Description

A very simple ndarray-based histogram class.

Author

Nicholas Devenish

Matplotlib histograms are geared around drawing, not data manipulation. Numpy direct support for histograms is extremely limited, and not very different from matpotlib. This is intended to turn into a set of very lightweight classes for shuffling data around. This is very much a work-in-progress.

The only required depenency is numpy, and the package is designed to work for python >= 2.6

A summary of usage, taken from the hists.py docstring follows:

Importing:

>>> from simplehist import Hist

Initialise with bin indices:

>>> a = Hist([0, 1, 2, 3]) >>> a.bincount 3 >>> a.bins (0, 1, 2, 3) >>> a.data array([ 0., 0., 0.])

Optionally include data:

>>> a = Hist([0, 1, 2, 3], data=[1, 0.2, 3]) >>> a.data array([ 1. , 0.2, 3. ])

Or just specify the blank data type:

>>> a = Hist([0, 1, 2, 3], dtype=int) >>> a.data array([0, 0, 0])

You can do arithmetic operations in place or seperately:

>>> a = Hist([0, 1, 2, 3], data=[1, 0.2, 3]) >>> b = a + a >>> b -= a >>> a.data == b.data array([ True, True, True], dtype=bool)

And you can fill bins from values:

>>> a = Hist([0,1,2,3]) >>> a.fill(1.4, weight=3) >>> a.data array([ 0., 3., 0.])

Even out of range:

>>> a = Hist([0,1]) >>> a.fill(-10) >>> a.underflow 1.0

If you use pyROOT, you can convert from 1D histograms:

>>> type(source) <class 'ROOT.TH1D'> >>> convert = fromTH1(source) >>> type(convert) <class 'simplehist.hists.Hist'>

And you can draw histograms, using any of the options that can be passed to matplotlib.pyplot.hist:

>>> hist_object.draw_hist(lw=2)