RegressionMadeSimple (RMS)

A minimalist machine learning wrapper for lazy/skilled devs who want to skip the boilerplate. Built as a clean backdoor to scikit-learn.

---

🚀 Quickstart

import regressionmadesimple as rms

# Load dataset
df = rms.Preworks.readcsv("./your_data.csv")

# Train a linear regression model
model = rms.Linear(dataset=df, colX="feature", colY="target")

# Predict new values
predicted = model.predict([[5.2], [3.3]])

# Plot prediction vs. test
model.plot_predict([[5.2], [3.3]], predicted).show()

---

📦 Features

• 🧠 Wraps sklearn's most-used regression model(s) in a friendly API
• 📊 Built-in Plotly visualizations -- planned later support for matplotlib (? on this)
• 🔬 Designed for quick prototyping and educational use
• 🧰 Utility functions via preworks:
  • readcsv(path) — load CSV
  • create_random_dataset(...) — create random datasets (for demos)
• One-liner regression setup
• .summary() and .plot() for quick insight
• Global config system: rms.options
• Accepts pre-split data (X_train, y_train, etc.)
• Easily extendable — logistic, trees, etc. coming soon
• MIT Licensed

---

Project LINK

https://unknownuserfrommars.github.io/regressionmadesimple/

PS: Changelog also can be accessed from there

---

✅ Installation

pip install regressionmadesimple

Or install the dev version:

git clone https://github.com/Unknownuserfrommars/regressionmadesimple.git
cd regressionmadesimple
pip install -e .

---

📁 Project Structure

regressionmadesimple/
├── __init__.py
├── base_class.py
├── linear.py
├── logistic.py        # (soon)
├── tree.py            # (soon)
├── utils_preworks.py

---

🧪 Tests

Coming soon under a /tests folder using pytest

---

📜 License

MIT License

---

🛠 Author

Made with ❤️ by Unknownuserfrommars :)

---

🌌 Ideas for Future Versions

• Logistic() and DecisionTree() models
• .summary() for all models
• Export/save models
• Visual explainability (feature importance, SHAP)

---

⭐ Star this project if you like lazy ML.