On 08/30/2012 04:55 PM, Peter Prettenhofer wrote: > 2012/8/30 Olivier Grisel <[email protected]>: >> 2012/8/30 Alexandre Gramfort <[email protected]>: >>> hi sklearners, >>> >>> I am coming back from the Biomag conference [1] >>> where emanuele (@emanuele on github) won the data analysis competition [2] >>> using the scikit-learn ! >>> >>> congrats @emanuele ! >> Congrats indeed! Which of the 2 competitions did you / he won? > Congratulations Emanuele! > > Looking forward to more information about the competition/solution. > >
Dear All, Thanks for your kind words. I am not the only one in the team, which is composed by three people: - me (Emanuele Olivetti) - Yuan Tao (PhD student at University of Trento) - Nathan Weisz (Principal Investigator of our MEG lab) We participated to competition 1, i.e. discriminating living vs. non-living categories for auditory and visual stimuli from MEG recordings. The dataset comprised data from 2 subjects. I haven't seen the final result of all the teams (I am not in Paris, Nathan is) so I cannot comment on the details now. My impression is that our results are not great in absolute value, in the sense that we had above chance results on one subject only. Moreover the test sets were quite small for a 4 class problem so the variance of the result might be large enough to say that luck might have helped us a bit. Anyway it seems this was enough to beat other teams :) We used mainly scikit-learn (Random Forests) and for this reason we wanted to mention explicitly you all for your great work. I attempted many things during the competition so the code is a total mess at the moment. I am very swamped in this period but sooner or later I'll publish a clean-up version of the last code. Thanks again to you all, Emanuele ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
