{
  "id": 505535,
  "title": "Will there be a huge shake up?",
  "url": "/competitions/home-credit-credit-risk-model-stability/discussion/505535",
  "author_name": "",
  "post_date": "2024-05-18T01:38:54.790067300Z",
  "votes": 5,
  "comment_count": 1,
  "views": 0,
  "content": "<ul>\n<li><p>Are high scores on the leaderboard due to metric hacking? If so, do you think the rankings will be different on the private test data? </p></li>\n<li><p>Is it possible that those currently ranked 1000th or 2000th end up winning the competition?</p></li>\n<li><p>Should we then select submissions with the highest scores on local evaluation datasets?</p></li>\n</ul>",
  "messages": [
    {
      "id": "2821194",
      "postDate": "05/18/2024 01:38:54",
      "content": "<ul>\n<li><p>Are high scores on the leaderboard due to metric hacking? If so, do you think the rankings will be different on the private test data? </p></li>\n<li><p>Is it possible that those currently ranked 1000th or 2000th end up winning the competition?</p></li>\n<li><p>Should we then select submissions with the highest scores on local evaluation datasets?</p></li>\n</ul>",
      "rawMarkdown": "Are high scores on the leaderboard due to metric hacking? If so, do you think the rankings will be different on the private test data? \n\n- Is it possible that those currently ranked 1000th or 2000th end up winning the competition?\n\n- Should we then select submissions with the highest scores on local evaluation datasets?",
      "votes": null
    },
    {
      "id": "2821394",
      "postDate": "05/18/2024 05:16:07",
      "content": "<p>I think proper metric hacking should stick along but I am doubtful with the latest round of hack (the one-liner) <a href=\"https://www.kaggle.com/jeannkouagou\" target=\"_blank\">@jeannkouagou</a> </p>",
      "rawMarkdown": "I think proper metric hacking should stick along but I am doubtful with the latest round of hack (the one-liner) @jeannkouagou",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2821394,
      "author_name": "ravi20076",
      "author_url": "",
      "post_date": "05/18/2024 05:16:07",
      "content": "<p>I think proper metric hacking should stick along but I am doubtful with the latest round of hack (the one-liner) <a href=\"https://www.kaggle.com/jeannkouagou\" target=\"_blank\">@jeannkouagou</a> </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2821194": "Are high scores on the leaderboard due to metric hacking? If so, do you think the rankings will be different on the private test data? \n\n- Is it possible that those currently ranked 1000th or 2000th end up winning the competition?\n\n- Should we then select submissions with the highest scores on local evaluation datasets?",
    "2821394": "I think proper metric hacking should stick along but I am doubtful with the latest round of hack (the one-liner) @jeannkouagou"
  },
  "source": "meta"
}