{
  "id": 358987,
  "title": " 📌 Feature Importances by using featimp 🔥🔥",
  "url": "/competitions/tabular-playground-series-oct-2022/discussion/358987",
  "author_name": "",
  "post_date": "2022-10-10T10:28:05.395517900Z",
  "votes": 9,
  "comment_count": 6,
  "views": 0,
  "content": "<p><code>pip install featimp</code></p>\n<p><code>fi_df = get_feature_importances(data=sample_df, \n                                                        num_features=num_features,\n                                                         target='team_A_scoring_within_10sec', \n                                                         task='clf_binary', \n                                                         ml_model_name='LGBM', \n                                                         method='all')</code></p>\n<p><code>display_feature_importances(data=fi_df)</code><br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F3381067%2F0e565fc0644ddf09eccfd78013a133cc%2Ffi.png?generation=1665397329991726&amp;alt=media\" alt=\"\"></p>",
  "messages": [
    {
      "id": "1980701",
      "postDate": "10/10/2022 10:28:05",
      "content": "<p><code>pip install featimp</code></p>\n<p><code>fi_df = get_feature_importances(data=sample_df, \n                                                        num_features=num_features,\n                                                         target='team_A_scoring_within_10sec', \n                                                         task='clf_binary', \n                                                         ml_model_name='LGBM', \n                                                         method='all')</code></p>\n<p><code>display_feature_importances(data=fi_df)</code><br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F3381067%2F0e565fc0644ddf09eccfd78013a133cc%2Ffi.png?generation=1665397329991726&amp;alt=media\" alt=\"\"></p>",
      "rawMarkdown": "`pip install featimp`\n\n```fi_df = get_feature_importances(data=sample_df, \n                                                        num_features=num_features,\n                                                         target='team_A_scoring_within_10sec', \n                                                         task='clf_binary', \n                                                         ml_model_name='LGBM', \n                                                         method='all')```\n\n`display_feature_importances(data=fi_df)`\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F3381067%2F0e565fc0644ddf09eccfd78013a133cc%2Ffi.png?generation=1665397329991726&alt=media)",
      "votes": null
    },
    {
      "id": "1983052",
      "postDate": "10/11/2022 18:54:44",
      "content": "<p>Thanks for sharing, I didn’t know this package. I will try to use it in my notebooks, looks very useful.</p>",
      "rawMarkdown": "Thanks for sharing, I didn’t know this package. I will try to use it in my notebooks, looks very useful.",
      "votes": null
    },
    {
      "id": "1983639",
      "postDate": "10/12/2022 06:35:50",
      "content": "<p>There are lots of feature importances techniques. You can calculate all feature importances and rank them by using <a href=\"https://github.com/Hasan-Basri-Akcay/featimp\" target=\"_blank\">featimp</a> 👍.</p>",
      "rawMarkdown": "There are lots of feature importances techniques. You can calculate all feature importances and rank them by using [featimp](https://github.com/Hasan-Basri-Akcay/featimp) 👍.",
      "votes": null
    },
    {
      "id": "1986117",
      "postDate": "10/13/2022 20:07:10",
      "content": "<p>Thanks for sharing <a href=\"https://www.kaggle.com/hasanbasriakcay\" target=\"_blank\">@hasanbasriakcay</a> 👍</p>",
      "rawMarkdown": "Thanks for sharing @hasanbasriakcay 👍",
      "votes": null
    },
    {
      "id": "1987503",
      "postDate": "10/14/2022 18:09:35",
      "content": "<p>How does it find the important features?</p>",
      "rawMarkdown": "How does it find the important features?",
      "votes": null
    },
    {
      "id": "1989363",
      "postDate": "10/15/2022 21:25:30",
      "content": "<p>Surely will use that as a reference! Great Work, keep it up</p>",
      "rawMarkdown": "Surely will use that as a reference! Great Work, keep it up",
      "votes": null
    },
    {
      "id": "1990389",
      "postDate": "10/16/2022 14:04:49",
      "content": "<p>\"There are a lot of feature importance techniques and each technique calculates different importance. Some of them are suitable for numerical to numerical importance, some of them are ideal for categorical to the numerical significance and some of them are suitable for categorical to categorical importance. featimp automatically calculates feature importances and ranks them for you.\" </p>\n<p>source: <a href=\"https://github.com/Hasan-Basri-Akcay/featimp\" target=\"_blank\">github</a></p>",
