LlamaIndex

Log inLog into community

Find answers from the community

Updated 3 months ago

any idea how I can extend this to PDFs

any idea how I can extend this to PDFs

·

any idea how I can extend this to PDFs instead of web pages?

L

t

7 comments

I think you can just load PDFs instead of web files?

i.e. in this section: https://docs.llamaindex.ai/en/latest/examples/agent/multi_document_agents.html#setup-and-download-data

Just do

Plain Text

data = {}
pdf_paths = [...]
for pdf_path in pdf_paths:
  data[pdf_paths] = SimpleDirectoryReader(input_files=[pdf_path]).load_data()

oh, wow. amazing.

do the pdf's have to be turned into vector embeddings?

that happens later on in the notebook 👍

It creates a vector store index for each path essentially

brilliant

this is my chat with my friend Laurent 🙂

Attachment

Screenshot_2023-12-29_at_4.30.58_PM.png

Add a reply

Sign up and join the conversation on Discord