scikitelearn-collections

Elegant, production-ready extensions for Scikit-learn pipelines
Save time, build faster, scale better 🚀

---

🔍 Overview

scikitelearn-collections is a curated collection of robust utilities, transformers, wrappers, and experiment tools built on top of the Scikit-learn ecosystem. It helps you streamline model development, experiment tracking, and pipeline customization — all with full Scikit-learn compatibility.

---

✨ Features

• ✅ Plug-and-play Pipeline and ColumnTransformer components
• ✅ Drop-in feature generators (dates, text, outliers, etc.)
• ✅ Advanced custom transformers and meta-estimators
• ✅ Support for nested cross-validation and custom scorers
• ✅ Compatible with GridSearchCV and RandomizedSearchCV
• ✅ Simple model evaluation wrappers with logging
• ✅ Utility functions for feature selection, data cleaning, and split strategies
• ✅ Modular design for experimentation & reproducibility
• ✅ Clean, tested, and production-grade Python code
• ✅ 100% compatible with Scikit-learn’s API & best practices

---

📦 Installation

Requirements

• Python 3.8+
• scikit-learn >= 1.0
• numpy, pandas, joblib

Install via pip (PyPI release coming soon)

pip install scikitelearn-collections

Until then, you can clone manually:

git clone https://github.com/your-username/scikitelearn-collections.git
cd scikitelearn-collections
pip install -e .

---

🚀 Quick Start

from sklearn.pipeline import Pipeline
from scikitelearn_collections.transformers import DateFeatureGenerator, OutlierRemover
from sklearn.linear_model import LogisticRegression

pipeline = Pipeline([
    ("date_features", DateFeatureGenerator(columns=["signup_date"])),
    ("remove_outliers", OutlierRemover(method="zscore", threshold=3.0)),
    ("classifier", LogisticRegression())
])

pipeline.fit(X_train, y_train)

---

🧠 Modules & Components

Module	Description
transformers/	Custom transformers (dates, outliers, encodings, etc.)
pipelines/	Reusable ML pipelines with preprocessing and modeling
wrappers/	Model wrappers for enhanced evaluation, prediction, and logging
validators/	Custom cross-validation strategies and metric calculators
utils/	Helper utilities for splits, selection, diagnostics
examples/	Real-world usage examples in Jupyter notebooks

---

📁 Project Structure

scikitelearn-collections/
│
├── transformers/         # Custom transformers
├── pipelines/            # Ready-to-use ML pipelines
├── wrappers/             # Model and metric wrappers
├── utils/                # Helper functions and classes
├── validators/           # Scoring & validation strategies
├── examples/             # Example notebooks and scripts
├── tests/                # Unit tests
└── README.md             # You're here!

---

🧪 Examples

Explore the examples/ directory for practical Jupyter notebooks:

• ✅ Binary classification with preprocessing
• ✅ Regression with feature engineering
• ✅ Outlier detection & removal
• ✅ Cross-validation with custom scoring
• ✅ Hyperparameter tuning with pipeline integration

---

✅ Contributing

We ❤️ contributions! To contribute:

1. Fork this repository
2. Create a new branch: git checkout -b feature/your-feature
3. Write clean, tested code
4. Ensure all tests pass with pytest
5. Submit a pull request 🚀

---

🧪 Testing

All modules include unit tests in the tests/ directory. Run:

pytest

We use Black for code formatting and expect all code to follow PEP8 guidelines.

---

📄 License

This project is licensed under the MIT License.

---

🙌 Acknowledgements

• Built with ❤️ using Scikit-learn
• Inspired by real-world ML use-cases in research & production
• Thanks to open-source contributors and community ideas

---

📬 Contact

Have questions or suggestions? Open an issue or start a discussion!

---

Let your pipelines be elegant, reusable, and powerful. — scikitelearn-collections