Introducing SimuBayes

---

SimuBayes is a powerful and user-friendly machine learning toolkit designed to streamline complex data analysis workflows—from preprocessing to advanced statistical modeling. Tailored for researchers, engineers, and data scientists, SimuBayes offers an integrated pipeline for data cleaning, Deep Gaussian Process (DeepGP) modeling, prediction, forward uncertainty quantification (UQ), and sensitivity analysis.

✅ Smart Data Cleaning
Effortlessly prepare and preprocess datasets for reliable and efficient modeling.

✅ Deep Gaussian Process (DeepGP) Modeling
Capture intricate, high-dimensional relationships using flexible Bayesian non-parametric models.

✅ Predictive Modeling with Uncertainty Bounds
Deliver accurate forecasts with quantified confidence intervals, essential for risk-sensitive applications.

✅ Forward Uncertainty Quantification (UQ)
Propagate input uncertainties through your models to support robust, informed decision-making.

✅ Sensitivity Analysis
Identify the most influential parameters using Morris screening (for quick filtering) and Sobol indices (for global sensitivity insights).

SimuBayes stands out by combining Bayesian robustness, computational efficiency, and usability in one seamless platform:

🔹 Integrated Workflow
No need to juggle multiple tools—go from raw data to interpretable insights in a single environment.

🔹 Bayesian Confidence
Built on DeepGPs, SimuBayes delivers trustworthy, uncertainty-aware predictions ideal for high-impact decision domains.

🔹 Insightful Sensitivity Analysis
Move beyond black-box predictions with interpretable insights into which inputs matter most.

🔹 User-Friendly & Scalable
Whether you're a beginner or an expert, SimuBayes offers an intuitive interface and scalable performance for large datasets.

• DeepGP Modeling:  https://youtu.be/2UqKFge2KR0
• Prediction, Forward UQ, and Sensitivity Analysis: https://youtu.be/DgXdScMfdLk

We welcome contributions, suggestions, and issues—feel free to open a pull request or contact us below or on GitHub!