{
  "id": 18341,
  "title": "Human help in algorithm",
  "url": "/competitions/second-annual-data-science-bowl/discussion/18341",
  "author_name": "",
  "post_date": "2016-01-10T23:52:18.970Z",
  "votes": null,
  "comment_count": 2,
  "views": 555,
  "content": "<p>Reading up on some methods for calculating the volumes that I'd like to try (or consider to at least), I found one that requires human input in the form of contours drawn on one image per case by a human observer.</p>\n\n<p>Would input like this be allowed for this competition? </p>\n\n<p>My guess is that it would not be allowed, though I would like to check what others think.</p>",
  "messages": [
    {
      "id": "104249",
      "postDate": "01/10/2016 23:52:18",
      "content": "<p>Reading up on some methods for calculating the volumes that I'd like to try (or consider to at least), I found one that requires human input in the form of contours drawn on one image per case by a human observer.</p>\n\n<p>Would input like this be allowed for this competition? </p>\n\n<p>My guess is that it would not be allowed, though I would like to check what others think.</p>",
      "rawMarkdown": "Reading up on some methods for calculating the volumes that I'd like to try (or consider to at least), I found one that requires human input in the form of contours drawn on one image per case by a human observer.\r\n\r\nWould input like this be allowed for this competition? \r\n\r\nMy guess is that it would not be allowed, though I would like to check what others think.",
      "votes": null
    },
    {
      "id": "104250",
      "postDate": "01/11/2016 00:22:32",
      "content": "<p>You can do whatever you want to the training data. But, when it comes to the test data, you won't be able to do any type of hand labeling.</p>",
      "rawMarkdown": "You can do whatever you want to the training data. But, when it comes to the test data, you won't be able to do any type of hand labeling.",
      "votes": null
    },
    {
      "id": "104266",
      "postDate": "01/11/2016 07:09:25",
      "content": "<p>Thanks  @inversion, \nSo that clearly requires me to find out a way to automatically 'label' the LV. \nI am going to need to think about it.</p>",
      "rawMarkdown": "Thanks  @inversion, \r\nSo that clearly requires me to find out a way to automatically 'label' the LV. \r\nI am going to need to think about it.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 104250,
      "author_name": "inversion",
      "author_url": "",
      "post_date": "01/11/2016 00:22:32",
      "content": "<p>You can do whatever you want to the training data. But, when it comes to the test data, you won't be able to do any type of hand labeling.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 104266,
      "author_name": "florensdewit",
      "author_url": "",
      "post_date": "01/11/2016 07:09:25",
      "content": "<p>Thanks  @inversion, \nSo that clearly requires me to find out a way to automatically 'label' the LV. \nI am going to need to think about it.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "104249": "Reading up on some methods for calculating the volumes that I'd like to try (or consider to at least), I found one that requires human input in the form of contours drawn on one image per case by a human observer.\r\n\r\nWould input like this be allowed for this competition? \r\n\r\nMy guess is that it would not be allowed, though I would like to check what others think.",
    "104250": "You can do whatever you want to the training data. But, when it comes to the test data, you won't be able to do any type of hand labeling.",
    "104266": "Thanks  @inversion, \r\nSo that clearly requires me to find out a way to automatically 'label' the LV. \r\nI am going to need to think about it."
  },
  "source": "meta"
}