text_histogram 0.0.5

A dependency-free library to quickly make ascii histograms from data.

text_histogram

Histograms are great. This is Bit.ly’s data_hacks histogram.py repackaged for convenient script use.

>>> from text_histogram import histogram
>>> import random
>>> histogram([random.gauss(50, 20) for _ in xrange(100)])
# NumSamples = 100; Min = 1.42; Max = 87.36
# Mean = 51.848095; Variance = 332.055832; SD = 18.222399; Median 53.239251
# each ∎ represents a count of 1
    1.4221 -    10.0159 [     3]: ∎∎∎
   10.0159 -    18.6098 [     3]: ∎∎∎
   18.6098 -    27.2036 [     6]: ∎∎∎∎∎∎
   27.2036 -    35.7974 [     4]: ∎∎∎∎
   35.7974 -    44.3913 [    17]: ∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎
   44.3913 -    52.9851 [    16]: ∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎
   52.9851 -    61.5789 [    17]: ∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎
   61.5789 -    70.1728 [    20]: ∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎∎
   70.1728 -    78.7666 [     8]: ∎∎∎∎∎∎∎∎
   78.7666 -    87.3604 [     6]: ∎∎∎∎∎∎

Installation

$ pip install data_hacks

Why?

Histograms are great for exploring data, but numpy and matplotlib are heavy and overkill for quick analysis. They also can’t be easily used on remote servers or over ssh. Don’t even get me started on installing them.

data_hacks is pretty great, but difficult to use from python code because it directly requires a optparse.OptionParser to pass histogram options.