Support for loading differnet document types (xlsx,.doc,.txt)

[Krishnateja244]

Hi,

I have been using this package for a couple of days to develop a vectorcypher retriever-based chatbot. I am working on a data lake consisting of Excel sheets. From the api documentation, I only saw the support for processing PDF files, but not other file types. So, I would like to know if there is a module to process different file types to extract data from it?

[stellasia]

Hi @Krishnateja244 ,

Sorry for the late reply! At the moment this package is only for text-to-KG, text being raw text or in a PDF file. If your excel file has a "text" column you want to perform entity and relation extraction from, you can write your own DataLoader. But if you excel file is mostly structured, you can try the data importer.