search

arc53 / DocsGPT

GPT-powered chat for documentation, chat with your documents
https://app.docsgpt.cloud/
MIT License
14.45k stars 1.45k forks source link

🚀 Feature: Does it support Nought If not then addition of it. #528

Open rajveer43 opened 9 months ago

rajveer43 commented 9 months ago

🔖 Feature description

Nougat: Neural Optical Understanding for Academic Documents Scientific knowledge is predominantly stored in books and scientific journals, often in the form of PDFs However, the PDF format leads to a loss of semantic information, particularly for mathematical expressions), a Visual Transformer model that performs an Optical Character Recognition (OCR) task for processing scientific documents into a markup language, and demonstrate the effectiveness of our model on a new dataset of scientific documents. The proposed approach offers a promising solution to enhance the accessibility of scientific knowledge in the digital age, by bridging the gap between human-readable documents and machine-readable text. We release the models and code to accelerate future work on scientific text recognition.

🎤 Why is this feature needed ?

Nought has the potential to be a valuable tool for DocsGPT users

✌️ How do you aim to achieve this?

https://github.com/facebookresearch/nougat

🔄️ Additional Information

-

👀 Have you spent some time to check if this feature request has been raised before?

Are you willing to submit PR?

Yes I am willing to submit a PR!

staticGuru commented 9 months ago

Hi @rajveer43, Thank you for submitting your feature request. Could you please provide detailed guidance on how these requested features can be effectively integrated into our system? We would greatly appreciate your insights and explanations.

rajveer43 commented 9 months ago

Hi @rajveer43, Thank you for submitting your feature request. Could you please provide detailed guidance on how these requested features can be effectively integrated into our system? We would greatly appreciate your insights and explanations.

Hi its a LLM tool built for parsing scientific docs. so lets discuss first how it will be integrated and what additional thing will require.

dartpain commented 9 months ago

I think if we integrate it has to link to the file inside worker.py, ingest_worker.

  1. check if pdf has any text inside
  2. If no text parse it save as txt
  3. Delete uploaded pdf
  4. give it as txt into simple document reader.