newton-musyimi / multi-pdf-bot

# Multi-Pdf-Bot This code is a Streamlit application that allows users to chat with multiple PDF documents. It utilizes various libraries and modules to process the PDFs, extract text from them, create a conversational AI model, and provide responses to user queries based on the content of the PDFs. Let's go through the code step by step: 1. Importing Required Libraries: - `streamlit`: Used to create interactive web applications. - `dotenv`: Used to load environment variables. - `PdfReader` from `PyPDF2`: Used to read PDF documents. - Various modules from the `langchain` library: Used for text splitting, embeddings, vector stores, chat models, memory, chains, and LLMS (Language Learning Models). - `pandas`: Used for data manipulation and analysis. 2. Function Definitions: - `get_pdf_text(pdf_docs)`: Takes a list of PDF documents and extracts the text from them using `PdfReader`. - `get_text_chunks(text)`: Takes the extracted text as input and splits it into smaller chunks using a `CharacterTextSplitter` from the `langchain` library. This is done to enable efficient processing and retrieval of information. - `get_vectorstore(text_chunks)`: Takes the text chunks as input and creates a vector store using `OpenAIEmbeddings` from the `langchain` library. - `get_conversation_chain(vectorstore)`: Takes the vector store as input and creates a conversation chain using a chat model (`ChatOpenAI`) from the `langchain` library, along with a memory component (`ConversationBufferMemory`). This chain allows the system to engage in conversational interactions with the user based on the content of the PDFs. 3. `handle_userinput(user_question)`: Handles the user's input question. It sends the question to the conversation chain and retrieves the response. The conversation history is stored in the `st.session_state.chat_history` variable. The user's question and the chat history are displayed using HTML templates. 4. `main()`: The main function that serves as the entry point for the Streamlit application. It sets up the Streamlit page, handles user input, and processes the PDFs. - It first loads the environment variables and sets the page configuration. - It initializes the session state variables for conversation and chat history. - It creates a Streamlit sidebar where the user can upload multiple PDF documents. - When the user clicks the "Process" button, the PDFs are processed. - The PDFs' text is extracted using `get_pdf_text`. - The text is split into chunks using `get_text_chunks`. - A vector store is created using `get_vectorstore`. - A conversation chain is created using `get_conversation_chain`. - The chat history is cleared. - If a conversation chain exists, it checks if the chat history is empty. If so, it greets the user with a default message. - The user's question is input via a text input box, and the `handle_userinput` function is called to process the question. 5. The `if __name__ == '__main__':` block ensures that the `main()` function is executed when the script is run directly. Overall, this code sets up a Streamlit application that allows users to upload multiple PDF documents, processes the documents to enable conversational interactions, and provides responses based on the content of the PDFs using a conversational AI model. Run the Script: `streamlit run app.py`  # Multi-Pdf-Bot