---
language: en
license: mit
tags:
- intent-classification
- assistant
- scikit-learn
- sentence-transformers
model-type: text-classification
pipeline_tag: text-classification
base_model: sentence-transformers/all-MiniLM-L6-v2
widget:
  - text: "increase volume"
  - text: "open gmail"
  - text: "restart system"
---

# 🧠 Jarvis Intent Classifier

A lightweight CPU-friendly intent classification model that detects user commands  
and returns both a natural response and a system action command.

This model is ideal for building local AI assistants that can interact with your operating system  
(e.g., open apps, control volume, restart, sleep, or play music).

---

## 🏗️ Model Overview

| Component | Description |
|------------|--------------|
| **Base Model** | `sentence-transformers/all-MiniLM-L6-v2` (384-dim embeddings) |
| **Classifier** | `LogisticRegression` (scikit-learn) |
| **Dataset** | Custom OS command dataset with examples, responses, and Python actions |
| **Frameworks** | `sentence-transformers`, `scikit-learn` |
| **Runs on** | CPU (no GPU needed) |
| **License** | MIT |

---

## 🧩 Example Input → Output

**Input:**
```text
restart my computer

{
  "intent": "system_restart",
  "response": "System restart initiated...",
  "action": "os.system('shutdown /r /t 1')"
}


## 📜 License

This model is released under the **MIT License**.  
You may use, modify, and distribute it freely with attribution.

---

## 📖 Citation

If you use this model, please cite:

@software{jarvis_intent_classifier,
  title={Jarvis Intent Classifier},
  author={Arul Krishnan},
  year={2025},
  publisher={Hugging Face Hub},
  url={https://huggingface.co/KrishnanDevilking/jarvis-intent-classifier}
}