{
  "id": 327984,
  "title": "Evaluation Metric in Datatable (6x speedup over Pandas) 🔥",
  "url": "/competitions/amex-default-prediction/discussion/327984",
  "author_name": "",
  "post_date": "2022-05-30T10:49:02.106248200Z",
  "votes": 13,
  "comment_count": 2,
  "views": 0,
  "content": "<p>The implementation of the evaluation metric using Python Datatable: <a href=\"https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Datatable-Implementation\" target=\"_blank\">https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Datatable-Implementation</a></p>\n<p>It is about 6x faster than the official pandas implementation as compared <a href=\"https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Speed-Comparison\" target=\"_blank\">here</a> with some other implementations too.</p>\n<p>If you find better ways to optimize the datatable version, please share here and I'll be glad to update the notebook!</p>",
  "messages": [
    {
      "id": "1805595",
      "postDate": "05/30/2022 10:49:02",
      "content": "<p>The implementation of the evaluation metric using Python Datatable: <a href=\"https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Datatable-Implementation\" target=\"_blank\">https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Datatable-Implementation</a></p>\n<p>It is about 6x faster than the official pandas implementation as compared <a href=\"https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Speed-Comparison\" target=\"_blank\">here</a> with some other implementations too.</p>\n<p>If you find better ways to optimize the datatable version, please share here and I'll be glad to update the notebook!</p>",
      "rawMarkdown": "The implementation of the evaluation metric using Python Datatable: [https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Datatable-Implementation](https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Datatable-Implementation)\n\nIt is about 6x faster than the official pandas implementation as compared [here](https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Speed-Comparison) with some other implementations too.\n\nIf you find better ways to optimize the datatable version, please share here and I'll be glad to update the notebook!",
      "votes": null
    },
    {
      "id": "1805815",
      "postDate": "05/30/2022 14:23:16",
      "content": "<p><a href=\"https://www.kaggle.com/competitions/amex-default-prediction/discussion/328020\" target=\"_blank\">https://www.kaggle.com/competitions/amex-default-prediction/discussion/328020</a></p>\n<p>I post another numpy implementation here~</p>",
      "rawMarkdown": "https://www.kaggle.com/competitions/amex-default-prediction/discussion/328020\n\nI post another numpy implementation here~",
      "votes": null
    },
    {
      "id": "1806488",
      "postDate": "05/31/2022 08:25:26",
      "content": "<p>Thanks a lot! I've added it to the notebook linking your post and profile 👍</p>",
      "rawMarkdown": "Thanks a lot! I've added it to the notebook linking your post and profile 👍",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1805815,
      "author_name": "yunchonggan",
      "author_url": "",
      "post_date": "05/30/2022 14:23:16",
      "content": "<p><a href=\"https://www.kaggle.com/competitions/amex-default-prediction/discussion/328020\" target=\"_blank\">https://www.kaggle.com/competitions/amex-default-prediction/discussion/328020</a></p>\n<p>I post another numpy implementation here~</p>",
      "votes": null,
      "replies": [
        {
          "id": 1806488,
          "author_name": "rohanrao",
          "author_url": "",
          "post_date": "05/31/2022 08:25:26",
          "content": "<p>Thanks a lot! I've added it to the notebook linking your post and profile 👍</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1805595": "The implementation of the evaluation metric using Python Datatable: [https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Datatable-Implementation](https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Datatable-Implementation)\n\nIt is about 6x faster than the official pandas implementation as compared [here](https://www.kaggle.com/code/rohanrao/amex-competition-metric-implementations/notebook#Speed-Comparison) with some other implementations too.\n\nIf you find better ways to optimize the datatable version, please share here and I'll be glad to update the notebook!",
    "1805815": "https://www.kaggle.com/competitions/amex-default-prediction/discussion/328020\n\nI post another numpy implementation here~",
    "1806488": "Thanks a lot! I've added it to the notebook linking your post and profile 👍"
  },
  "source": "meta"
}