readme updated finalv1
Browse files
README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
---
|
| 2 |
title: Pdfsense
|
| 3 |
-
emoji:
|
| 4 |
colorFrom: red
|
| 5 |
colorTo: red
|
| 6 |
sdk: streamlit
|
|
@@ -11,5 +11,59 @@ license: apache-2.0
|
|
| 11 |
short_description: PDF Answering Assistant
|
| 12 |
---
|
| 13 |
|
| 14 |
-
# PDFSense
|
| 15 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
title: Pdfsense
|
| 3 |
+
emoji: 📜
|
| 4 |
colorFrom: red
|
| 5 |
colorTo: red
|
| 6 |
sdk: streamlit
|
|
|
|
| 11 |
short_description: PDF Answering Assistant
|
| 12 |
---
|
| 13 |
|
| 14 |
+
# 📜 PDFSense : PDF Question Answering Assistant with Chat History
|
| 15 |
+
|
| 16 |
+
PDFSense is an LLM-powered Streamlit application that enables users to upload PDFs and ask questions based on the document's content. It uses a Retrieval-Augmented Generation (RAG) approach to provide accurate, context-aware answers by incorporating previous chat history of the current session.
|
| 17 |
+
|
| 18 |
+
[App in Hugging Face Space](https://huggingface.co/spaces/AkashVD26/pdfsense)
|
| 19 |
+
|
| 20 |
+
## 🚀 Features
|
| 21 |
+
- Upload and analyze PDF documents.
|
| 22 |
+
- Ask questions about the uploaded PDF in natural language.
|
| 23 |
+
- Retrieve answers using LangChain, FAISS indexing, and Hugging Face embeddings.
|
| 24 |
+
- Maintain conversation context for coherent responses.
|
| 25 |
+
|
| 26 |
+
## 📚 How It Works
|
| 27 |
+
- Upload PDF: Drag and drop your PDF file into the uploader.
|
| 28 |
+
- Ask Questions: Type a question about the PDF's content.
|
| 29 |
+
- Contextual Answers: PDFSense retrieves answers using FAISS and LLMs while maintaining chat history for context.
|
| 30 |
+
|
| 31 |
+
## 🛠️ Technologies Used
|
| 32 |
+
- Streamlit: Interactive web application framework.
|
| 33 |
+
- LangChain: Framework for creating LLM-based applications.
|
| 34 |
+
- FAISS: Vector search for efficient retrieval.
|
| 35 |
+
- Hugging Face: Pretrained embeddings for document processing.
|
| 36 |
+
- Groq: LLM used for generating responses.
|
| 37 |
+
- PyPDFLoader: Document loader for processing PDFs.
|
| 38 |
+
|
| 39 |
+
## 🧩 Prerequisites
|
| 40 |
+
Make sure you have the following prerequisites:
|
| 41 |
+
|
| 42 |
+
- [Python 3.8 and above](https://www.python.org)
|
| 43 |
+
- [Hugging Face account](https://huggingface.co)
|
| 44 |
+
- [Hugging Face Access Token](https://huggingface.co/settings/tokens)
|
| 45 |
+
- [Groq API key](https://console.groq.com/keys)
|
| 46 |
+
|
| 47 |
+
## 📦 Installation
|
| 48 |
+
If you want to use this locally on your system:
|
| 49 |
+
|
| 50 |
+
```
|
| 51 |
+
git clone https://github.com/Akashvarma26/PDFSense.git
|
| 52 |
+
```
|
| 53 |
+
|
| 54 |
+
```
|
| 55 |
+
pip install -r requirements.txt
|
| 56 |
+
```
|
| 57 |
+
|
| 58 |
+
## ▶️ Usage
|
| 59 |
+
Run the Streamlit app locally:
|
| 60 |
+
```
|
| 61 |
+
streamlit run app.py
|
| 62 |
+
```
|
| 63 |
+
|
| 64 |
+
## 🙋♂️ Acknowledgments
|
| 65 |
+
- [LangChain](https://www.langchain.com)
|
| 66 |
+
- [Hugging Face](https://huggingface.co)
|
| 67 |
+
- [FAISS](https://ai.meta.com/tools/faiss/)
|
| 68 |
+
- [Groq](https://groq.com)
|
| 69 |
+
- [streamlit](https://www.langchain.com)
|