Skip to content

Google summer of code (GSOC) 2015

Jump to bottom
Andreas Mueller edited this page Mar 5, 2015 · 35 revisions

This is the page for coordination of the GSoC for scikit-learn.

Scikit-learn is a machine learning module in Python. See http://scikit-learn.org for more details.

Scikit-learn is taking part of the GSoC trough the Python Software Foundation: http://wiki.python.org/moin/SummerOfCode

Instructions to student: achieving a good proposal

Difficulty: Scikit-learn is a technical project. Contributing via a GSoC requires a number of expertise in Python coding as well as numerical and machine learning algorithms.

Important: Read: Expectations for prospective students

Application template: https://wiki.python.org/moin/SummerOfCode/ApplicationTemplate2015 Please follow this template.

Also important: A letter from Gaël to former applicants. His suggestions are just as relevant this year.

Hi folks,

The deadline for applications is nearing. I'd like to stress that the scikit-learn will only be accepting high-quality application: it is a challenging, though rewarding, project to work with. To maximize the quality of your application, here are a few advice:

  1. First discuss on the mailing list a pre-proposal. Make sure that both the scikit-learn team and yourself are entousiastic about the idea. Try to have one or two possible mentors that hold a dialog with you.

  2. Satisfy the PSF requirements (http://wiki.python.org/moin/SummerOfCode/Expectations) briefly:

    • Demonstrate to your prospective mentor(s) that you are able to complete the project you've proposed
    • Blog for your GSoC project.
    • Contribute at least one patch to the project

I'd add the the patch should be somewhat substantial, not just fixing typos.

To contribute patch, please have a look at the [contribution guide] (http://scikit-learn.org/dev/developers/index.html#contributing-code) and the Easy issues in the tracker.

  1. In parallel with 2, start a online document (google doc, for instance) to elaborate your final proposal, and if you manage to convince mentors, you can get feedback on it.

As a final note, I want to stress that GSOC projects are ambitious: we are talking about a few months of full time work. Thus the ideas proposed are idea challenging, and the students are supposed to draw a battle plan, with difficult variants and less difficult variants. The GSOC is a full major set of contributions, not a single pull request.

Good luck, I am looking forward to seeing the proposals. You'll see, the scikit is a big friendly and enthousiastic community,

Gaël

A list of topics for a Google summer of code (GSOC) 2015

Disclaimer: This list of topics is currently being updated from last year's, and some information (like the names of possible mentors) is not definitive. Please e-mail the list with any questions.

Improve GMM

Possible mentors: Andreas Mueller, Gael Varoquaux, Vlad Niculae

Possible candidate:

  • Reimplement VBGMM and DPGMM based on text-book derivations
  • Refurbish current GMM implementation and testing, clean up the API.
  • Implement a core-set strategy for GMM (optional)

http://las.ethz.ch/files/feldman11scalable-long.pdf http://videolectures.net/nips2011_faulkner_coresets/

Issue to get started : https://github.com/scikit-learn/scikit-learn/issues/4202

Generalized Additive Models ( GAMs )

Possible mentor: Paolo Losi, Alex Gramfort, (others?)

Possible candidate: ragv

Goal: Add the additive ( or additive_model ) directory and implement a few additive models.

  • Help finishing up the PR by jcrudy on including pyearth into scikit
  • Add Generalized Additive Model ( GAM )
  • Add SpAM ( Sparse Additive Model )
  • Add GAMLSS ( GAM for Location Scale and Shape )
  • Add LISO ( LASSO ISOtone for High Dimensional Additive Isotonic Regression)

References:

Metric learning

Possible mentor:

Possible candidate: Barmaley-exe

Goal: add some of metric learning algorithms (like NCA, ITML, LMNN) to be used with KNNs and as transformers. Brian Kulis has a survey and a tutorial on metric learning, that seem to be a good place to start.

Multiple metric support for cross-validation and gridsearches

Possible mentor: Andreas Mueller

**Possible candidate: **

Goal: Allow the simultaneous evaluation of multiple metric functions (like accuracy, AUC, recall) with a single run on cross-validation or Grid Search.

This project is quite heavy on API design, and will touch many places in scikit-learn. See #1850 and #2759.

Cross-validation and Meta-Estimators for semi-supervised learning

** Possible mentor: **

** Possible candidate: **

Goal: Improve support for semi-supervised learning, implement a meta-estimator for self-taught learning.

The idea is to make semi-supervised learning easier in scikit-learn, by improving the support in cross-validation. Cross-validation objects need to be aware of which samples are labeled, and which are not, to produce meaningful test errors. There are several semi-supervised algorithms that would be good to implement, the easiest being a self-taught learning meta-estimator #1243.

Possible mentors

Here are people that have said that they might be available for mentoring:

Gaël Varoquaux, Vlad Niculae, Olivier Grisel, Andreas Mueller, Alexandre Gramfort, Michael Eickenberg.

Clone this wiki locally