I made a little 'bihist' function to transform histograms to bi-histograms (e.g. http://www.itl.nist.gov/div898/handbook/eda/section3/bihistog.htm, see also Antonio Gonzalez's version, http://article.gmane.org/gmane.comp.python.matplotlib.general/7309). The chosen approach is just a small 'post-production' routine that still allows full control over plotted histograms.
Let me know if it's useful.
PS: would there be any "centralized" repository (besides the mailing list and the already over-crowded example list) where to gather all these kinds of matplotlib bits? Or a dedicated toolkit, where power functions could mature before possibly being integrated into plain matplotlib?
bihist.py (3.13 KB)
.~. Yannick COPIN (o:>* Doctus cum libro
/V\ Institut de physique nucléaire de Lyon
// \\ === *** AWAY FROM THE OFFICE *** ===
/( )\ AIM: YnCopin ICQ: 236931013