This is pyhistuples, a set of python modules containing easy-to-use, pure-python classes representing histograms and NTuples. NTuples are a data structure well known in high-energy physics, where this project had its origins. It has been released within the DaVinci software project of the LHCb experiment at CERN. As far as I know, it hasn't been extensively used. See release.notes for more information.

The rationale here has been to provide intuitive and extensible classes for simple data processing and manipulation. Performance has not been a consideration so far, but it is on the to-do list. There is a simple persistency mechanism using python pickle. Among things to inversigate is the use of the shelve module for ntuples that are too large to keep in memory.

Lesser GPL, see COPYING and COPYING.LESSER.

• nose for all unit tests.
• matplotlib for histogram plotting.

This assumes matplotlib has already been installed into the system.

Clone repoo and pip install from source:

git clone https://github.com/juanchopanza/pyhistuples.git
pip install pyhistuples 

Pip install from repo

pip install git+https://github.com/juanchopanza/pyhistuples.git

• Completely experimental.

• The unit tests give a good indication of how to use the different components. Documentation will follow some time in the future.

Juan Palacios 14/05/2011