Trying to load unstructured data into your Weaviate database with high-quality and enrichment? Check out this post on the Weaviate blog, which runs through how to use Aryn's Sycamore engine for document ETL. It goes over the steps for processing a PDF dataset, including chunking, enriching, cleaning, embedding, and loading vectors and metadata into a Weaviate vector index. You can also check out the Sycamore script for the ETL job in this notebook.
If you are interested in learning more about the Sycamore connector for Weaviate, visit the documentation.
コメント