Quantcast
Channel: Weaviate Community Forum - Latest posts
Viewing all articles
Browse latest Browse all 3590

Using DataType.TEXT_ARRAY as a datatype for a feature/column causes problems with LangChain's Document and retrieval systems

$
0
0

Hi @kumaran14,

It’s lovely to have you here—welcome to the community!

LangChain’s Document class is designed to handle individual text entries, with the page_content attribute specifically expecting a single string (str).

https://api.python.langchain.com/en/latest/documents/langchain_core.documents.base.Document.html

This design choice aligns with the framework’s focus on processing and analyzing individual documents or text segments efficiently.

When integrating with Weaviate, if your data is stored as a TEXT_ARRAY, it doesn’t directly match the expected input type for LangChain’s Document class. This mismatch lead to validation errors, as the system anticipates a single string rather than an array.

Regards,
Mohamed Shahin,
Weaviate Support Engineer


Viewing all articles
Browse latest Browse all 3590

Trending Articles