{
  "id": 173298,
  "title": "Run Time Requirement Question",
  "url": "/competitions/landmark-retrieval-2020/discussion/173298",
  "author_name": "",
  "post_date": "2020-08-08T16:29:18.446535400Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I'm wondering if my model is able to run within the 9 hours requirement now, would it be guaranteed to finish within 9 hours during the private leaderboard evaluation? </p>\n\n<p>To be more specific, is the public score we get right now run on the entire 100% of the data or just 33% of the data? And similarly, during the private leaderboard evaluation, is the model run on 66% of the data or 100% of the data?</p>\n\n<p>If the model is only run on 33% of the data now, does it mean we need to run within 4.5 or 3 hours to make sure we can finish within required 9 hours during private leaderboard evaluation? </p>",
  "messages": [
    {
      "id": "963065",
      "postDate": "08/08/2020 16:29:18",
      "content": "<p>I'm wondering if my model is able to run within the 9 hours requirement now, would it be guaranteed to finish within 9 hours during the private leaderboard evaluation? </p>\n\n<p>To be more specific, is the public score we get right now run on the entire 100% of the data or just 33% of the data? And similarly, during the private leaderboard evaluation, is the model run on 66% of the data or 100% of the data?</p>\n\n<p>If the model is only run on 33% of the data now, does it mean we need to run within 4.5 or 3 hours to make sure we can finish within required 9 hours during private leaderboard evaluation? </p>",
      "rawMarkdown": "I'm wondering if my model is able to run within the 9 hours requirement now, would it be guaranteed to finish within 9 hours during the private leaderboard evaluation? \n\nTo be more specific, is the public score we get right now run on the entire 100% of the data or just 33% of the data? And similarly, during the private leaderboard evaluation, is the model run on 66% of the data or 100% of the data?\n\nIf the model is only run on 33% of the data now, does it mean we need to run within 4.5 or 3 hours to make sure we can finish within required 9 hours during private leaderboard evaluation?",
      "votes": null
    },
    {
      "id": "963142",
      "postDate": "08/08/2020 17:34:40",
      "content": "<p>I think the model runs on whole data and we only see the public result. Recently, Kaggle adapted this way rather than re-runs.</p>",
      "rawMarkdown": "I think the model runs on whole data and we only see the public result. Recently, Kaggle adapted this way rather than re-runs.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 963142,
      "author_name": "aerdem4",
      "author_url": "",
      "post_date": "08/08/2020 17:34:40",
      "content": "<p>I think the model runs on whole data and we only see the public result. Recently, Kaggle adapted this way rather than re-runs.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "963065": "I'm wondering if my model is able to run within the 9 hours requirement now, would it be guaranteed to finish within 9 hours during the private leaderboard evaluation? \n\nTo be more specific, is the public score we get right now run on the entire 100% of the data or just 33% of the data? And similarly, during the private leaderboard evaluation, is the model run on 66% of the data or 100% of the data?\n\nIf the model is only run on 33% of the data now, does it mean we need to run within 4.5 or 3 hours to make sure we can finish within required 9 hours during private leaderboard evaluation?",
    "963142": "I think the model runs on whole data and we only see the public result. Recently, Kaggle adapted this way rather than re-runs."
  },
  "source": "meta"
}