{
  "id": 186270,
  "title": "Is public leaderboard based on a different test data?",
  "url": "/competitions/osic-pulmonary-fibrosis-progression/discussion/186270",
  "author_name": "",
  "post_date": "2020-09-23T20:05:04.436866500Z",
  "votes": 2,
  "comment_count": 5,
  "views": 0,
  "content": "<p>Is the test data used for the public LB the same as provided in the Data section? or is it completely different?</p>\n<p>I couldn't find any notes on this.</p>",
  "messages": [
    {
      "id": "1024381",
      "postDate": "09/23/2020 20:05:04",
      "content": "<p>Is the test data used for the public LB the same as provided in the Data section? or is it completely different?</p>\n<p>I couldn't find any notes on this.</p>",
      "rawMarkdown": "Is the test data used for the public LB the same as provided in the Data section? or is it completely different?\n\nI couldn't find any notes on this.",
      "votes": null
    },
    {
      "id": "1024383",
      "postDate": "09/23/2020 20:10:54",
      "content": "<p>check the evaluation criteria. sometimes public LB based on 30% test data or 50% test data</p>",
      "rawMarkdown": "check the evaluation criteria. sometimes public LB based on 30% test data or 50% test data",
      "votes": null
    },
    {
      "id": "1024403",
      "postDate": "09/23/2020 20:34:29",
      "content": "<p>That is stated, but most kernel-competitions use a completely different test data from the data provided publicly in the Data section.</p>",
      "rawMarkdown": "That is stated, but most kernel-competitions use a completely different test data from the data provided publicly in the Data section.",
      "votes": null
    },
    {
      "id": "1024436",
      "postDate": "09/23/2020 21:28:15",
      "content": "<p>The \"visible\" test data is just 5 patients from the train data. It is just to provide the structure of the test.csv file and directories.</p>\n<p>The Leaderboard is based on 15% of the \"hidden\" test data. The final score is based on the other 85% (if I read the heading in the Leaderboard correctly).</p>",
      "rawMarkdown": "The \"visible\" test data is just 5 patients from the train data. It is just to provide the structure of the test.csv file and directories.\n\nThe Leaderboard is based on 15% of the \"hidden\" test data. The final score is based on the other 85% (if I read the heading in the Leaderboard correctly).",
      "votes": null
    },
    {
      "id": "1024454",
      "postDate": "09/23/2020 21:59:47",
      "content": "<p>Thank you <a href=\"https://www.kaggle.com/richardepstein\" target=\"_blank\">@richardepstein</a> </p>",
      "rawMarkdown": "Thank you @richardepstein",
      "votes": null
    },
    {
      "id": "1025431",
      "postDate": "09/24/2020 14:47:21",
      "content": "<p>Amazing!</p>",
      "rawMarkdown": "Amazing!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1024383,
      "author_name": "akash14",
      "author_url": "",
      "post_date": "09/23/2020 20:10:54",
      "content": "<p>check the evaluation criteria. sometimes public LB based on 30% test data or 50% test data</p>",
      "votes": null,
      "replies": [
        {
          "id": 1024403,
          "author_name": "imoore",
          "author_url": "",
          "post_date": "09/23/2020 20:34:29",
          "content": "<p>That is stated, but most kernel-competitions use a completely different test data from the data provided publicly in the Data section.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1024436,
      "author_name": "richardepstein",
      "author_url": "",
      "post_date": "09/23/2020 21:28:15",
      "content": "<p>The \"visible\" test data is just 5 patients from the train data. It is just to provide the structure of the test.csv file and directories.</p>\n<p>The Leaderboard is based on 15% of the \"hidden\" test data. The final score is based on the other 85% (if I read the heading in the Leaderboard correctly).</p>",
      "votes": null,
      "replies": [
        {
          "id": 1024454,
          "author_name": "imoore",
          "author_url": "",
          "post_date": "09/23/2020 21:59:47",
          "content": "<p>Thank you <a href=\"https://www.kaggle.com/richardepstein\" target=\"_blank\">@richardepstein</a> </p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1025431,
      "author_name": "salmaneunus",
      "author_url": "",
      "post_date": "09/24/2020 14:47:21",
      "content": "<p>Amazing!</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1024381": "Is the test data used for the public LB the same as provided in the Data section? or is it completely different?\n\nI couldn't find any notes on this.",
    "1024383": "check the evaluation criteria. sometimes public LB based on 30% test data or 50% test data",
    "1024403": "That is stated, but most kernel-competitions use a completely different test data from the data provided publicly in the Data section.",
    "1024436": "The \"visible\" test data is just 5 patients from the train data. It is just to provide the structure of the test.csv file and directories.\n\nThe Leaderboard is based on 15% of the \"hidden\" test data. The final score is based on the other 85% (if I read the heading in the Leaderboard correctly).",
    "1024454": "Thank you @richardepstein",
    "1025431": "Amazing!"
  },
  "source": "meta"
}