )
48
49    for obj in response.objects:
50        print(f"ID: {obj.uuid}")
51        print(f"Data: {json.dumps(obj.properties, indent=2)}")
52
53finally:
54    client.close()  # Close the connection