plot-shift 0.2.0

A Python package for advanced plotting utilities.

PLOT-SHIFT

A growing collection of Python utilities for advanced data visualization, with a focus on exploratory data analysis. This project is a personal effort to gather and refine useful plotting functions developed during my data science studies and research.

Purpose

The goal of this package is to provide reusable, well-documented visualization tools that support a variety of data science workflows. Over time, new plotting functions and utilities will be added as they are created or improved.

Features

• Visualizations for cluster analysis, including silhouette diagram matrices
• Designed for integration with scikit-learn and other common data science libraries
• Usable in research, notebooks, and production code

Installation

Install via PDM (recommended):

pdm add plot-shift

Or with pip (after publishing to PyPI):

pip install plot-shift

Note: If you are using the code locally (not from PyPI), make sure your PYTHONPATH includes the src/ directory, or install the package in editable mode:

pip install -e ./src

Usage

Import and use the available plotting functions in your data science projects. Example for clustering visualization:

from plot_shift.silhouette_diagram import silhouette_diagram

# Example usage:
silhouette_diagram(
    k,                          # Number of clusters
    y_pred,                     # Predicted cluster labels
    silhouette_coefficients,    # Silhouette coefficients for sample
    silhouette_score,           # Silhouette score for sample
    ax,                         # Matplotlib Axes object
)

More plotting functions will be added over time. See the notebooks/ directory for practical examples and demonstrations.

Contributing

Contributions and suggestions are welcome! If you have a useful plotting function or improvement, feel free to open an issue or pull request.

License

MIT License

Author

Koen van Esterik