This package implements statistical methods useful for Datazilla.

Currently the only methods implemented are Welch's t-test (dzmetrics.ttest.welchs_ttest) and Benjamini-Hochberg false discovery rate control (dzmetrics.fdr.rejector). See the docstrings of those functions for details.

Datazilla-metrics requires scipy and thus also numpy. If installed via pip, these must be installed sequentially (first numpy then scipy), and scipy requires the BLAS and LAPACK libraries and a Fortran compiler. For instance, to get everything in place on an Ubuntu 12.04 system, run:

sudo apt-get install libblas-dev liblapack-dev gfortran
virtualenv dzmetrics
pip install numpy
pip install scipy
pip install git://github.com/mozilla/datazilla-metrics

It may be easier to just use the system packages for numpy and scipy; in that case to import it from within a virtualenv you'd need to create the virtualenv with the --system-site-packages flag:

sudo apt-get install python-scipy
virtualenv --system-site-packages dzmetrics
pip install git://github.com/mozilla/datazilla-metrics

python setup.py test