{
  "id": 200951,
  "title": "[Xgboost] - Running using TPU",
  "url": "/competitions/riiid-test-answer-prediction/discussion/200951",
  "author_name": "",
  "post_date": "2020-12-02T14:00:10.937882600Z",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hello kagglers, </p>\n<p>Does the GPU hist implementation of Xgboost use TPU if it is on ? <br>\nThank u </p>\n<p>Happy kaggling. </p>",
  "messages": [
    {
      "id": "1099641",
      "postDate": "12/02/2020 14:00:10",
      "content": "<p>Hello kagglers, </p>\n<p>Does the GPU hist implementation of Xgboost use TPU if it is on ? <br>\nThank u </p>\n<p>Happy kaggling. </p>",
      "rawMarkdown": "Hello kagglers, \n\nDoes the GPU hist implementation of Xgboost use TPU if it is on ? \nThank u \n\nHappy kaggling.",
      "votes": null
    },
    {
      "id": "1102813",
      "postDate": "12/05/2020 10:47:42",
      "content": "<p>Hi,</p>\n<p>I don't think that would work. There is nothing on XGBoost docs about it. Plus, TPU's are super fast doing matrix calculations (the kind of calculations neural networks perform), but it is not the way XGBoost works, so I doubt it is compatible</p>",
      "rawMarkdown": "Hi,\n\nI don't think that would work. There is nothing on XGBoost docs about it. Plus, TPU's are super fast doing matrix calculations (the kind of calculations neural networks perform), but it is not the way XGBoost works, so I doubt it is compatible",
      "votes": null
    },
    {
      "id": "1490930",
      "postDate": "08/26/2021 02:31:21",
      "content": "<p>Thanks a lot 👍 to both <a href=\"https://www.kaggle.com/benayas\" target=\"_blank\">@benayas</a> and <a href=\"https://www.kaggle.com/rpygamer\" target=\"_blank\">@rpygamer</a> </p>",
      "rawMarkdown": "Thanks a lot :+1: to both @benayas and @rpygamer",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1102813,
      "author_name": "benayas",
      "author_url": "",
      "post_date": "12/05/2020 10:47:42",
      "content": "<p>Hi,</p>\n<p>I don't think that would work. There is nothing on XGBoost docs about it. Plus, TPU's are super fast doing matrix calculations (the kind of calculations neural networks perform), but it is not the way XGBoost works, so I doubt it is compatible</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1490930,
      "author_name": "keagle",
      "author_url": "",
      "post_date": "08/26/2021 02:31:21",
      "content": "<p>Thanks a lot 👍 to both <a href=\"https://www.kaggle.com/benayas\" target=\"_blank\">@benayas</a> and <a href=\"https://www.kaggle.com/rpygamer\" target=\"_blank\">@rpygamer</a> </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1099641": "Hello kagglers, \n\nDoes the GPU hist implementation of Xgboost use TPU if it is on ? \nThank u \n\nHappy kaggling.",
    "1102813": "Hi,\n\nI don't think that would work. There is nothing on XGBoost docs about it. Plus, TPU's are super fast doing matrix calculations (the kind of calculations neural networks perform), but it is not the way XGBoost works, so I doubt it is compatible",
    "1490930": "Thanks a lot :+1: to both @benayas and @rpygamer"
  },
  "source": "meta"
}