{{lastupdated_at}} by {{lastupdated_by}}

h1. Sixth ctools coding sprint

The sixth ctools coding sprint will take place at IRAP, Toulouse in the week 3-7 October 2016. We will start around noon on the first day and stop around noon on the last day so that you can travel on the same day to Toulouse (we may adapt the exact time of start on stop to your travel schedule).

Follow us on twitter (hash tag #ctools6).

!6th-coding-sprint.jpg!

h2. Audience and Participants

Everyone is welcome; if you have used ctools for CTA science simulations or IACT data analysis, if you are interested in contributing to the ctools and GammaLib software code development, or if you are already an experienced developer, you can contribute. You will need programming skills in Python or C++, best is of course if you are experienced in both. We will split into groups and work onto a number of issues. Bring your laptop. If possible, install git before coming and git clone ctools and GammaLib. For new folks: you can get a head start also by making an account on Redmine and taking a look at the code documentation (http://cta.irap.omp.eu/ctools/ and http://cta.irap.omp.eu/gammalib/).

If you plan to attend, simply add your name to the list below:

* Jürgen Knödlseder (IRAP)
* Francois Brun (CENBG)
* Michael Mayer (HU Berlin)
* Thierry Louge (IRAP)
* Johan Bregeon (LUPM)
* Thierry Reposeur (CENBG) 
* Alexander Ziegler (ECAP)
* Domenico Tiziani (ECAP)
* Marianne Lemoine-Goumard (CENBG)
* Marie-Hélène Grondin (CENBG)
* Matteo Balbo (ISDC)
* Lili Yang (UNG)

h2. Practical information

The meeting will take place at IRAP-Roche, information of how to reach the location can be found at http://cta.irap.omp.eu/toulouse2011/practical.html (the bus schedule may be a bit outdated! Take the bus 78 from Faculté de Pharmacie and exit at stop LAAS, then it's less then 5 min walk to IRAP).

A suggestion of possible hotels is on http://cta.irap.omp.eu/toulouse2011/accomodation.html (also this information dates back to 2011, but I guess most hotels are still there). Note that since recently we have a tramway connection to the airport. You may take the tram to go to the terminus (Palais de Justice) and from there take the metro to Faculté de Pharmacie (direction Ramonville) to go to the lab.

Another link with information about IRAP is here: http://www.irap.omp.eu/en/presentation/infos-pratiques


h2. Prerequisites for the sprint

* Have Git installed on your laptop and familiarize with it's use (see https://git-scm.com). You may use a GUI front end for Git. Personally I use SmartGit which I find extremely convenient to forget about the detailed syntax of Git commands (see http://www.syntevo.com/smartgit/)
* Learn about the ctools & GammaLib Git workflow and learn how to use GitLab (see http://cta.irap.omp.eu/ctools/develop/git/index.html)
* If you never have done a commit, try to create a fake commit in your GitLab user space as follows
** create forks of the ctools and GammaLib projects in your GitLab user space
** create a feature branch in the fork (either in ctools or GammaLib)
** clone the fork to your machine
** edit the code
** compile the code
** commit the code
* Read our GammaLib & ctools reference paper (http://www.aanda.org/articles/aa/abs/2016/09/aa28822-16/aa28822-16.html)

h2. Tentative agenda

Monday, 3 October:
* 14:00 - 15:00: Introduction to GammaLib and ctools (Jürgen) attachment:6th-coding-sprint.pdf
* 15:00 - 15:15: CTA Data Challenge: #1853 (Jürgen) attachment:data-challenge.pdf
* 15:15 - 15:30: User feedback from HESS collaboration (Michael) attachment:feedback_from_hess_MM_6th.pdf
* 15:30 - 15:45: Support composite model (or linking parameters): #1706 (Michael) attachment:composite_models_MM_6th.pdf
* 15:45 - 16:00: Error bands, ctbutterfly for non power law models (Alexander) attachment:ctools_IRAP.pdf
* _16:00 - 16:30: Coffee break_
* 16:30 - 16:45: VO Table display, event list (Thierry) attachment:GammaLibCodingSprintOctober2016_VOAspects.pdf
* 16:45 - 17:00: Status and evolution of On/Off analysis (Pierrick)
* 17:00 - 17:15: Tools for periodicity search (Matteo) attachment:Searching_for_periodicity.pdf

Tuesday, 4 October:
* Sprint features and issues (before starting today we should form groups; please add your name and create an issue if it does not yet exist):
** Pointing simulation tool (preparation of CTA Data Challenge): #1853 (Lili, Jürgen)
** Composite spectral and spatial models: #1706 (Domenico, Michael)
** Multiplicative spectral models: #1861 (Domenico, Michael)
** Covariance matrix, ctbutterfly: #1863, #1862 (Alexander, Matteo)
** VO table (Thierry L.)
** On/Off analysis (pull distributions, improvements) (François, Pierrick, Johan)
** Periodicity search, tempo2 plugin (Thierry R., Marianne, Marie-Hélène)
* 9:00-18:00: Coding, Testing, Documenting

