Using Python for Astronomical Data Analysis

227th AAS Meeting
Monday, 4 January 2016
9:00am - 12:00pm, 1:30pm - 4:30pm
Room: St. George 114
Fee: $35
Instructors: Larry Bradley, Nadia Dencheva, Tim Pickering, and Erik Tollerud
Speaker: Jason Kalirai

This workshop will cover the use of Python tools to analyze astronomical data, with the focus primarily on Optical, IR and UV data analysis tools. The primary tools that will be covered are those available in the Astropy library and affiliated packages. The specific tools to be covered will be:

• Physical units and quantities
• Basics on accessing data files, both FITS and ascii tables
• Coordinate utilities
• Modeling and Fitting
• Photometric tools
• Interactive visualization and analysis tools:
  • Glue
  • imexam
  • specview

There will be time spent on hands-on exercises. Instructions on installing the necessary software will be provided before the workshop and help will be available at the workshop for those that experience problems with installations.

The prerequisites are a familiarity with astronomical data analysis. Basic Python experience is highly recommended to be able to participate in the exercises. Those without Python experience will still get much useful information about the capabilities for data analysis in Python. Experience with Python scientific libraries, particularly numpy and matplotlib, is helpful, but not required.

Installation Requirements

To follow along with the tutorials and to do the exercises, you will need to have the following packages installed:

• Python 2.6 (>= 2.6.5), 2.7, or >= 3.3
• IPython (>= 4.0) and Jupyter Notebook (>= 1.0)
• Numpy >= 1.6
• Scipy >= 0.15
• Pandas >= 0.17
• xlwt >= 1.0
• matplotlib >= 1.3
• Astropy 1.1
• Photutils 0.2
• scikit-image >= 0.11
• astroquery >= 0.2.6 (not strictly needed, but makes the coordinates tutorial more straightforward)
• Glue
• imexam
• astroquery

You can run the check_env.py script to check your Python environment for the required dependencies:

% python check_env.py

Some quirks you might run into:

• glue is installed as glueviz (e.g. conda install glueviz)
• skimage is scikit-image (e.g. conda install scikit-image)
• If you're using anaconda or miniconda, you should use the astropy channel for conda to install photutils and astroquery:

conda install -c http://conda.anaconda.org/astropy photutils astroquery

Schedule

Time	Topic
8:00 - 9:00 am	Installation help
8:30 - 9:00 am	Breakfast
9:00 - 9:15 am	Astropy Overview (Erik T; 15 min)
9:15 - 10:00 am	Numpy Intro, Units (Larry; 45 min)
10:00 - 10:30 am	Tables, FITS, ASCII tables Part 1 (Tim; 30 min)
10:30 - 10:45 am	Break (15 min)
10:45 - 11:15 am	Tables, FITS, ASCII tables Part 2 (Tim; 30 min)
11:15 am - 12 pm	Coordinates (Erik T; 45 min)
12:00 - 1:30 pm	Lunch
1:30 - 1:45 pm	Cosmology (Erik T; 15 min)
1:45 - 2:40 pm	Modeling, Fitting, WCS (Nadia; 55 min)
2:40 - 3:00 pm	JWST: Preparing for Cycle 1 (Jason; 20 min)
3:00 - 3:15 pm	Break (15 min)
3:15 - 4:00 pm	Photutils (Larry; 45 min)
4:00 - 4:30 pm	GUI (Glue, imexam, specview) (Tim; 30 min)

AAS Survey

If you participated in the workshop, the AAS asks that you please fill out the survey at http://bit.ly/aas227profdev