Generate Embeddings: SQL and PL/SQL Examples

In these examples, you can see how to generate one or more vector embeddings from text strings and PDF documents.

The resulting embeddings are numerical representations of the input data, and capture the semantic meaning or context of that data.

• Convert Text String to Embedding Within and Outside Oracle Database
  Perform a text-to-embedding transformation by accessing either a vector embedding model in ONNX format (stored in the database) or a third-party vector embedding model.
• Convert File to Text to Chunks to Embeddings Within Oracle Database
  First convert a PDF file to text, split the text into chunks, and then create vector embeddings on each chunk by accessing a vector embedding model stored in the database.
• Convert File to Embeddings Within Oracle Database
  Directly extract vector embeddings from a PDF document, using a single-step statement, by accessing a vector embedding model stored in the database.
• Generate and Use Embeddings for an End-to-End Search
  First generate vector embeddings from textual content by using a vector embedding model stored in the database, and then populate and query a vector index. At query time, you also vectorize the query criteria on the fly.