Wednesday, 5 October:
* 9:00-18:00: Coding, Testing, Documenting
* 20:30: Coding sprint dinner:
Bois et Charbon (see http://www.boisetcharbon.net)
64 rue de la Colombette (see https://www.google.fr/maps/place/64+Rue+de+la+Colombette,+31000+Toulouse/@43.604797,1.4524952,17z/data=!3m1!4b1!4m5!3m4!1s0x12aebc97312e5719:0xb2677edc66ef7da6!8m2!3d43.604797!4d1.4546839)
Nearby (400-500 m) subway stations: Jean Jaurès

Thursday, 6 October:
* 9:00-18:00: Coding, Testing, Documenting

Friday, 7 October:
* 9:00 - 12:00: Sprint wrap up
** Progress on sprint features and issues:
*** Pointing simulation tool (preparation of CTA Data Challenge): #1853 (Lili, Jürgen)
*** Composite spectral and spatial models: #1706 (Domenico, Michael)
*** Multiplicative spectral models: #1861 (Domenico, Michael)
*** Covariance matrix, ctbutterfly: #1863, #1862 (Alexander, Matteo)
*** VO table (Thierry L.)
*** On/Off analysis (pull distributions, improvements) (François, Pierrick, Johan)
*** Periodicity search, tempo2 plugin (Thierry R., Marianne, Marie-Hélène)
** Roadmap for release 1.2
*** Priorities (following list to be discussed and amended):
**** CTA Data Challenge (think about what the Consortium will get and how they will be able to analyse the data; do we want a better support of classical analysis for the DC?)
***** Input from PHYS group would be interesting
***** Tools for variable sky (e.g. simulation of AGN flare, GRB)
**** Enabling a first H.E.S.S. (VERITAS, MAGIC) ctools science publication (what additional stuff you need in the software for that?)
***** Background modeling (but maybe outside ctools / GammaLib, maybe a script that computes the probability that a background model is okay)
***** Significance maps (e.g. implement Li & Ma formula)
***** ctools not much used for extragalactic science (tools to search for variability)
**** Fixing remaining bugs (don't forget to create issues for the stuff mentioned on Monday; they should have the highest priority, we don't want buggy software)
**** Increasing User friendliness (user interface, documentation)
**** ... (anything else?)
*** Issues
**** GammaLib: https://cta-redmine.irap.omp.eu/projects/gammalib/issues
**** ctools: https://cta-redmine.irap.omp.eu/projects/ctools/issues
*** Roadmap
**** GammaLib: https://cta-redmine.irap.omp.eu/projects/gammalib/roadmap
**** ctools: https://cta-redmine.irap.omp.eu/projects/ctools/roadmap

h2. Next sprint

Probably in Erlangen, during the "Vorlesungsfreie Zeit" (no lectures), somewhere second half of February, beginning March



h2. Collection of issues to be addressed during the sprint

* A central theme of this coding sprint could be the implementation of features needed for the CTA Data Challenge that is planned for beginning of 2017. The preliminary goals of the CTA Data Challenge are:
** "End-to-end" testing of analysis software
** Familiarise team with data content, formats, tools and realistic details of analysis issues (both instrumental and astrophysical)
** Develop additional methods for analysing CTA data, encouraging alternatives that are compatible within the planned analysis framework
** Provide feedback to the DATA group on what works and what is missing from the data formats and tools
** Re-evaluate sensitivities (possibly there are systematic effects we are not aware of yet)
** Grow the science analysis team - start to work together
** *Support readiness to do early science*

* The preliminary proposed scope of the CTA Data Challenge is
** start from gamma-ray response functions (+ residual background)
** Perform survey of Galactic plane, Galactic center and extragalactic survey
*** Realistic observation pattern for first-year survey(s) - including dead time
*** Sky model - probably derived from Fermi-LAT and TeVCat
*** Good response function for gamma rays and background at all zenith angles

* Other ideas for the sprint include:
** Implement models for variable source (lightcurves, pulsars, gamma-ray binaries)
** Implement source search script or tool (should find candidate sources in a data set and create a test XML model)
** Any issue from ctools release 1.2 roadmap: https://cta-redmine.irap.omp.eu/projects/ctools/roadmap
** Any issue from GammaLib release 1.2 roadmap: https://cta-redmine.irap.omp.eu/projects/gammalib/roadmap
** Support composite model (or linking parameters): #1706