{
  "id": 114989,
  "title": "Wind speed and direction ",
  "url": "/competitions/ashrae-energy-prediction/discussion/114989",
  "author_name": "",
  "post_date": "2019-10-30T15:07:43.377983400Z",
  "votes": 1,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Did anyone find either of them useful ? In my 5-fold cross validation in all the 5 models they came last in place, apparently with 0 feature importance. Same goes for precip_depth_1_hr. I wonder still why some of the good public models are using them.  Did anyone found a way to convert them into something useful ?</p>",
  "messages": [
    {
      "id": "661667",
      "postDate": "10/30/2019 15:07:43",
      "content": "<p>Did anyone find either of them useful ? In my 5-fold cross validation in all the 5 models they came last in place, apparently with 0 feature importance. Same goes for precip_depth_1_hr. I wonder still why some of the good public models are using them.  Did anyone found a way to convert them into something useful ?</p>",
      "rawMarkdown": "Did anyone find either of them useful ? In my 5-fold cross validation in all the 5 models they came last in place, apparently with 0 feature importance. Same goes for precip_depth_1_hr. I wonder still why some of the good public models are using them.  Did anyone found a way to convert them into something useful ?",
      "votes": null
    },
    {
      "id": "661703",
      "postDate": "10/30/2019 16:01:33",
      "content": "<p>You can use it to introduce some weather indexes. I tried to use it.</p>",
      "rawMarkdown": "You can use it to introduce some weather indexes. I tried to use it.",
      "votes": null
    },
    {
      "id": "661711",
      "postDate": "10/30/2019 16:10:35",
      "content": "<p>Did it improve results ? How did it come out in feature importance ?</p>",
      "rawMarkdown": "Did it improve results ? How did it come out in feature importance ?",
      "votes": null
    },
    {
      "id": "661715",
      "postDate": "10/30/2019 16:22:27",
      "content": "<p>I hope, but I did not make it into submission seriously. There are some difficulties to submit my works (computing power etc...)  :(</p>",
      "rawMarkdown": "I hope, but I did not make it into submission seriously. There are some difficulties to submit my works (computing power etc...)  :(",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 661703,
      "author_name": "hanjoonchoe",
      "author_url": "",
      "post_date": "10/30/2019 16:01:33",
      "content": "<p>You can use it to introduce some weather indexes. I tried to use it.</p>",
      "votes": null,
      "replies": [
        {
          "id": 661711,
          "author_name": "khairulislam",
          "author_url": "",
          "post_date": "10/30/2019 16:10:35",
          "content": "<p>Did it improve results ? How did it come out in feature importance ?</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 661715,
          "author_name": "hanjoonchoe",
          "author_url": "",
          "post_date": "10/30/2019 16:22:27",
          "content": "<p>I hope, but I did not make it into submission seriously. There are some difficulties to submit my works (computing power etc...)  :(</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "661667": "Did anyone find either of them useful ? In my 5-fold cross validation in all the 5 models they came last in place, apparently with 0 feature importance. Same goes for precip_depth_1_hr. I wonder still why some of the good public models are using them.  Did anyone found a way to convert them into something useful ?",
    "661703": "You can use it to introduce some weather indexes. I tried to use it.",
    "661711": "Did it improve results ? How did it come out in feature importance ?",
    "661715": "I hope, but I did not make it into submission seriously. There are some difficulties to submit my works (computing power etc...)  :("
  },
  "source": "meta"
}