{
  "id": 9398,
  "title": "Reduced feature set.",
  "url": "/competitions/decoding-the-human-brain/discussion/9398",
  "author_name": "maveric",
  "post_date": "2014-06-05T21:24:04.583000",
  "votes": 2,
  "comment_count": 0,
  "views": 970,
  "content": "<p>As discussed elsewhere, and probably a bit late, there is some potential to reduce the size of feature set. A simple frequency domain transform (on a per sensor, 0.5 sec post stimuls time signal), limited to 25% of single sided spectrum yields features that gives 0.65930 on leaderboard (vowpal wabbit). The total number of features accorss all 306 sensor streams is ~5k. &nbsp;While the error score is not particularly impressive, the smaller feature allows trying out a variety of models quickly. A similar albeit a little bit lower score is obtained by applying a fc=0.25 filter in time domain, followed by down-sampling by an order of 4. A sklearn logistic regression seemed &nbsp;to take ~4gb ram.&nbsp;</p>\n<p>Similar reduction in feature set should be possible by spatial decomposition across sensors.</p>\n<p>There is a risk that this reduced set of feature will loose the discriminating information useful to capture variability accross train&nbsp;and test subjects., though the LB score does not seem to indicate it.</p>",
  "messages": [
    {
      "id": 48738,
      "postDate": "2014-06-05T21:24:04.583Z",
      "content": "<p>As discussed elsewhere, and probably a bit late, there is some potential to reduce the size of feature set. A simple frequency domain transform (on a per sensor, 0.5 sec post stimuls time signal), limited to 25% of single sided spectrum yields features that gives 0.65930 on leaderboard (vowpal wabbit). The total number of features accorss all 306 sensor streams is ~5k. &nbsp;While the error score is not particularly impressive, the smaller feature allows trying out a variety of models quickly. A similar albeit a little bit lower score is obtained by applying a fc=0.25 filter in time domain, followed by down-sampling by an order of 4. A sklearn logistic regression seemed &nbsp;to take ~4gb ram.&nbsp;</p>\n<p>Similar reduction in feature set should be possible by spatial decomposition across sensors.</p>\n<p>There is a risk that this reduced set of feature will loose the discriminating information useful to capture variability accross train&nbsp;and test subjects., though the LB score does not seem to indicate it.</p>",
      "votes": 2
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "48738": ""
  }
}