{
  "id": 3677,
  "title": "asdipoistion.csv data for PublicLeaderboardSet",
  "url": "/competitions/flight/discussion/3677",
  "author_name": "",
  "post_date": "2013-01-23T19:16:54.827Z",
  "votes": null,
  "comment_count": 2,
  "views": 2422,
  "content": "<p>Hello</p>\r\n<p><span style=\"line-height:1.4em\">Out of the 26459 flights to be predicted, it seems that about half&nbsp;of them have no position record available in asdiposition.csv</span></p>\r\n<p>I get that the position data should be filtered so that it doesn't reveal the actual arrival times, but I didn't expect that half of them will have absolutely no position record available.. Since the flights to be predicted are already in air, shouldn't\r\n there be at least one for each flight?</p>\r\n<p>If this was something intentional, can you confirm that it will be similar for the final evaluation set as well?</p>\r\n<p>or am I missing something?</p>\r\n<p>&nbsp;</p>\r\n<p>Thank you in advance</p>",
  "messages": [
    {
      "id": "19635",
      "postDate": "01/23/2013 19:16:54",
      "content": "<p>Hello</p>\r\n<p><span style=\"line-height:1.4em\">Out of the 26459 flights to be predicted, it seems that about half&nbsp;of them have no position record available in asdiposition.csv</span></p>\r\n<p>I get that the position data should be filtered so that it doesn't reveal the actual arrival times, but I didn't expect that half of them will have absolutely no position record available.. Since the flights to be predicted are already in air, shouldn't\r\n there be at least one for each flight?</p>\r\n<p>If this was something intentional, can you confirm that it will be similar for the final evaluation set as well?</p>\r\n<p>or am I missing something?</p>\r\n<p>&nbsp;</p>\r\n<p>Thank you in advance</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "19640",
      "postDate": "01/23/2013 20:17:39",
      "content": "<p>yes; you are correct. I find the same thing</p>\r\n<p>ftr<em>groundspeed ftr</em>altitude <br>\r\n0.436070902 0.436070902 </p>\r\n<p>43% of them are missing.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "19646",
      "postDate": "01/23/2013 23:45:29",
      "content": "<p>The reason is that the position data is truncated by default. Even in training set most of position data starts int he middle. See\r\n<a href=\"http://www.gequest.com/c/flight/forums/t/3670/flights-positions-are-incomplete-in-asdiposition-csv\">\r\nthis post</a></p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 19640,
      "author_name": "rkirana",
      "author_url": "",
      "post_date": "01/23/2013 20:17:39",
      "content": "<p>yes; you are correct. I find the same thing</p>\r\n<p>ftr<em>groundspeed ftr</em>altitude <br>\r\n0.436070902 0.436070902 </p>\r\n<p>43% of them are missing.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 19646,
      "author_name": "vspinu",
      "author_url": "",
      "post_date": "01/23/2013 23:45:29",
      "content": "<p>The reason is that the position data is truncated by default. Even in training set most of position data starts int he middle. See\r\n<a href=\"http://www.gequest.com/c/flight/forums/t/3670/flights-positions-are-incomplete-in-asdiposition-csv\">\r\nthis post</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "19635": "",
    "19640": "",
    "19646": ""
  },
  "source": "meta"
}