[Question]: Weaviate cross-references

[pedrocr83]

Question Validation

• I have searched both the documentation and discord for an answer.

Question

I want to create two collections in Weaviate DB and use cross-references. How to implement this and save documents/nodes using VectorStoreIndex?

[dosubot]

To create two collections in Weaviate DB and save documents with cross-references using VectorStoreIndex, follow these steps:

1. Define Your Collections (Schemas): First, define the schema for each collection. For example:

   schema = {
       "classes": [
           {
               "class": "CollectionOne",
               "properties": [
                   {"name": "name", "dataType": ["string"]},
                   {"name": "referenceToTwo", "dataType": ["CollectionTwo"]}
               ],
           },
           {
               "class": "CollectionTwo",
               "properties": [
                   {"name": "description", "dataType": ["string"]},
                   {"name": "referenceToOne", "dataType": ["CollectionOne"]}
               ],
           }
       ]
   }
   client.schema.create(schema)

   This creates two collections with cross-references between them.

2. Add Documents to Collections: Prepare your documents, ensuring each has a unique identifier and the necessary fields, including references. Use the add method to add documents:

   # Example document for CollectionOne
   document_one = {
       "name": "Example Name",
       "referenceToTwo": [
           {
               "beacon": "weaviate://localhost/CollectionTwo/<uuid_of_collection_two_document>"
           }
       ]
   }
   
   # Example document for CollectionTwo
   document_two = {
       "description": "Example Description",
       "referenceToOne": [
           {
               "beacon": "weaviate://localhost/CollectionOne/<uuid_of_collection_one_document>"
           }
       ]
   }
   
   # Add documents to Weaviate
   client.data_object.create(data_object=document_one, class_name="CollectionOne")
   client.data_object.create(data_object=document_two, class_name="CollectionTwo")

   Replace <uuid_of_collection_one_document> and <uuid_of_collection_two_document> with the actual UUIDs of the documents you are referencing.

3. Query the Index: After adding documents, you can query the index using Weaviate's query capabilities to perform vector searches or hybrid searches based on your requirements.

Ensure you have weaviate-client installed and set up a connection to your Weaviate instance before proceeding with these steps. This approach will help you effectively manage data and create cross-references between collections in Weaviate.

Details

Sources

• [Question]: Help!! Using mongodb as document store and chroma as vector store for Auto Merging Retriever
• docs/docs/examples/vector_stores/WeaviateIndexDemo-Hybrid.ipynb
• llama-index-integrations/vector_stores/llama-index-vector-stores-weaviate/llama_index/vector_stores/weaviate/base.py
• llama-index-legacy/llama_index/legacy/vector_stores/weaviate.py
• docs/docs/examples/vector_stores/WeaviateIndex_auto_retriever.ipynb
• llama-index-integrations/vector_stores/llama-index-vector-stores-weaviate/llama_index/vector_stores/weaviate/utils.py
• llama-index-packs/llama-index-packs-sub-question-weaviate/llama_index/packs/sub_question_weaviate/base.py
• llama-index-integrations/readers/llama-index-readers-weaviate/llama_index/readers/weaviate/base.py

About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

To continue the conversation, mention @dosu.