{
  "id": 337488,
  "title": "Example request: LightGBM w/ Early Stopping",
  "url": "/competitions/amex-default-prediction/discussion/337488",
  "author_name": "",
  "post_date": "2022-07-16T10:25:16.044637600Z",
  "votes": 3,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Hello everyone, LGBM seems to be an important piece in this puzzle, but getting early stopping to work has been a struggle so far. There has been a lot of issues logged within the LightGBM GitHub repo:<br>\ne.g.: <a href=\"https://github.com/microsoft/LightGBM/issues/5354\" target=\"_blank\">https://github.com/microsoft/LightGBM/issues/5354</a></p>\n<p>Does anyone have a working flow of early stopping? <br>\nPS: I am using lightgbm 3.3.2</p>\n<p>Thanks in advance :) </p>",
  "messages": [
    {
      "id": "1857725",
      "postDate": "07/16/2022 10:25:16",
      "content": "<p>Hello everyone, LGBM seems to be an important piece in this puzzle, but getting early stopping to work has been a struggle so far. There has been a lot of issues logged within the LightGBM GitHub repo:<br>\ne.g.: <a href=\"https://github.com/microsoft/LightGBM/issues/5354\" target=\"_blank\">https://github.com/microsoft/LightGBM/issues/5354</a></p>\n<p>Does anyone have a working flow of early stopping? <br>\nPS: I am using lightgbm 3.3.2</p>\n<p>Thanks in advance :) </p>",
      "rawMarkdown": "Hello everyone, LGBM seems to be an important piece in this puzzle, but getting early stopping to work has been a struggle so far. There has been a lot of issues logged within the LightGBM GitHub repo:\ne.g.: https://github.com/microsoft/LightGBM/issues/5354\n\nDoes anyone have a working flow of early stopping? \nPS: I am using lightgbm 3.3.2\n\nThanks in advance :)",
      "votes": null
    },
    {
      "id": "1857903",
      "postDate": "07/16/2022 13:36:33",
      "content": "<p>DART is a promising boosting method for this competition. However, with DART you cannot use early stopping. check out <a href=\"https://www.kaggle.com/competitions/amex-default-prediction/discussion/332575\" target=\"_blank\">DART, LGBM, Saving Best Models [Callbacks Code Snippet]</a> for more information!</p>",
      "rawMarkdown": "DART is a promising boosting method for this competition. However, with DART you cannot use early stopping. check out [DART, LGBM, Saving Best Models [Callbacks Code Snippet]](https://www.kaggle.com/competitions/amex-default-prediction/discussion/332575) for more information!",
      "votes": null
    },
    {
      "id": "1857926",
      "postDate": "07/16/2022 14:12:53",
      "content": "<p>Thank you I checked it out and its exactly what i needed to learn :) </p>",
      "rawMarkdown": "Thank you I checked it out and its exactly what i needed to learn :)",
      "votes": null
    },
    {
      "id": "1857944",
      "postDate": "07/16/2022 14:24:40",
      "content": "<p>good luck <a href=\"https://www.kaggle.com/leo1988\" target=\"_blank\">@leo1988</a> !</p>",
      "rawMarkdown": "good luck @leo1988 !",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1857903,
      "author_name": "mohammadrahmati",
      "author_url": "",
      "post_date": "07/16/2022 13:36:33",
      "content": "<p>DART is a promising boosting method for this competition. However, with DART you cannot use early stopping. check out <a href=\"https://www.kaggle.com/competitions/amex-default-prediction/discussion/332575\" target=\"_blank\">DART, LGBM, Saving Best Models [Callbacks Code Snippet]</a> for more information!</p>",
      "votes": null,
      "replies": [
        {
          "id": 1857926,
          "author_name": "leo1988",
          "author_url": "",
          "post_date": "07/16/2022 14:12:53",
          "content": "<p>Thank you I checked it out and its exactly what i needed to learn :) </p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1857944,
          "author_name": "mohammadrahmati",
          "author_url": "",
          "post_date": "07/16/2022 14:24:40",
          "content": "<p>good luck <a href=\"https://www.kaggle.com/leo1988\" target=\"_blank\">@leo1988</a> !</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1857725": "Hello everyone, LGBM seems to be an important piece in this puzzle, but getting early stopping to work has been a struggle so far. There has been a lot of issues logged within the LightGBM GitHub repo:\ne.g.: https://github.com/microsoft/LightGBM/issues/5354\n\nDoes anyone have a working flow of early stopping? \nPS: I am using lightgbm 3.3.2\n\nThanks in advance :)",
    "1857903": "DART is a promising boosting method for this competition. However, with DART you cannot use early stopping. check out [DART, LGBM, Saving Best Models [Callbacks Code Snippet]](https://www.kaggle.com/competitions/amex-default-prediction/discussion/332575) for more information!",
    "1857926": "Thank you I checked it out and its exactly what i needed to learn :)",
    "1857944": "good luck @leo1988 !"
  },
  "source": "meta"
}