douglasleetucker / pyexcal

A simple Python standard star fitter inspired by the SDSS mtpipe excal code.

pyExcal

A Python standard star fitter inspired by the SDSS mtpipe excal code.

FAQ

• How to run pyExcal?
• How to request updates to the code or to the FAQ?

---

How to run pyExcal?

1. Ensure that you have the necessary dependencies installed on your computer:
   • Python 2.7 or higher
   • The numpy python module (pyExcal definitely works with numpy 1.11.1; other versions should work, too.)
   • The scipy python module (pyExcal definitely works with numpy 0.18.0; other versions should work, too.)
   A fairly recent version of Ureka (run "ur_setup") or astroconda should set up all the necessary versions of Python, numpy, and scipy.
2. Run the appropriate setup script in the pyExcal bin directory, depending on which shell you are using (bash or tcsh):

    source pyExcal/bin/setup.bash

   or

    source pyExcal/bin/setup.csh

   **Note: you will need to modify the value for PYEXCAL_DIR in the setup script before the first time you ever run it.**
3. Run the following command:

    pyExcal.py --inputFile $PYEXCAL_DIR/test/std-rguiz-test.g.csv --band g

You should get the following output to the screen, as well as two QA plots in png form:

bash$ pyExcal.py --inputFile $PYEXCAL_DIR/test/std-rguiz-test.g.csv --band g

Initial parameter values:   [1.6509999999999998, 0.0, 0.0]
Converged
Converged with chi squared  0.481103286398
degrees of freedom, dof  40
RMS of residuals (i.e. sqrt(chisq/dof))  0.109670334001
Reduced chisq (i.e. variance of residuals)  0.01202758216

Fitted parameters at minimum, with 68% C.I.:
a                1.34904 +/-     0.0629574   (4.666840 percent)
b            -0.00829816 +/-     0.0502757   (605.865115 percent)
k               0.214706 +/-     0.0391504   (18.234481 percent)

Correlation matrix:
                 a          b          k
a            1.000000
b           -0.101816   1.000000
k           -0.904639  -0.244643   1.000000


Your fit equation is:
    g_inst - g_std = 1.349 + -0.008*((g-r) - 0.530) + 0.215*X

Outputting QA plot qa-std-rguiz-test.g.csv_airmass.g-band.png
Outputting QA plot qa-std-rguiz-test.g.csv_color.g-band.png


That's all, folks!

The a, b, and k parameters are the familiar a, b, and k from the mtpipe photometric equations. Currently, I have ignored the "c" term coefficient (a.k.a., the second-order extinction), but that could be added if needed. (The concern is that it is so small that we won't get a significant fit for it, and will just end up adding noise to the fit.)

Back to top.

---

How to request updates to the code or to the FAQ?

Please use the issues to post requests for additions or other updates to the code or to this FAQ .

Back to top.