Assignment #1 Development of a RAG App

Arnav Joshi's submission for Assignment #1 of INFO 5940.

---

🛠️ Prerequisites

Before starting, ensure you have the following installed on your system:

• Docker (Ensure Docker Desktop is running)
• VS Code
• VS Code Remote - Containers Extension
• Git
• OpenAI API Key

---

🚀 Setup Guide

The installation for this assignment should be nearly identical to cloning and setting up the repository straight from GitHub. However, due to minor issues in the setup that currently exist in the repository (and some inefficiencies), the instructions are in a different order and slightly modified.

---

1️⃣ Clone the Repository

Open a terminal and run:

git clone https://github.com/joshiarnav/INFO5940-A1.git
cd INFO5940-A1

2️⃣ Configure OpenAI API Key

Since docker-compose.yml expects environment variables, follow these steps:

➤ Option 1: Set the API Key in .env (Recommended)

1. Inside the project folder, create a .env file:

   ni .env

   or for Mac:

   touch .env

2. Add your API key and base URL:

   OPENAI_API_KEY=your-api-key-here
   OPENAI_BASE_URL=https://api.ai.it.cornell.edu/
   TZ=America/New_York
   

Now, your API key will be automatically loaded inside the container.

3️⃣ Open in VS Code with Docker

1. Open VS Code, navigate to the INFO5940-A1 folder.
2. Open the Command Palette (Ctrl+Shift+P or Cmd+Shift+P on Mac) and search for:

   Dev Containers: Rebuild and Reopen in Container
   

3. Select this option. VS Code will build and open the project inside the container.

📌 Note: If you don’t see this option, ensure that the Dev Containers extension is installed.

4️⃣ Run Streamlit App

1. Navigate to the INFO5940-A1 folder.
2. Run:

   streamlit run advanced_chat.py

3. Open a browser and navigate to http://localhost:8501 (or the port number shown in the terminal if different).
4. Upload and chat! Supports PDFs and text files (plural) and utilizes FAISS for vectorization, vector store, and vector similarity search. Utilizes OpenAI's Embeddings (to create vector embeddings) and LLMs (for chat).

---

📌 Changes Made

• Deleted old poetry.lock file (there was a dependency error on all tested machines in the existing poetry.lock file in the lecture-05 branch of INFO-5940).
• Added faiss-cpu package to the poetry packages (poetry add faiss-cpu).
• Added PyPDF2 package to the poetry packages (poetry add PyPDF2).
• Deleted any extraneous folders/files for the assignment:
  • /notebooks
  • /data
  • chat_with_pdf.py
  • chat_with_rag.py
  • Chatbot.py
  • poetry.lock (explained above)
  • Old README.md
  • summary.py
  • tokens.py
• Modified README.md:
  • Modified name of the extension for Dev Containers (Dev Containers: Rebuild and Reopen in Container). The original instructions in the lecture-05 branch of INFO-5940 were for Remote - Containers which is deprecated.
  • Moved API key setup before starting the container (this avoids rebuilding the container to add the API key).

---

📄 Sources

• Course Material (lecture-05 branch of INFO-5940)
• FAISS Documentation
• OpenAI Platform Page

---

🛠️ Troubleshooting

Container Fails to Start?

• Ensure Docker Desktop is running.

• Run docker-compose up --build again.

• If errors persist, delete existing containers with:

  docker-compose down

  Then restart:

  docker-compose up --build

Cannot Access Jupyter Notebook from outside VS Code?

• Ensure you’re using the correct port (8888).
• Run docker ps to check if the container is running.

OpenAI API Key Not Recognized?

• Check if .env is correctly created.
• Ensure docker-compose.yml includes env_file: - .env.
• Restart the container after making changes (docker-compose up --build).

---