Find answers from the community

Updated 4 months ago

I am unable to get the page label and

At a glance

The community member is unable to get the page label and file name for PDFs in the llamaparse library. Other community members suggest trying the JSON mode and accessing the source nodes in the response object to retrieve the document details, including metadata and segmentation information. They indicate that the response object format may depend on the LLM (Language Model) or Vector DB used. One community member confirms that the source nodes contain the required information and provides an example of how to access them.

Useful resources
I am unable to get the page label and file name in for PDFs in llamaparse
W
v
d
10 comments
Did you try the JSON mode?
Is there a similar method for getting json response back (with the used document details) from query()?
I don't think so! All the used nodes can be accessed via source nodes in the response object itself.
ok, that sounds familiar. Those source nodes will contain the various metadata, document details, and segmentation (where in file) information?
Is there a reference to what the response object looks like? Does it depend on the LLM / Vector DB?
When you do .query() and get response you can check all the retrieved source nodes like this: print(response.source_nodes)
It worked for me
Add a reply
Sign up and join the conversation on Discord