Kaggle team, Hope everyone's okay with communicating general stuff here, as per request.
I was thinking that we could start simply, by just feeding in the stream of solar energy readings to the CLA sequentially. This will allow us to see how well it does at predicting solar energy from just the temporal sequence, and the time of day, day of week, etc. In parallel, someone else can learn to extract what we need from the GEFS weather data, correlating the weather variables with each solar energy reading time. We can build a CSV file of all the variables together, and run a swarm. How does this approach sound? I can volunteer to set up the project and start feeding the solar energy with temporal information to the CLA. Whoever wants to help is welcome to. Anyone up for massaging the GEFS data as described above? Thanks, Chetan P.S. We still need a team name! On Mon, Oct 21, 2013 at 9:31 PM, Tim McNamara <[email protected]>wrote: > +1 > > Very interested in observing this project > On 22/10/2013 12:08 AM, "Matthew Taylor" <[email protected]> wrote: > >> Pedro et al, >> >> Unless you're actually going to be dealing with proprietary >> information, I encourage you to have your Kaggle discussions here on >> the mailing list. I'm sure there are others out there interested in >> what you're doing, but unable to take part (like me). >> >> Feel free to start a new thread, or use this one, but please use this >> list if you can. No one has complained about too many emails... yet >> ;). >> >> --------- >> Matt Taylor >> OS Community Flag-Bearer >> Numenta >> >> >> On Fri, Oct 18, 2013 at 11:39 AM, Pedro Tabacof <[email protected]> >> wrote: >> > Hello, >> > >> > Thank you all for the interest. I think now it's time to create a >> private >> > thread for the competition. The Solar Energy competition was the most >> voted >> > one so it can be our first Kaggle attempt. >> > >> > If anyone else is also interested in joining the competition, just send >> me a >> > message. >> > >> > Pedro. >> > >> > >> > On Fri, Oct 18, 2013 at 3:13 PM, Doug King <[email protected]> wrote: >> >> >> >> I'm interested in joining the team. I looked at the Kaggle contests >> and I >> >> agree that the best choice would be Solar Energy Prediction, but I >> wouldn't >> >> discount the Phone accelerometer contest. I think its a good fit for >> the CLA >> >> and and interesting application. The forum posts for that contest seem >> to >> >> indicate the sponsors are willing to judge a 'clean' (no leakage) >> version of >> >> the contest. >> >> >> >> >> >> On Thu, Oct 17, 2013 at 10:45 AM, Jay Gokhale <[email protected]> >> wrote: >> >>> >> >>> I'd love to use CLA's on a kaggle project! >> >>> >> >>> If Chetan's description is correct, it seems to me that the Solar >> Energy >> >>> Prediction contest would be the most straight forward one to work on. >> >>> >> >>> Jay Gokhale >> >>> >> >>> >> >>> On Wed, Oct 16, 2013 at 9:07 PM, Matthew Taylor <[email protected]> >> wrote: >> >>>> >> >>>> These also sound like good hackathon projects! >> >>>> >> >>>> Matt >> >>>> >> >>>> Sent from my MegaPhone >> >>>> >> >>>> On Oct 16, 2013, at 8:03 PM, Chetan Surpur <[email protected]> >> wrote: >> >>>> >> >>>> I'm interested in participating! >> >>>> >> >>>> I haven't decided on one yet, but I wanted to point out that the CLA >> >>>> looks ideally suited for the AMS 2013-2014 Solar Energy Prediction >> contest. >> >>>> From what I can tell, it's about predicting the daily energy output >> of solar >> >>>> farms from 12, 15, 18, 21, 24-hourly training data. There's 29 days >> to go. >> >>>> >> >>>> >> >>>> On Wed, Oct 16, 2013 at 7:31 PM, Pedro Tabacof <[email protected]> >> >>>> wrote: >> >>>>> >> >>>>> Hello, >> >>>>> >> >>>>> After the warm reception of the Kaggle competition idea, I decided >> to >> >>>>> create the thread asap. For those who are not aware, Kaggle is a >> data mining >> >>>>> competition website. It has been hyped on the machine learning >> media and has >> >>>>> some great people participating in it. It rewards with different >> prizes, but >> >>>>> my focus is not earn money, but