AI-Based Code Generator — Setup Guide (Beginner Friendly)

This guide assumes you have never built a Python/AI project before. Follow every step in order.

---

Step 1: Install Python

1. Go to https://www.python.org/downloads/
2. Download Python 3.10 or higher
3. During install, check the box "Add Python to PATH" (Windows) — this step is easy to miss and causes most beginner errors
4. Verify install by opening a terminal (Command Prompt / PowerShell / Terminal) and typing:

   python --version
   

   You should see something like Python 3.11.4

---

Step 2: Get the project files

Put app.py, requirements.txt, and this README.md in one folder, e.g.:

ai-code-generator/
├── app.py
├── requirements.txt
└── README.md

---

Step 3: Open a terminal in that folder

• Windows: open the folder, click the address bar, type cmd, press Enter
• Mac/Linux: right-click the folder → "Open Terminal Here" (or cd into it manually)

---

Step 4: Create a virtual environment (recommended, keeps things clean)

python -m venv venv

Activate it:

• Windows: venv\Scripts\activate
• Mac/Linux: source venv/bin/activate

You'll know it worked because your terminal line will now start with (venv).

---

Step 5: Install the required libraries

pip install -r requirements.txt

This installs:

• streamlit → for the web UI
• openai → to talk to the AI model

---

Step 6: Get an OpenAI API key

1. Go to https://platform.openai.com/signup and create an account
2. Go to https://platform.openai.com/api-keys
3. Click "Create new secret key" and copy it (starts with sk-...)
4. Important: OpenAI requires a small prepaid balance (as little as $5) to use the API. Add billing at https://platform.openai.com/account/billing

💡 Don't want to pay? See the "Free Alternative" section at the bottom — you can use Hugging Face's free models instead.

---

Step 7: Run the app

In your terminal (with venv still active):

streamlit run app.py

Your browser will automatically open http://localhost:8501 showing your app.

---

Step 8: Use it

1. Paste your API key into the sidebar (or set it as an environment variable — see below)
2. Pick a language (Python, Java, C++, etc.)
3. Type a task, e.g. "function to check if a string is a palindrome"
4. Click Generate
5. See your code appear with an explanation below it

---

Optional: Set your API key permanently (so you don't paste it every time)

Instead of typing it in the sidebar each time, set it as an environment variable:

• Windows (PowerShell):

  setx OPENAI_API_KEY "sk-your-key-here"
  

  (then close and reopen your terminal)

• Mac/Linux:

  export OPENAI_API_KEY="sk-your-key-here"
  

  (add this line to your ~/.bashrc or ~/.zshrc to make it permanent)

---

Free Alternative: Hugging Face (no payment needed)

If you don't want to pay OpenAI, you can swap the generate_code() function to call a free Hugging Face model instead, e.g. bigcode/starcoder2-15b or Salesforce/codegen-350M-mono, using their free Inference API:

import requests

def generate_code_hf(prompt, hf_token):
    API_URL = "https://api-inference.huggingface.co/models/bigcode/starcoder2-15b"
    headers = {"Authorization": f"Bearer {hf_token}"}
    response = requests.post(API_URL, headers=headers, json={"inputs": prompt})
    return response.json()[0]["generated_text"]

Get a free Hugging Face token at https://huggingface.co/settings/tokens (free tier has rate limits, but is fine for a student project demo).

---

What to screenshot for your project report

1. The sidebar with settings
2. A generated Python example (simple, e.g. factorial or palindrome)
3. A generated Java or C++ example (shows multi-language support)
4. The History section with 2-3 past generations
5. One "failure case" — a vague or tricky prompt that produces imperfect code (great for your "Limitations" section)

---

Common errors and fixes

Error	Fix
streamlit: command not found	Your venv isn't activated — repeat Step 4
AuthenticationError	Your API key is wrong or has no billing set up
ModuleNotFoundError: No module named 'openai'	Run pip install -r requirements.txt again
App opens but blank	Refresh the browser tab, or check terminal for red error text

---

Next steps / ways to extend this project (for extra marks)

• Add a "Fix my code" mode: paste broken code, get a corrected version
• Add a "Convert language" mode: paste Python, get the same logic in Java
• Add syntax highlighting using streamlit-ace or st.code(result, language=lang)
• Deploy it for free on https://streamlit.io/cloud so you can share a live link
• Log all prompts/outputs to a CSV file to analyze accuracy for your report