Sherpa Installation

In order to follow along with the Sherpa examples presented in this workshop, you can use Sherpa in CIAO, or install the stand-alone version described below.


Unfortunately we have not updated the binary versions to match the Sherpa 4.4 release - they are still for Sherpa 4.3 - but you can try the source build of Sherpa 4.4.

  • Sherpa Dependencies (Sherpa 4.3)
    • Python 2.6 or 2.7 (not Python 3.x)
    • NumPy >= 1.3
    • Matplotlib >= 0.99
    • pyFITS >= 1.3
    • DS9 >= 5


EPD users, you have already satisfied the installation requirements above. Continue with the installation notes below. If you have been to previous workshops and are able to complete the examples, your installation is most likely sufficient. See the Sherpa installation notes below to install Sherpa version 4.3.0.

Sherpa downloads

The Sherpa source tar file and pre-built binaries can be found here:


Installed the Enthought EPD 7.0 following the instructions for MacPorts. Then download the appropriate Sherpa disk image for your version of OSX:

OSX EPD disk image Sherpa disk image
Leopard 10.5 epd-7.0-2-macosx-i386.dmg sherpa-4.3.0-EPD-7.0-i386.dmg
Snow Leopard 10.6 epd-7.0-2-macosx-x86_64.dmg sherpa-4.3.0-EPD-7.0-x86_64.dmg

Double-click on the disk image and follow the instructions in the install wizard.

Ubuntu (root)

Users who use apt-get to manage their Python distribution can download the appropriate Ubuntu package for their operating system.

Double-click on the Debian package file (.deb) and follow the instructions in the install wizard.

EPD Linux

Install the EPD on Linux by following the instructions for Non-root setup using virtualenv. Then download the appropriate Sherpa egg for your Linux architecture

Architecture EPD Linux installer Sherpa egg
Linux i686 (32-bit) sherpa-4.3.0-EPD-7.0-linux-x86.egg
Linux x86_64 (64-bit) sherpa-4.3.0-EPD-7.0-linux-x86_64.egg

Install Sherpa into your EPD installation using easy_install:

easy_install sherpa-4.3.0-EPD-7.0-linux-x86.egg

Try It Out

Try importing the Sherpa high level UI with:

from sherpa.astro.ui import *


If you see the following error messages

WARNING: failed to import sherpa.plot.chips_backend; plotting routines will not be available

WARNING: failed to import; FITS I/O routines will not be available

Be sure to edit your ~/.sherpa.rc file and indicate

plot_pkg : pylab

io_pkg : pyfits

If you continue to see these messages, you should install pyFITS and matplotlib.


If you see the following error message, your installation of XSPEC may be incomplete.

WARNING: failed to import sherpa.astro.xspec; XSPEC models will not be available


The following error message indicates that Sherpa is unable to find your DS9 or XPA.

WARNING: imaging routines will not be available, failed to import sherpa.image.ds9_backend due to ‘RuntimeErr: DS9Win unusable: Could not find ds9 on your PATH’