Skip to content

Installation

Standard installation

pip install pycatdap

This installs the core CATDAP-01 / CATDAP-02 functionality with numpy and pandas as the only required dependencies.

With visualization support

pip install "pycatdap[plot]"

Adds matplotlib and statsmodels for the existing v0.2 plotting API.

pip install "pycatdap[plotly]"

Adds plotly and jinja2 for interactive figures and HTML reports. Required for the modern EDA workflow (profile(), aic_heatmap(), error_analysis()).

Future extras

The following extras will become available as the roadmap progresses:

  • pycatdap[widget] — anywidget for interactive Jupyter widgets
  • pycatdap[data] — requests and scikit-learn for fetch_* data loaders
  • pycatdap[all] — everything

Supported Python versions

  • Python 3.10
  • Python 3.11
  • Python 3.12
  • Python 3.13

Verifying the installation

import pycatdap
print(pycatdap.__version__)

Development install (from source)

git clone https://github.com/nbx-liz/pycatdap
cd pycatdap
uv sync --all-groups

This installs all development dependencies, including documentation tooling (docs group).

Optional: R catdap for cross-validation

If you want to run the R cross-validation suite (pytest -m slow), you'll need R with the catdap package:

install.packages("catdap")

These tests are excluded from default CI but required before release tags.