Welcome to gemmr’s documentation!

gemmr (short for GEnerative Modeling of Multivariate Relationships) calculates sample sizes required to run a canonical correlation analysis (CCA) or partial least squares (PLS). To that end, it also provides

• scikit-learn-style estimators for CCA and PLS,

• a scikit-learn-style Python wrapper for PMA, a sparse CCA implementation

• functionality to generate synthetic data for use with CCA and PLS, and

• code to determine their parameter dependencies.

See also

• source code repository on github