Welcome to the PDF Chat Assistant! This application allows you to upload PDF files and chat with them using the power of AI. Powered by LangChain, FAISS, and Groq's LLM, it brings your static documents to life by giving you answers to your questions based on their content!
To get started with the PDF Chat Assistant, you will need to have:
- Python 3.8+
- Streamlit
- Hugging Face Transformers
- FAISS
- Groq API Token
-
Clone the Repository
git clone https://github.com/MonsieurNam/ChatWF.git
-
Install the Required Packages
Install all dependencies by running:
pip install -r requirements.txt
-
Set up Environment Variables
Create a
.envfile in the root directory and add your Groq API Token:GROQ_API_TOKEN=your_groq_api_key_here -
Setup vectorsotre file
Run preprocess_storevector.py to make vectorstore.pkl
-
Run the Application
Start the application using the following command:
streamlit run app.py for original ver streamlit run newapp.py for fix ver with some edit streamlit run mix.py for static ver
-
Upload PDFs
Use the document uploader on the right side of the page to upload your PDF files.
-
Ask Questions
Once your documents are processed, you can start asking questions based on the content of your PDFs in the chat interface.
-
Customize the AI
Modify the system prompt to change how the AI responds. For example, you can ask it to be more casual, technical, or creative!
- Streamlit: Frontend interface and chat system.
- LangChain: Manages conversational logic and connects your queries with document content.
- Hugging Face Embeddings: Transforms document text into meaningful vector representations.
- FAISS: Efficiently stores and retrieves document vectors to match relevant content with user questions.
- Groq API: The powerhouse behind generating intelligent responses based on document data.
- Upload PDFs: The user uploads PDFs that need to be processed.
- Text Processing: Text is extracted and split into chunks for easier embedding.
- Vectorization: The document chunks are embedded using Hugging Face models.
- Conversational Retrieval: The AI searches through the document vectors to retrieve relevant information and answers questions.
- System Prompt: Customize the behavior of the AI by changing the system prompt.
- Temperature: Adjust the temperature to control the creativity of responses.
- Python: Core programming language.
- Streamlit: For the web interface.
- PyPDF2: For extracting text from PDFs.
- LangChain & FAISS: For conversational AI and document vectorization.
- Groq API: Powers the conversational responses.
If you encounter any issues, don't worry! The app is equipped with robust error handling for:
- PDF Processing Failures: Receive user-friendly error messages if a document fails to process.
- Embedding Issues: Alerts you when there are issues creating embeddings for your document.
- API Problems: Notifies you if there’s a problem connecting to the Groq API.
The chat interface is built for maximum readability and style:
- User Messages: Dark theme with light text.
- AI Responses: Highlighted in a contrasting color to distinguish between user and bot messages.
- Smooth UI: Responsive and interactive, designed for ease of use!
- Multi-Language Support: Expanding the tool to support documents in multiple languages.
- Improved Query Handling: Adding support for more complex queries and contextual conversations.
- Larger Document Handling: Optimizing performance for bigger datasets and more complex documents.
Feedback and contributions to improve this project are welcomed! Feel free to submit pull requests or open issues.
This project is licensed under the MIT License. See the LICENSE file for more details.
Enjoy chatting with your PDFs! 🚀