copulas

What it is and what it does

Copulas is a Python library for learning multivariate distributions from numerical data and generating synthetic data that follows the same statistical patterns. You provide a table of real data, fit a copula model (choosing from options like Gaussian Copula, Vine Copulas, or Archimedian Copulas), and then sample new synthetic records that preserve the correlations and marginal distributions of the original. The library integrates with numpy, pandas, scipy for computation and plotly for visualization, allowing you to compare real and synthetic data side-by-side in 1D, 2D, and 3D plots.

The package is part of the Synthetic Data Vault Project and targets developers and data engineers building synthetic data pipelines. It exposes the learned model parameters for inspection and tuning, making it suitable for both exploratory work and production use where you need control over the generation process. The codebase is actively maintained, supports Python 3.9 through 3.14, and has no known security vulnerabilities.

Use it for:

• Generate privacy-preserving synthetic datasets for testing and development without exposing real customer or sensitive data
• Create balanced training datasets for machine learning by sampling from learned multivariate distributions
• Augment small datasets by learning their statistical structure and generating additional synthetic records
• Validate data pipelines and analytics by producing synthetic data with known statistical properties
• Compare real versus synthetic data distributions visually to verify model quality before deployment

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Copulas models multivariate statistical distributions and generates synthetic tabular data that preserves the statistical properties of the original dataset using copula functions.

Yes, with conditions. The package is actively maintained, has low install friction, and solves a real need for synthetic tabular data generation. However, the BUSL-1.1 license restricts commercial use until a future date—verify the license terms against your use case before committing. The pre-alpha development status suggests the API may change; for production use, pin the version and monitor releases.

Install

copulas on PyPI

pip install copulas

uv add copulas

poetry add copulas

Installing copulas

Before you install

Installation is straightforward with low friction; the package has four common runtime dependencies (numpy, pandas, plotly, scipy) and maintains active development with a recent release.

License in practice

The package uses BUSL-1.1 (Business Source License), which restricts commercial use until a future date and may require review before adoption in commercial projects.

Quickstart

pip install copulas

from copulas.datasets import sample_trivariate_xyz
from copulas.multivariate import GaussianMultivariate

real_data = sample_trivariate_xyz()
copula = GaussianMultivariate()
copula.fit(real_data)
synthetic_data = copula.sample(len(real_data))

Verify before relying

• Whether BUSL-1.1 license restrictions apply to your intended use case and timeline
• Performance characteristics and scalability limits for large datasets
• Specific copula types supported beyond Gaussian, Vine, and Archimedian mentioned in description

Package facts

Evidence: copulas-0.14.1-py3-none-any.whl

Keywords: copulas

Tags

More Artificial Intelligence packages