{
  "id": 552294,
  "title": "Ensemble of public notebooks",
  "url": "/competitions/child-mind-institute-problematic-internet-use/discussion/552294",
  "author_name": "",
  "post_date": "2024-12-18T22:01:14.694202Z",
  "votes": 1,
  "comment_count": 4,
  "views": 0,
  "content": "<p>There is a notebook <a href=\"https://www.kaggle.com/code/vyacheslavbolotin/cmi-ensemble-of-solutions\" target=\"_blank\">CMI | Ensemble of solutions</a> that combines the top-rated submissions by creating a weighted ensemble. This person also claims of the effectiveness of this concept in previous competitions. Is this strategy worth a shot in the final hours or is it just a fluke. Interested to know your thoughts on this?</p>",
  "messages": [
    {
      "id": "3075453",
      "postDate": "12/18/2024 22:01:14",
      "content": "<p>There is a notebook <a href=\"https://www.kaggle.com/code/vyacheslavbolotin/cmi-ensemble-of-solutions\" target=\"_blank\">CMI | Ensemble of solutions</a> that combines the top-rated submissions by creating a weighted ensemble. This person also claims of the effectiveness of this concept in previous competitions. Is this strategy worth a shot in the final hours or is it just a fluke. Interested to know your thoughts on this?</p>",
      "rawMarkdown": "There is a notebook [CMI | Ensemble of solutions](https://www.kaggle.com/code/vyacheslavbolotin/cmi-ensemble-of-solutions) that combines the top-rated submissions by creating a weighted ensemble. This person also claims of the effectiveness of this concept in previous competitions. Is this strategy worth a shot in the final hours or is it just a fluke. Interested to know your thoughts on this?",
      "votes": null
    },
    {
      "id": "3075570",
      "postDate": "12/19/2024 03:18:45",
      "content": "<p>We will know soon</p>",
      "rawMarkdown": "We will know soon",
      "votes": null
    },
    {
      "id": "3075612",
      "postDate": "12/19/2024 04:20:12",
      "content": "<p>Let's at least try it!</p>",
      "rawMarkdown": "Let's at least try it!",
      "votes": null
    },
    {
      "id": "3075678",
      "postDate": "12/19/2024 06:05:38",
      "content": "<p><a href=\"https://www.kaggle.com/nikhil1e9\" target=\"_blank\">@nikhil1e9</a> I suggest you to look at the data and build a model yourself. Simply resubmitting a public kernel is not the correct way to approach any competition. </p>\n<p>This competition is doable in 1 day, so you still have some time to structure a good approach.</p>",
      "rawMarkdown": "nikhil1e9 I suggest you to look at the data and build a model yourself. Simply resubmitting a public kernel is not the correct way to approach any competition. \n\nThis competition is doable in 1 day, so you still have some time to structure a good approach.",
      "votes": null
    },
    {
      "id": "3075820",
      "postDate": "12/19/2024 09:59:29",
      "content": "<p>I also won't trust these dubious methods <a href=\"https://www.kaggle.com/ravi20076\" target=\"_blank\">@ravi20076</a> but the CV-LB scores are so inconsistent for me that I am not so sure about the correct way to approach this anymore. The top public notebooks are doomed for sure though.</p>",
      "rawMarkdown": "I also won't trust these dubious methods @ravi20076 but the CV-LB scores are so inconsistent for me that I am not so sure about the correct way to approach this anymore. The top public notebooks are doomed for sure though.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3075570,
      "author_name": "trcnveli",
      "author_url": "",
      "post_date": "12/19/2024 03:18:45",
      "content": "<p>We will know soon</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 3075612,
      "author_name": "mrsimple07",
      "author_url": "",
      "post_date": "12/19/2024 04:20:12",
      "content": "<p>Let's at least try it!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 3075678,
      "author_name": "ravi20076",
      "author_url": "",
      "post_date": "12/19/2024 06:05:38",
      "content": "<p><a href=\"https://www.kaggle.com/nikhil1e9\" target=\"_blank\">@nikhil1e9</a> I suggest you to look at the data and build a model yourself. Simply resubmitting a public kernel is not the correct way to approach any competition. </p>\n<p>This competition is doable in 1 day, so you still have some time to structure a good approach.</p>",
      "votes": null,
      "replies": [
        {
          "id": 3075820,
          "author_name": "nikhil1e9",
          "author_url": "",
          "post_date": "12/19/2024 09:59:29",
          "content": "<p>I also won't trust these dubious methods <a href=\"https://www.kaggle.com/ravi20076\" target=\"_blank\">@ravi20076</a> but the CV-LB scores are so inconsistent for me that I am not so sure about the correct way to approach this anymore. The top public notebooks are doomed for sure though.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3075453": "There is a notebook [CMI | Ensemble of solutions](https://www.kaggle.com/code/vyacheslavbolotin/cmi-ensemble-of-solutions) that combines the top-rated submissions by creating a weighted ensemble. This person also claims of the effectiveness of this concept in previous competitions. Is this strategy worth a shot in the final hours or is it just a fluke. Interested to know your thoughts on this?",
    "3075570": "We will know soon",
    "3075612": "Let's at least try it!",
    "3075678": "nikhil1e9 I suggest you to look at the data and build a model yourself. Simply resubmitting a public kernel is not the correct way to approach any competition. \n\nThis competition is doable in 1 day, so you still have some time to structure a good approach.",
    "3075820": "I also won't trust these dubious methods @ravi20076 but the CV-LB scores are so inconsistent for me that I am not so sure about the correct way to approach this anymore. The top public notebooks are doomed for sure though."
  },
  "source": "meta"
}