{
  "id": 502078,
  "title": "👀 Will shakeup happen? 🌀",
  "url": "/competitions/home-credit-credit-risk-model-stability/discussion/502078",
  "author_name": "",
  "post_date": "2024-05-12T00:30:57.415396500Z",
  "votes": 4,
  "comment_count": 4,
  "views": 0,
  "content": "<p>It's almost time.</p>\n<p>There is a lot of talk about metric hacking right now, but is it also effective for private LBs? Isn't that so? The selection of submissions is important…</p>\n<p>I'm curious about the results of this unique competition.</p>",
  "messages": [
    {
      "id": "2807905",
      "postDate": "05/12/2024 00:30:57",
      "content": "<p>It's almost time.</p>\n<p>There is a lot of talk about metric hacking right now, but is it also effective for private LBs? Isn't that so? The selection of submissions is important…</p>\n<p>I'm curious about the results of this unique competition.</p>",
      "rawMarkdown": "It's almost time.\n\nThere is a lot of talk about metric hacking right now, but is it also effective for private LBs? Isn't that so? The selection of submissions is important...\n\nI'm curious about the results of this unique competition.",
      "votes": null
    },
    {
      "id": "2808069",
      "postDate": "05/12/2024 04:28:26",
      "content": "<p>At the end of the competition, each person is allowed to submit two notebooks. I think they can submit a hacking version and another normal training model version. I don't think this will have any impact on them.</p>",
      "rawMarkdown": "At the end of the competition, each person is allowed to submit two notebooks. I think they can submit a hacking version and another normal training model version. I don't think this will have any impact on them.",
      "votes": null
    },
    {
      "id": "2808759",
      "postDate": "05/12/2024 11:44:47",
      "content": "<p>Maybe some shakeup will happen in this competition based on the model, but perhaps  due to the hacking element too <a href=\"https://www.kaggle.com/archerpan\" target=\"_blank\">@archerpan</a> <a href=\"https://www.kaggle.com/mitsuyasuhoshino\" target=\"_blank\">@mitsuyasuhoshino</a> </p>",
      "rawMarkdown": "Maybe some shakeup will happen in this competition based on the model, but perhaps  due to the hacking element too @archerpan @mitsuyasuhoshino",
      "votes": null
    },
    {
      "id": "2810222",
      "postDate": "05/13/2024 07:14:25",
      "content": "<p><a href=\"https://www.kaggle.com/ravi20076\" target=\"_blank\">@ravi20076</a> I agree that shakeup occur to a greater or lesser degree.<br>\nI'm very interested in <a href=\"https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/discussion/501744\" target=\"_blank\">this post</a> because I don't think host haven't thought about this at all.</p>",
      "rawMarkdown": "ravi20076 I agree that shakeup occur to a greater or lesser degree.\nI'm very interested in [this post](https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/discussion/501744) because I don't think host haven't thought about this at all.",
      "votes": null
    },
    {
      "id": "2839142",
      "postDate": "05/27/2024 11:58:35",
      "content": "<p>Now it's time!! Choose your two submissions wisely.<br>\nMy unfounded prediction is that a fairly large shake will occur (Submissions that involve metric hacking or unnecessary post-processing will fail).<br>\nIt suggests the \"terrible\" silence of recent hosts.</p>",
      "rawMarkdown": "Now it's time!! Choose your two submissions wisely.\nMy unfounded prediction is that a fairly large shake will occur (Submissions that involve metric hacking or unnecessary post-processing will fail).\nIt suggests the \"terrible\" silence of recent hosts.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2808069,
      "author_name": "archerpan",
      "author_url": "",
      "post_date": "05/12/2024 04:28:26",
      "content": "<p>At the end of the competition, each person is allowed to submit two notebooks. I think they can submit a hacking version and another normal training model version. I don't think this will have any impact on them.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2808759,
          "author_name": "ravi20076",
          "author_url": "",
          "post_date": "05/12/2024 11:44:47",
          "content": "<p>Maybe some shakeup will happen in this competition based on the model, but perhaps  due to the hacking element too <a href=\"https://www.kaggle.com/archerpan\" target=\"_blank\">@archerpan</a> <a href=\"https://www.kaggle.com/mitsuyasuhoshino\" target=\"_blank\">@mitsuyasuhoshino</a> </p>",
          "votes": null,
          "replies": [
            {
              "id": 2810222,
              "author_name": "mitsuyasuhoshino",
              "author_url": "",
              "post_date": "05/13/2024 07:14:25",
              "content": "<p><a href=\"https://www.kaggle.com/ravi20076\" target=\"_blank\">@ravi20076</a> I agree that shakeup occur to a greater or lesser degree.<br>\nI'm very interested in <a href=\"https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/discussion/501744\" target=\"_blank\">this post</a> because I don't think host haven't thought about this at all.</p>",
              "votes": null,
              "replies": []
            }
          ]
        }
      ]
    },
    {
      "id": 2839142,
      "author_name": "mitsuyasuhoshino",
      "author_url": "",
      "post_date": "05/27/2024 11:58:35",
      "content": "<p>Now it's time!! Choose your two submissions wisely.<br>\nMy unfounded prediction is that a fairly large shake will occur (Submissions that involve metric hacking or unnecessary post-processing will fail).<br>\nIt suggests the \"terrible\" silence of recent hosts.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2807905": "It's almost time.\n\nThere is a lot of talk about metric hacking right now, but is it also effective for private LBs? Isn't that so? The selection of submissions is important...\n\nI'm curious about the results of this unique competition.",
    "2808069": "At the end of the competition, each person is allowed to submit two notebooks. I think they can submit a hacking version and another normal training model version. I don't think this will have any impact on them.",
    "2808759": "Maybe some shakeup will happen in this competition based on the model, but perhaps  due to the hacking element too @archerpan @mitsuyasuhoshino",
    "2810222": "ravi20076 I agree that shakeup occur to a greater or lesser degree.\nI'm very interested in [this post](https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/discussion/501744) because I don't think host haven't thought about this at all.",
    "2839142": "Now it's time!! Choose your two submissions wisely.\nMy unfounded prediction is that a fairly large shake will occur (Submissions that involve metric hacking or unnecessary post-processing will fail).\nIt suggests the \"terrible\" silence of recent hosts."
  },
  "source": "meta"
}