README.md

metadata

title: DevDeCode
emoji: 💡
colorFrom: indigo
colorTo: purple
sdk: gradio
sdk_version: 5.32.0
app_file: app.py
pinned: false

💡 DevDeCode: Explain Code with LLM

DevDeCode is a simple but powerful code explanation tool powered by Google's Gemma 2B and LangChain. Paste any code snippet (Java, Python, etc.), and get a natural-language explanation to help you understand what it does.

🧠 Built using Hugging Face Transformers, LangChain, and Gradio.

---

🚀 Features

• 📜 Explain small code snippets using LLM
• 🧩 Supports languages like Java, Python (others can be added)
• 🌐 Web-based interface via Gradio
• 🔄 Can be extended with memory & history tracking

---

🛠 Tech Stack

Tool / Library	Purpose
transformers	Load & run Gemma 2B
langchain	LLM integration & prompt logic
huggingface_hub	Auth & model access
gradio	Frontend UI
torch	Model backend (PyTorch)
langchain-huggingface	LangChain-HF wrapper

---

🧪 How to Use

Just paste a code snippet into the textbox and click submit. The model will explain what the code does.

⚠️ Note: The response time may vary depending on your GPU and input length. Model is gated — ensure you're authenticated.

---

🧩 Example Prompt

public class Factorial { public static int factorial(int n) { if (n <= 1) return 1; return n * factorial(n - 1); } }

👆 Get a clear explanation of recursion, factorial logic, and base conditions.

---

🔐 Hugging Face Token

This model uses a gated Google Gemma 2B model. Make sure your Hugging Face token is set via:

• Hugging Face Spaces secrets (HF_TOKEN)
• Or via huggingface-cli login locally.

---

🧠 Coming Soon

• Memory / Chat history support
• Upload files for batch code analysis
• Multi-language explanation support

---

👨‍💻 Created by @itzsudipta 🧪 Try it now: https://huggingface.co/spaces/sudiptaverse/DataSynth