Installation

The package is available on Test PyPI and can be built from source.

From Test PyPI

The package is currently available on Test PyPI as pyfastcpd:

# Install Armadillo first (required for building C++ extension)
# macOS:
brew install armadillo

# Ubuntu/Debian:
sudo apt-get install libarmadillo-dev

# Then install pyfastcpd
pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ pyfastcpd

Note

Test PyPI currently only has source distribution (no pre-built wheels). You must install Armadillo before running pip install.

From Source

For development or if you want the latest features:

System Requirements:

  • Python ≥ 3.8

  • C++17 compiler

  • Armadillo library (required for C++ compilation)

  • BLAS/LAPACK

Installation Steps:

# 1. Install Armadillo (required!)
# macOS:
brew install armadillo

# Ubuntu/Debian:
sudo apt-get install libarmadillo-dev

# 2. Clone repository
git clone https://github.com/zhangxiany-tamu/fastcpd_Python.git
cd fastcpd_Python

# 3. Install with editable mode
pip install -e .

# 4. Optional extras for examples/benchmarks/time series
pip install -e .[dev,test,benchmark,timeseries]

# 5. Optional: Install Numba for 7-14x GLM speedup
pip install numba

Supported Platforms:

  • Linux (x86_64, aarch64)

  • macOS (Intel x86_64, Apple Silicon arm64)

  • Windows (experimental, source builds only)

Optional Dependencies

For Additional Performance

# Numba for 7-14x speedup on GLM models (binomial, poisson)
pip install numba

For Time Series Models

# ARMA models
pip install statsmodels

# GARCH models
pip install arch

# Or install both with:
pip install pyfastcpd[timeseries]

For Visualization

pip install matplotlib

For Development

# Install all development dependencies
pip install pyfastcpd[dev]

# This includes:
# - pytest, pytest-cov (testing)
# - black, ruff (formatting/linting)
# - mypy (type checking)

Verifying Installation

Test your installation:

import fastcpd
from fastcpd.segmentation import mean
import numpy as np

# Check version
print(f"fastcpd version: {fastcpd.__version__}")

# Run simple test
data = np.concatenate([
    np.random.normal(0, 1, 100),
    np.random.normal(5, 1, 100)
])
result = mean(data)
print(f"Detected change points: {result.cp_set}")

Expected output:

fastcpd version: 0.18.3
Detected change points: [100]

Troubleshooting

C++ Extension Import Error

If you see ImportError: C++ extension '_fastcpd_impl' is not available:

  1. Make sure Armadillo is installed:

    # macOS
brew list armadillo

# Ubuntu/Debian
dpkg -l | grep armadillo

  2. Reinstall the package:

    pip install --force-reinstall pyfastcpd

CMake Error: Could NOT find Armadillo

You need to install the Armadillo library before building:

# macOS
brew install armadillo

# Ubuntu/Debian
sudo apt-get update
sudo apt-get install libarmadillo-dev

Compilation Errors on Windows

Windows support is experimental. You’ll need:

For easier installation on Windows, wait for pre-built wheels to be available on PyPI.

Next Steps