      "rawMarkdown": "\"There are a lot of feature importance techniques and each technique calculates different importance. Some of them are suitable for numerical to numerical importance, some of them are ideal for categorical to the numerical significance and some of them are suitable for categorical to categorical importance. featimp automatically calculates feature importances and ranks them for you.\" \n\nsource: [github](https://github.com/Hasan-Basri-Akcay/featimp)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1983052,
      "author_name": "mrgabrielblins",
      "author_url": "",
      "post_date": "10/11/2022 18:54:44",
      "content": "<p>Thanks for sharing, I didn’t know this package. I will try to use it in my notebooks, looks very useful.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1983639,
          "author_name": "hasanbasriakcay",
          "author_url": "",
          "post_date": "10/12/2022 06:35:50",
          "content": "<p>There are lots of feature importances techniques. You can calculate all feature importances and rank them by using <a href=\"https://github.com/Hasan-Basri-Akcay/featimp\" target=\"_blank\">featimp</a> 👍.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1986117,
      "author_name": "nadiate",
      "author_url": "",
      "post_date": "10/13/2022 20:07:10",
      "content": "<p>Thanks for sharing <a href=\"https://www.kaggle.com/hasanbasriakcay\" target=\"_blank\">@hasanbasriakcay</a> 👍</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1987503,
      "author_name": "himanshuashp77",
      "author_url": "",
      "post_date": "10/14/2022 18:09:35",
      "content": "<p>How does it find the important features?</p>",
      "votes": null,
      "replies": [
        {
          "id": 1990389,
          "author_name": "hasanbasriakcay",
          "author_url": "",
          "post_date": "10/16/2022 14:04:49",
          "content": "<p>\"There are a lot of feature importance techniques and each technique calculates different importance. Some of them are suitable for numerical to numerical importance, some of them are ideal for categorical to the numerical significance and some of them are suitable for categorical to categorical importance. featimp automatically calculates feature importances and ranks them for you.\" </p>\n<p>source: <a href=\"https://github.com/Hasan-Basri-Akcay/featimp\" target=\"_blank\">github</a></p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1989363,
      "author_name": "ahmedhelmey",
      "author_url": "",
      "post_date": "10/15/2022 21:25:30",
      "content": "<p>Surely will use that as a reference! Great Work, keep it up</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1980701": "`pip install featimp`\n\n```fi_df = get_feature_importances(data=sample_df, \n                                                        num_features=num_features,\n                                                         target='team_A_scoring_within_10sec', \n                                                         task='clf_binary', \n                                                         ml_model_name='LGBM', \n                                                         method='all')```\n\n`display_feature_importances(data=fi_df)`\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F3381067%2F0e565fc0644ddf09eccfd78013a133cc%2Ffi.png?generation=1665397329991726&alt=media)",
    "1983052": "Thanks for sharing, I didn’t know this package. I will try to use it in my notebooks, looks very useful.",
    "1983639": "There are lots of feature importances techniques. You can calculate all feature importances and rank them by using [featimp](https://github.com/Hasan-Basri-Akcay/featimp) 👍.",
    "1986117": "Thanks for sharing @hasanbasriakcay 👍",
    "1987503": "How does it find the important features?",
    "1989363": "Surely will use that as a reference! Great Work, keep it up",
    "1990389": "\"There are a lot of feature importance techniques and each technique calculates different importance. Some of them are suitable for numerical to numerical importance, some of them are ideal for categorical to the numerical significance and some of them are suitable for categorical to categorical importance. featimp automatically calculates feature importances and ranks them for you.\" \n\nsource: [github](https://github.com/Hasan-Basri-Akcay/featimp)"
  },
  "source": "meta"
}