Metadata-Version: 2.4
Name: quantiq
Version: 0.2.0
Summary: Confidence-aware AI toolkit with uncertainty estimation for Transformers and Deep Ensembles
Home-page: https://github.com/aryanator/QuantIQ
Author: Aryan Patil
Author-email: aryanator01@gmail.com
Project-URL: Documentation, https://github.com/aryanator/QuantIQ
Project-URL: Source, https://github.com/aryanator/QuantIQ
Project-URL: Bug Tracker, https://github.com/aryanator/QuantIQ/issues
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.7
Description-Content-Type: text/markdown
License-File: LICENCE
Requires-Dist: torch>=1.9.0
Requires-Dist: transformers>=4.30.0
Requires-Dist: matplotlib
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: license-file
Dynamic: project-url
Dynamic: requires-dist
Dynamic: requires-python
Dynamic: summary

# QuantIQ 🔍

**QuantIQ** is a unified Python library for building robust, uncertainty-aware deep learning systems. It brings together lightweight, modular tools to help researchers and practitioners gain insight into model reliability and risk through principled uncertainty estimation.
<img width="1024" height="939" alt="poster" src="https://github.com/user-attachments/assets/12d53dab-8cd9-4c2b-b5b8-7bf203007ff5" />


## 📦 What's Inside

Quantiq currently includes two powerful uncertainty quantification tools:

### 1. DropWise 🔁
A plug-and-play PyTorch/HuggingFace wrapper for Monte Carlo Dropout–based uncertainty estimation in Transformers.

- Supports classification, regression, QA, and token tagging
- Computes entropy, confidence, and class-wise variances
- Enables dropout during inference for Bayesian-style sampling

📖 [Full DropWise Documentation](https://github.com/aryanator/QuantIQ/blob/main/quantiq/dropwise/README.md)

---

### 2. SmartEnsemble 🧠
A deep ensemble wrapper for PyTorch models with support for adversarial training and dual-mode (epistemic + aleatoric) uncertainty estimation.

- Works with any PyTorch model
- Enables risk scoring and calibration
- Includes built-in visualization and prediction APIs

📖 [Full SmartEnsemble Documentation](https://github.com/aryanator/QuantIQ/blob/main/quantiq/smartensemble/README.md)

---

## 🔧 Installation

```bash
pip install quantiq
```

Or install from source:

```bash
git clone https://github.com/aryanator/QuantIQ.git
cd quantiq
pip install -e .
```

---

## 🧪 Use Cases

- Safety-critical predictions (medical AI, self-driving, finance)
- Uncertainty-aware active learning
- Robust ML pipelines with explainable confidence
- Research experiments involving confidence, entropy, risk

---

## 📚 Documentation & Examples

Explore examples, API usage, and task-specific walkthroughs in the GitHub repository:  
🔗 https://github.com/aryanator/QuantIQ


<img width="1189" height="590" alt="image" src="https://github.com/user-attachments/assets/16fcbc24-5f7e-463e-91af-f7b1806b90c5" />


---

## 📝 License

MIT License

---

Built by [Aryan Patil](https://github.com/aryanator) to make uncertainty estimation simpler, smarter, and production-ready.
