{
  "id": 5257,
  "title": "Train data question",
  "url": "/competitions/accelerometer-biometric-competition/discussion/5257",
  "author_name": "",
  "post_date": "2013-08-01T07:22:13.800Z",
  "votes": null,
  "comment_count": 6,
  "views": 2970,
  "content": "<p>I think I misunderstand smth but train data file has only 5 unique device ids ([ 7 8 9 12 23]). So how could we identify true/false statements from question file since there are&nbsp;387 ids?</p>",
  "messages": [
    {
      "id": "27977",
      "postDate": "08/01/2013 07:22:13",
      "content": "<p>I think I misunderstand smth but train data file has only 5 unique device ids ([ 7 8 9 12 23]). So how could we identify true/false statements from question file since there are&nbsp;387 ids?</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "27980",
      "postDate": "08/01/2013 07:29:43",
      "content": "<p>That's strange, I have exactly &nbsp;unique 387 ids in my training file. The file might have been updated.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "28000",
      "postDate": "08/01/2013 12:51:14",
      "content": "<p>The train file has not been updated as far as I know. Running this in your shell should return the right number (387 + 1 header = 388)</p>\n<p><code>cat train.csv | cut -d, -f5 | uniq | wc -l</code></p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "28032",
      "postDate": "08/02/2013 00:21:38",
      "content": "<p>[quote=adveboy;27977]</p>\n<p>I think I misunderstand smth but train data file has only 5 unique device ids ([ 7 8 9 12 23]). So how could we identify true/false statements from question file since there are&nbsp;387 ids?</p>\n<p>[/quote]</p>\n<p>are you using excel? if you do, keep in mind that it can't handle more than 1,048,576 rows.. so you're missing &quot;just&quot; others 29 millions rows! xD</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "28033",
      "postDate": "08/02/2013 01:02:34",
      "content": "<p>exactly Alessandro, i&nbsp;forgot that I read in matlab only the number of lines i saw in excel, baby bug :)</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "28437",
      "postDate": "08/08/2013 15:19:50",
      "content": "<p>I loaded the training data and got&nbsp;29,563,983 rows. &nbsp;This is not quite 30mil, am I missing something?</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "28438",
      "postDate": "08/08/2013 15:26:25",
      "content": "<p>You are not missing anything.&nbsp;</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 27980,
      "author_name": "issamlaradji",
      "author_url": "",
      "post_date": "08/01/2013 07:29:43",
      "content": "<p>That's strange, I have exactly &nbsp;unique 387 ids in my training file. The file might have been updated.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 28000,
      "author_name": "wcukierski",
      "author_url": "",
      "post_date": "08/01/2013 12:51:14",
      "content": "<p>The train file has not been updated as far as I know. Running this in your shell should return the right number (387 + 1 header = 388)</p>\n<p><code>cat train.csv | cut -d, -f5 | uniq | wc -l</code></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 28032,
      "author_name": "alzmcr",
      "author_url": "",
      "post_date": "08/02/2013 00:21:38",
      "content": "<p>[quote=adveboy;27977]</p>\n<p>I think I misunderstand smth but train data file has only 5 unique device ids ([ 7 8 9 12 23]). So how could we identify true/false statements from question file since there are&nbsp;387 ids?</p>\n<p>[/quote]</p>\n<p>are you using excel? if you do, keep in mind that it can't handle more than 1,048,576 rows.. so you're missing &quot;just&quot; others 29 millions rows! xD</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 28033,
      "author_name": "",
      "author_url": "",
      "post_date": "08/02/2013 01:02:34",
      "content": "<p>exactly Alessandro, i&nbsp;forgot that I read in matlab only the number of lines i saw in excel, baby bug :)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 28437,
      "author_name": "mchangun",
      "author_url": "",
      "post_date": "08/08/2013 15:19:50",
      "content": "<p>I loaded the training data and got&nbsp;29,563,983 rows. &nbsp;This is not quite 30mil, am I missing something?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 28438,
      "author_name": "geoff3",
      "author_url": "",
      "post_date": "08/08/2013 15:26:25",
      "content": "<p>You are not missing anything.&nbsp;</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "27977": "",
    "27980": "",
    "28000": "",
    "28032": "",
    "28033": "",
    "28437": "",
    "28438": ""
  },
  "source": "meta"
}