rather to see how good NuPIC would >> fare >> >>>>> against the state of the art. I found those following competitions >> to be a >> >>>>> viable start: >> >>>>> >> >>>>> 1) Accelerometer Biometric Competition >> >>>>> From real acceleration data you have to guess from which cell phone >> >>>>> came the test samples. >> >>>>> 36 days to go >> >>>>> Pros: >> >>>>> -Multivariable, temporal data seems ideal for the CLA >> >>>>> Cons: >> >>>>> -The best competitors are using data leaks (meta-information), so >> the >> >>>>> chance of winning fairly is zero (though they will give a >> consolation prize >> >>>>> to the best "honest" entry) >> >>>>> Questions: >> >>>>> -How fast would NuPIC go through a 1GB CSV file? >> >>>>> -The test data is not open-ended, that is, you have to say whether a >> >>>>> recording belongs to a specific cell phone or not, with 50% chance >> of being >> >>>>> right. Would NuPIC do well in this case? >> >>>>> >> >>>>> 2) Multi-Label Bird Species Classification >> >>>>> Try to guess the bird species present on some audio recordings >> >>>>> 38 days to go >> >>>>> Pros: >> >>>>> -Temporal data >> >>>>> Cons: >> >>>>> -Multiple labels, so I think we would need 87 different models, >> which >> >>>>> is kinda impractical >> >>>>> Question: >> >>>>> -Is there a better way to do multi-label classification with NuPIC? >> >>>>> >> >>>>> 3) Conway's Reverse Game of Life >> >>>>> Reverse the game of life for 1 to 5 steps >> >>>>> 4 months to go >> >>>>> Pros: >> >>>>> -Natural input sparseness (have to check on it) >> >>>>> -Temporal data >> >>>>> -Multistep classification >> >>>>> -We can generate our own training data if necessary >> >>>>> Cons: >> >>>>> -Maybe there are very straightforward ways to solve this (rule >> based, >> >>>>> brute force, etc) >> >>>>> -Too much time before the competition ends >> >>>>> Questions: >> >>>>> -How sparse does the input need to be for the CLA to work well? >> >>>>> -Is it possible to use a 2D arrangement on the cells connection >> matrix? >> >>>>> >> >>>>> There is also the AMS 2013-2014 Solar Energy Prediction contest, >> but I >> >>>>> didn't have time to go over it. The other competitions don't seem >> to be >> >>>>> suited for NuPIC. >> >>>>> >> >>>>> Who here is interested on participating and what competition do you >> >>>>> think to be the most promising one? >> >>>>> >> >>>>> Pedro. >> >>>>> -- >> >>>>> Pedro Tabacof, >> >>>>> Unicamp - Eng. de Computação 08. >> >>>>> >> >>>>> _______________________________________________ >> >>>>> nupic mailing list >> >>>>> [email protected] >> >>>>> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org >> >>>>> >> >>>> >> >>>> _______________________________________________ >> >>>> nupic mailing list >> >>>> [email protected] >> >>>> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org >> >>>> >> >>>> >> >>>> _______________________________________________ >> >>>> nupic mailing list >> >>>> [email protected] >> >>>> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org >> >>>> >> >>> >> >>> >> >>> _______________________________________________ >> >>> nupic mailing list >> >>> [email protected] >> >>> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org >> >>> >> >> >> >> >> >> _______________________________________________ >> >> nupic mailing list >> >> [email protected] >> >> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org >> >> >> > >> > >> > >> > -- >> > Pedro Tabacof, >> > Unicamp - Eng. de Computação 08. >> > >> > _______________________________________________ >> > nupic mailing list >> > [email protected] >> > http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org >> > >> >> _______________________________________________ >> nupic mailing list >> [email protected] >> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org >> > > _______________________________________________ > nupic mailing list > [email protected] > http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org > >
_______________________________________________ nupic mailing list [email protected] http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
