{
  "id": 5048,
  "title": "Combining all the Features",
  "url": "/competitions/belkin-energy-disaggregation-competition/discussion/5048",
  "author_name": "",
  "post_date": "2013-07-09T05:30:40.883Z",
  "votes": null,
  "comment_count": 2,
  "views": 1555,
  "content": "<p>Hi,</p>\r\n<p>In the processed data given, there are three main feature sets - L1, L2, HF. Each one of these feature sets are measured at different frequencies and therefore have different matrix dimensions. Appreciate if someone could give me some pointers on how to\r\n combine all of these different frequency features (L1,L2,HF) in MATLAB.</p>\r\n<p>Thanks.</p>\r\n<p>-Rahul</p>",
  "messages": [
    {
      "id": "26842",
      "postDate": "07/09/2013 05:30:40",
      "content": "<p>Hi,</p>\r\n<p>In the processed data given, there are three main feature sets - L1, L2, HF. Each one of these feature sets are measured at different frequencies and therefore have different matrix dimensions. Appreciate if someone could give me some pointers on how to\r\n combine all of these different frequency features (L1,L2,HF) in MATLAB.</p>\r\n<p>Thanks.</p>\r\n<p>-Rahul</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "27160",
      "postDate": "07/13/2013 21:40:17",
      "content": "<p>Some potential ideas:</p>\r\n<p>1. You could resample them. We have included TimeTicks for each of these matrices.</p>\r\n<p>2. You could build a new vector that uses some information from all three.</p>\r\n<p></p>\r\n<p>In general, I do not think combining the the way you described is a good idea. They should be combined in a more semantic way that includes additional information from electrical events.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "27228",
      "postDate": "07/15/2013 12:40:14",
      "content": "<p>Look into time series data structures and query each of the feature sets for the appropriate time &quot;slice&quot; instead of trying to join all these sets together like one big table.</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 27160,
      "author_name": "sidhantgupta",
      "author_url": "",
      "post_date": "07/13/2013 21:40:17",
      "content": "<p>Some potential ideas:</p>\r\n<p>1. You could resample them. We have included TimeTicks for each of these matrices.</p>\r\n<p>2. You could build a new vector that uses some information from all three.</p>\r\n<p></p>\r\n<p>In general, I do not think combining the the way you described is a good idea. They should be combined in a more semantic way that includes additional information from electrical events.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 27228,
      "author_name": "zacstewart",
      "author_url": "",
      "post_date": "07/15/2013 12:40:14",
      "content": "<p>Look into time series data structures and query each of the feature sets for the appropriate time &quot;slice&quot; instead of trying to join all these sets together like one big table.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "26842": "",
    "27160": "",
    "27228": ""
  },
  "source": "meta"
}