{
  "id": 19234,
  "title": "Test data and validation results",
  "url": "/competitions/second-annual-data-science-bowl/discussion/19234",
  "author_name": "",
  "post_date": "2016-02-29T12:07:59.950Z",
  "votes": 3,
  "comment_count": 3,
  "views": 599,
  "content": "<p>Since it would be good to minimize trivial changes to the model related with reading in the new test data and validation results, could you clarify the format of the released data?</p>\n\n<p>Specifically I wanted to ask:</p>\n\n<ol>\n<li><p>I guess that the test set will unzip into the directory <code>test</code> and would contain studies with consecutive ids. What would be the range of ids? (say 701-XXX?)  </p></li>\n<li><p>Would the systole and diastole results for the validation dataset be given in a separate file <code>validate.csv</code>  with the same format as <code>train.csv</code> or we would get some combined file with id's from 1 to 700?</p></li>\n</ol>",
  "messages": [
    {
      "id": "109681",
      "postDate": "02/29/2016 12:07:59",
      "content": "<p>Since it would be good to minimize trivial changes to the model related with reading in the new test data and validation results, could you clarify the format of the released data?</p>\n\n<p>Specifically I wanted to ask:</p>\n\n<ol>\n<li><p>I guess that the test set will unzip into the directory <code>test</code> and would contain studies with consecutive ids. What would be the range of ids? (say 701-XXX?)  </p></li>\n<li><p>Would the systole and diastole results for the validation dataset be given in a separate file <code>validate.csv</code>  with the same format as <code>train.csv</code> or we would get some combined file with id's from 1 to 700?</p></li>\n</ol>",
      "rawMarkdown": "Since it would be good to minimize trivial changes to the model related with reading in the new test data and validation results, could you clarify the format of the released data?\r\n\r\nSpecifically I wanted to ask:\r\n\r\n1. I guess that the test set will unzip into the directory `test` and would contain studies with consecutive ids. What would be the range of ids? (say 701-XXX?)  \r\n\r\n2. Would the systole and diastole results for the validation dataset be given in a separate file `validate.csv`  with the same format as `train.csv` or we would get some combined file with id's from 1 to 700?",
      "votes": null
    },
    {
      "id": "110370",
      "postDate": "03/04/2016 21:49:30",
      "content": "<p>I am also wondering about these two questions. Inconsistent format between the uploaded model and the released data will cause unnecessary errors. Can organizer provides the details ahead of time? Thank you! </p>",
      "rawMarkdown": "I am also wondering about these two questions. Inconsistent format between the uploaded model and the released data will cause unnecessary errors. Can organizer provides the details ahead of time? Thank you!",
      "votes": null
    },
    {
      "id": "110375",
      "postDate": "03/04/2016 22:35:31",
      "content": "<ol>\n<li>Correct!</li>\n<li>Yes, we will give a validate.csv that looks like train.csv</li>\n</ol>",
      "rawMarkdown": "1. Correct!\r\n 2. Yes, we will give a validate.csv that looks like train.csv",
      "votes": null
    },
    {
      "id": "110378",
      "postDate": "03/04/2016 23:11:28",
      "content": "<p>[quote=William Cukierski;110375]</p>\n\n<ol>\n<li>Correct!</li>\n<li>Yes, we will give a validate.csv that looks like train.csv</li>\n</ol>\n\n<p>[/quote]</p>\n\n<p>@William Cukierski - Thanks! Could you let us know what will be the range of id's in the test set? i.e. the upper limit in 701-XXX?</p>",
      "rawMarkdown": "[quote=William Cukierski;110375]\r\n\r\n 1. Correct!\r\n 2. Yes, we will give a validate.csv that looks like train.csv\r\n\r\n[/quote]\r\n\r\n@William Cukierski - Thanks! Could you let us know what will be the range of id's in the test set? i.e. the upper limit in 701-XXX?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 110370,
      "author_name": "boshuang",
      "author_url": "",
      "post_date": "03/04/2016 21:49:30",
      "content": "<p>I am also wondering about these two questions. Inconsistent format between the uploaded model and the released data will cause unnecessary errors. Can organizer provides the details ahead of time? Thank you! </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 110375,
      "author_name": "wcukierski",
      "author_url": "",
      "post_date": "03/04/2016 22:35:31",
      "content": "<ol>\n<li>Correct!</li>\n<li>Yes, we will give a validate.csv that looks like train.csv</li>\n</ol>",
      "votes": null,
      "replies": []
    },
    {
      "id": 110378,
      "author_name": "romualdj",
      "author_url": "",
      "post_date": "03/04/2016 23:11:28",
      "content": "<p>[quote=William Cukierski;110375]</p>\n\n<ol>\n<li>Correct!</li>\n<li>Yes, we will give a validate.csv that looks like train.csv</li>\n</ol>\n\n<p>[/quote]</p>\n\n<p>@William Cukierski - Thanks! Could you let us know what will be the range of id's in the test set? i.e. the upper limit in 701-XXX?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "109681": "Since it would be good to minimize trivial changes to the model related with reading in the new test data and validation results, could you clarify the format of the released data?\r\n\r\nSpecifically I wanted to ask:\r\n\r\n1. I guess that the test set will unzip into the directory `test` and would contain studies with consecutive ids. What would be the range of ids? (say 701-XXX?)  \r\n\r\n2. Would the systole and diastole results for the validation dataset be given in a separate file `validate.csv`  with the same format as `train.csv` or we would get some combined file with id's from 1 to 700?",
    "110370": "I am also wondering about these two questions. Inconsistent format between the uploaded model and the released data will cause unnecessary errors. Can organizer provides the details ahead of time? Thank you!",
    "110375": "1. Correct!\r\n 2. Yes, we will give a validate.csv that looks like train.csv",
    "110378": "[quote=William Cukierski;110375]\r\n\r\n 1. Correct!\r\n 2. Yes, we will give a validate.csv that looks like train.csv\r\n\r\n[/quote]\r\n\r\n@William Cukierski - Thanks! Could you let us know what will be the range of id's in the test set? i.e. the upper limit in 701-XXX?"
  },
  "source": "meta"
}