Google Summer of Code 2013

This wiki page will be the central hub of information about the R Project participation in the Google Summer of Code (GSoC) for 2013. Administrators are Toby Dylan Hocking, Brian Peterson, and Virgilio Gomez Rubio as a backup. Questions should be posed to the google group!

Everyone who wants to participate in this year’s Google Summer of Code with R please join our google group: gsoc-r@googlegroups.com make sure to explain your relationship to R and your interest in the program.... requests without any other identifying or qualifying information may be considered spam.

Overview

In short, each student will get paid $5000 to work on an R package for 3 months during the summer:

  • Mentors can add projects to this wiki page to give ideas to students.
  • Students should look at the list of projects and see if any project interests them. If so, email the project mentors to express your interest, and describe any prior experience. Students with ideas for R projects not listed below are encouraged to post a query on the gsoc-r google group.
  • Students will submit project applications directly to Google. The home page is http://www.google-melange.com/gsoc/homepage/google/gsoc2013
  • Google will award a certain number of student slots to the R project.
  • The GSOC-R administrators and mentors will rank projects in order of importance to the R project, and the top projects will be funded.
  • Students get paid $5000.
  • Mentors get code written for their project, but no money.

Previous projects are listed on the soc08, soc09, soc10, soc11, and soc12 pages.

Status and Timeline

Note the org admins have a bit more work to do this year than last year from May 6 to May 27.

  • Mar 18: Mentoring organizations can begin submitting applications to Google.
  • Mar 29: Mentoring organizations application deadline.
  • Apr 8: List of accepted mentoring organizations published on the Google Summer of Code site.
  • Apr 22: Student application period opens.
  • May 3: Student application deadline.
  • May 6: Mentoring organizations should have requested slots via their profile in Melange by this point.
  • May 22: First round of de-duplication checks happens; organizations work together to try to resolve as many duplicates as possible.
  • May 24 07:00 UTC: All mentors must be signed up and all student proposals matched with a mentor. Student acceptance choice deadline.
  • May 24 19:00 UTC: IRC meeting to resolve any outstanding duplicate accepted students #gsoc.
  • May 27: Accepted student proposals announced on the Google Summer of Code site.
  • Jun 17: Coding begins.
  • Aug 2: Midterm evaluation deadline.
  • Sep 27: Final evaluation deadline.
  • Oct 19 & 20: Mentor summit.

Project Ideas

Project proposals are welcome, and should be added to the list below. However, each proposal has a big barrier to pass: it will to need explain why this particular project is of use to the community, and it will have to show how it can be achieved in three months. At a minimum, a submission should also include a review of related packages that address the same or similar problems and discuss how these packages are not sufficient to the task.

Mentors Proposal Non-R languages? Status Students Results
mentors Use this to add new proposals by copying the page content languages status students results
Michael Rosemblum, Han Liu Exponential Random Graph Model with block structure accepted, passed Xiaolin Yang blkergm on CRAN implements block Exponential Random Graph Models with different dependence assumptions including block Erdos-Renyi, Beta, Markov random graph models. In the future, the package will be extended for more models in the ERGM family with block structures.
Han Liu MutTLeR: A Generalized Multi-Task Learning Toolkit accepted, passed Tour? camel package on CRAN implements several solvers, future work??
Y. Richet, backup John Nash Handle parallel (vectorized) objective functions in a new optimization wrapper package C accepted Qiang Kou vectoptim package on github and the first version on CRAN, remaining work for next year: more optimization methods
Toby Dylan Hocking Improve rendering of interactive, animated ggplots Javascript accepted, passed Susan Vander Plas animint package on github implements most geoms, legends, axes, and stats. TODOs for next year are listed in the NEWS file: facets, coord_equal, etc.
Paul Murrell Improve display of imported graphics Postscript, SVG, XML, C accepted, passed Simon Potter grConvert, grImport, and grImport2 packages on R-Forge, summary
Kris Boudt, Jonathan Cornelissen highfrequency add inferential methods to highfrequency C accepted, passed Giang Nguyen http://r-forge.r-project.org/R/?group_id=1409,http://github.com/jonathancornelissen/highfrequency, a large family of jump tests is included, as well as ex post cov estimator using preaveraging and a C-implementation of the heavy model with SE
Joshua Ulrich, Michael Weylandt xtsExtra multi-type xts objects accepted Anton Samoylov xtsExtra on GitHub (factor-support branch) xts/zoo objects are a matrix with an ordered index attribute, so they cannot contain mixed types like a data.frame. Anton created a great working prototype of an xts-like object that can contain multi-type columns. Work next year could include: improving performance, and fixing bugs and refining edge cases by exploring use-cases.
Claudia Beleites, Simon Fuller, Bryan Hanson spectral unmixing accepted, passed Conor McManus unmixR package on GitHub Conor implemented the “original” papers for N-FINDR and Vertex Component Analysis, plus some recent papers with improvements. In addition, we developed our own improved version of N-FINDR!
Peter Carl Addressing IID Assumptions in Finance: Autocorrelation and Drawdowns in Performance Analysis accepted two projects pulkit?, Shubhankit Mohan
David Ardia, Brian Peterson Collection of functionality ported from the MATLAB code of Attilio Meucci Matlab accepted, passed Xavier Valls Meucci package on R-Cran. Remaining and future work for next year described in the “to do” file
Doug Martin, Guy Yollin, Brian Peterson Improve constraints, usability, graphics for PortfolioAnalytics accepted Ross Bennett
Gergely Daróczi Reproducible statistical report templates in markdown accepted Dániel Nagy rapport package on GitHub
Andreas Alfons robgpu Highly parallelized implementations of robust statistical procedures using CUDA C++, CUDA accepted Roland Boubela
Hans W Borchers Non-smooth Optimization — Implement different approaches for optimization of non-smooth functions, as often encountered in Engineering and Operations Research R, Matlab, C or Fortran accepted, failed Abhirup Mallik SIDPSM, DFLS?
pbdR Core Team Profiling Tools for Parallel Computing with R — Develop an MPI profiling package C accepted, passed Gaurav Sehrawat pbdR homepage, github page, future work listed in TODO file
Jae W. Lee, Jung Hoon Lee RIGHT: R-Interactive-Graphics-via-HTml HTML5, Javascript accepted, passed Chungha Sung RIGHT package on Google code and CRAN package, remaining work for next year: support for additional visual elements, e.g. point and line types, and ggplot2-like R API
Eric Zivot Factor Models for Return Series accepted,passed Yi-An Chen factorAnalytics on R-forge , remaining work for next year: Complete Barra type performance attribution
Javier Otegui Biodiversity data visualization in R accepted, passed Vijay Barve bdvis source code on github. The package provides ways to efficiently visualize biodiversity data. Specifically, it is useful for plotting spatial and temporal distribution of biodiversity data, taxonomic hierarchies and to perform assessments of completeness on the data. Next year: improve current visualizations and add more assessments.
Claudia Beleites, Simon Fuller HyperSpec improve HyperSpec not accepted
Roger Bivand, Daniel McMillen Akima's algorithm for interpolation test C/C++ not accepted
(added by) Dirk Eddelbuettel (not mentoring) Port pymssql core to R and DBI as RMSSql C not accepted

 
developers/projects/gsoc2013.txt · Last modified: 2013/11/09 by cbeleites
 
Recent changes RSS feed R Wiki powered by Driven by DokuWiki and optimized for Firefox Creative Commons License