{
  "id": 3586,
  "title": "An odd date",
  "url": "/competitions/event-recommendation-engine-challenge/discussion/3586",
  "author_name": "",
  "post_date": "2013-01-13T03:40:49.390Z",
  "votes": null,
  "comment_count": 4,
  "views": 1568,
  "content": "<p>I haven't had a chance to look over everything, but there's a date in the events.csv that was in 1912 (line #1045210). Is this a typo, or should we account for oddities like this to crop up in the final model?</p>\r\n<p>&nbsp;</p>\r\n<p>Thanks,</p>\r\n<p>-Brett</p>",
  "messages": [
    {
      "id": "19251",
      "postDate": "01/13/2013 03:40:49",
      "content": "<p>I haven't had a chance to look over everything, but there's a date in the events.csv that was in 1912 (line #1045210). Is this a typo, or should we account for oddities like this to crop up in the final model?</p>\r\n<p>&nbsp;</p>\r\n<p>Thanks,</p>\r\n<p>-Brett</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "19267",
      "postDate": "01/13/2013 17:24:44",
      "content": "<p>Hello Brett,</p>\r\n<p>There may be some oddities like this in the final model, and I'll investigate a bit more on our end.</p>\r\n<p>Thanks,<br>\r\nevents4you</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "19277",
      "postDate": "01/14/2013 00:57:51",
      "content": "<p>On this note, the events.csv file seems to have several bad records towards the end. For example look at the last ten lines and you'll see an event ID &quot;27920n20n230587159&quot; and some with spaces in them.</p>\r\n<p>Also the first columns in test.csv and trains.csv are user/event, but are described as event / user on the data page.</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "19279",
      "postDate": "01/14/2013 01:22:46",
      "content": "<p>Hello Sean,</p>\r\n<p>I can't find the line in events.csv you're referring to -- perhaps your file was corrupted somehow?</p>\r\n<p>Thanks for pointing out the mixed up user/event description on the data page. It's been fixed.</p>\r\n<p>Thanks,<br>\r\nevents4you</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "19297",
      "postDate": "01/14/2013 16:27:27",
      "content": "<p>Disregard that. I downloaded again and it's fine. I suppose I would have been surprised if it's corruption, since it ought to have made it un-gunzippable. But, I would not be surprised if something I did munged the file at some point.</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 19267,
      "author_name": "events4you",
      "author_url": "",
      "post_date": "01/13/2013 17:24:44",
      "content": "<p>Hello Brett,</p>\r\n<p>There may be some oddities like this in the final model, and I'll investigate a bit more on our end.</p>\r\n<p>Thanks,<br>\r\nevents4you</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 19277,
      "author_name": "",
      "author_url": "",
      "post_date": "01/14/2013 00:57:51",
      "content": "<p>On this note, the events.csv file seems to have several bad records towards the end. For example look at the last ten lines and you'll see an event ID &quot;27920n20n230587159&quot; and some with spaces in them.</p>\r\n<p>Also the first columns in test.csv and trains.csv are user/event, but are described as event / user on the data page.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 19279,
      "author_name": "events4you",
      "author_url": "",
      "post_date": "01/14/2013 01:22:46",
      "content": "<p>Hello Sean,</p>\r\n<p>I can't find the line in events.csv you're referring to -- perhaps your file was corrupted somehow?</p>\r\n<p>Thanks for pointing out the mixed up user/event description on the data page. It's been fixed.</p>\r\n<p>Thanks,<br>\r\nevents4you</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 19297,
      "author_name": "",
      "author_url": "",
      "post_date": "01/14/2013 16:27:27",
      "content": "<p>Disregard that. I downloaded again and it's fine. I suppose I would have been surprised if it's corruption, since it ought to have made it un-gunzippable. But, I would not be surprised if something I did munged the file at some point.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "19251": "",
    "19267": "",
    "19277": "",
    "19279": "",
    "19297": ""
  },
  "source": "meta"
}