{
  "id": 90385,
  "title": "Automated parameter tuning in a single line of code",
  "url": "/competitions/LANL-Earthquake-Prediction/discussion/90385",
  "author_name": "",
  "post_date": "2019-04-23T11:30:23.468292100Z",
  "votes": 8,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hi everyone, I've created a streamlined Hyperopt pipeline you can use for automated hyperparameter optimisation:</p>\n\n<p><a href=\"https://www.kaggle.com/bigironsphere/parameter-tuning-in-one-function-with-hyperopt\">https://www.kaggle.com/bigironsphere/parameter-tuning-in-one-function-with-hyperopt</a></p>\n\n<p>It'll tune the parameters for your dataset with either LightGBM, XGBoost or CatBoost in a single line of code. I know many people, particularly beginners, can find parameter tuning overwhelming. Hopefully this will be a useful tool that'll save you some time. Hyperopt is powerful, but not exactly 'user-friendly', as I found when I was working with it! This takes out most of the hassle for you.</p>\n\n<p>I'm interested to know what any of the old pros on here think - I've tried to include a broad parameter range for each model, but are there any omissions you think should be rectified?</p>\n\n<p>Please let me know if this tool helped you increase your LB score!</p>",
  "messages": [
    {
      "id": "521755",
      "postDate": "04/23/2019 11:30:23",
      "content": "<p>Hi everyone, I've created a streamlined Hyperopt pipeline you can use for automated hyperparameter optimisation:</p>\n\n<p><a href=\"https://www.kaggle.com/bigironsphere/parameter-tuning-in-one-function-with-hyperopt\">https://www.kaggle.com/bigironsphere/parameter-tuning-in-one-function-with-hyperopt</a></p>\n\n<p>It'll tune the parameters for your dataset with either LightGBM, XGBoost or CatBoost in a single line of code. I know many people, particularly beginners, can find parameter tuning overwhelming. Hopefully this will be a useful tool that'll save you some time. Hyperopt is powerful, but not exactly 'user-friendly', as I found when I was working with it! This takes out most of the hassle for you.</p>\n\n<p>I'm interested to know what any of the old pros on here think - I've tried to include a broad parameter range for each model, but are there any omissions you think should be rectified?</p>\n\n<p>Please let me know if this tool helped you increase your LB score!</p>",
      "rawMarkdown": "Hi everyone, I've created a streamlined Hyperopt pipeline you can use for automated hyperparameter optimisation:\n\nhttps://www.kaggle.com/bigironsphere/parameter-tuning-in-one-function-with-hyperopt\n\nIt'll tune the parameters for your dataset with either LightGBM, XGBoost or CatBoost in a single line of code. I know many people, particularly beginners, can find parameter tuning overwhelming. Hopefully this will be a useful tool that'll save you some time. Hyperopt is powerful, but not exactly 'user-friendly', as I found when I was working with it! This takes out most of the hassle for you.\n\nI'm interested to know what any of the old pros on here think - I've tried to include a broad parameter range for each model, but are there any omissions you think should be rectified?\n\nPlease let me know if this tool helped you increase your LB score!",
      "votes": null
    },
    {
      "id": "521799",
      "postDate": "04/23/2019 12:48:03",
      "content": "<p>Nicely done <a href=\"/bigironsphere\">@bigironsphere</a> !</p>",
      "rawMarkdown": "Nicely done @bigironsphere !",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 521799,
      "author_name": "vjcalling",
      "author_url": "",
      "post_date": "04/23/2019 12:48:03",
      "content": "<p>Nicely done <a href=\"/bigironsphere\">@bigironsphere</a> !</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "521755": "Hi everyone, I've created a streamlined Hyperopt pipeline you can use for automated hyperparameter optimisation:\n\nhttps://www.kaggle.com/bigironsphere/parameter-tuning-in-one-function-with-hyperopt\n\nIt'll tune the parameters for your dataset with either LightGBM, XGBoost or CatBoost in a single line of code. I know many people, particularly beginners, can find parameter tuning overwhelming. Hopefully this will be a useful tool that'll save you some time. Hyperopt is powerful, but not exactly 'user-friendly', as I found when I was working with it! This takes out most of the hassle for you.\n\nI'm interested to know what any of the old pros on here think - I've tried to include a broad parameter range for each model, but are there any omissions you think should be rectified?\n\nPlease let me know if this tool helped you increase your LB score!",
    "521799": "Nicely done @bigironsphere !"
  },
  "source": "meta"
}