Skip to content

Streaming plotting programs: scatterplot, histogram, sparkline, and timeline

Notifications You must be signed in to change notification settings

skuehn/stream-plots

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

OVERVIEW

This package provides simple and fast utilities for creating
figures. Programs operate on data streams, succeed silently, and try
not to make many assumptions about formats or parameters.  The goal is
ease of use in scripting, so a user can quickly chain together a
series of visualizations.

CONTENTS

After installation, the plotting programs will be located under bin/.
Information about running each program can be found by running the
program with the '--help' flag.  A brief overview of the four
programs:

o hist:     Create a histogram.  User must specify the high and low
            bins.  This program is a wrapper for gsl-histogram.
o scatter:  Compares two variables via scatterplot.  The program takes
            a file stream input consisting of two tab-delimited
            columns of numeric values.
o series:   Plot a series of data.  Multiple columns of input will plot
            multiple series.
o spark:    A simplified version of 'series'.  Check out
  	    http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0001OR&topic_id=1
	    for an in-depth explanation of the theory and purpose of sparklines.

EXAMPLES

Take a look at the scripts in share/ for some samples.
singe-instances.sh shows a few standard examples. A more involved
plotting pipeline is demonstrated in multi-instances.sh

INSTALLATION INSTRUCTIONS

Install a copy of the program by running 'python setup.py install',
from the root directory of this distribution.  A useful option for
installing the package in user space is '--prefix'.  For example, to
install into the local direction $HOME/foo, run 
'python setup.py install --prefix=$HOME/foo'.  

DEPENDENCIES

The plotting scripts make use of several other packages:
o The GNU Scientific Library (GSL), http://www.gnu.org/software/gsl.
  Version >= 1.9.  GSL binaries must be in the users PATH env variable
o Gnuplot.py version >= 1.7. Debian/ubuntu users: this package is
  available for most distros, as the 'python-gnuplot' package.
  Install manually from http://gnuplot-py.sourceforge.net. The
  Gnuplot.py package must be in the users PYTHONPATH env variable.
o Python, http://docs.python.org. Version >= 1.4

Installation will fail if either of these dependencies is not met.

TODO

o Fix program input to better handle negative valued paramaters
o Add a utility for matrix 'heatmap' plotting, or wrap matrix2png
o Add utilities for lift-charts and roc-curves
o Control scatterplot fitting output

About

Streaming plotting programs: scatterplot, histogram, sparkline, and timeline

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published