{
  "id": 55935,
  "title": "Unable to use trackml library with gpu",
  "url": "/competitions/trackml-particle-identification/discussion/55935",
  "author_name": "Siddharth Yadav",
  "post_date": "2018-05-03T09:47:29.115000",
  "votes": 1,
  "comment_count": 3,
  "views": 0,
  "content": "<p>For some reason I am unable to use the GPU provided by kaggle with the trackml library. Either one works alone but none at the same time. What might be the issue?</p>",
  "messages": [
    {
      "id": 322617,
      "postDate": "2018-05-03T09:47:29.117Z",
      "content": "<p>For some reason I am unable to use the GPU provided by kaggle with the trackml library. Either one works alone but none at the same time. What might be the issue?</p>",
      "rawMarkdown": "For some reason I am unable to use the GPU provided by kaggle with the trackml library. Either one works alone but none at the same time. What might be the issue?",
      "votes": 1
    },
    {
      "id": 327298,
      "postDate": "2018-05-11T08:10:12.683Z",
      "content": "<p>It seems that enabling the GPU suppport on Kaggle kernels prohibits the use of a custom docker image. That means you can only either install a custom package or use the GPU. You should still be able to read the files directly with Pandas.</p>",
      "rawMarkdown": "It seems that enabling the GPU suppport on Kaggle kernels prohibits the use of a custom docker image. That means you can only either install a custom package or use the GPU. You should still be able to read the files directly with Pandas.",
      "votes": 1,
      "replies": [
        {
          "id": 357984,
          "postDate": "2018-07-17T09:22:35.790Z",
          "content": "<p>I'll let you know if I find something out.</p>",
          "rawMarkdown": "I'll let you know if I find something out."
        }
      ]
    },
    {
      "id": 357902,
      "postDate": "2018-07-17T05:30:58.060Z",
      "rawMarkdown": "",
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 327298,
      "author_name": "Moritz Kiehn",
      "author_url": "",
      "post_date": "2018-05-11T08:10:12.683000",
      "content": "<p>It seems that enabling the GPU suppport on Kaggle kernels prohibits the use of a custom docker image. That means you can only either install a custom package or use the GPU. You should still be able to read the files directly with Pandas.</p>",
      "votes": 1,
      "replies": [
        {
          "id": 357984,
          "author_name": "Moritz Kiehn",
          "author_url": "",
          "post_date": "2018-07-17T09:22:35.790000",
          "content": "<p>I'll let you know if I find something out.</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 357902,
      "author_name": "",
      "author_url": "",
      "post_date": "2018-07-17T05:30:58.060000",
      "content": "",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "322617": "For some reason I am unable to use the GPU provided by kaggle with the trackml library. Either one works alone but none at the same time. What might be the issue?",
    "327298": "It seems that enabling the GPU suppport on Kaggle kernels prohibits the use of a custom docker image. That means you can only either install a custom package or use the GPU. You should still be able to read the files directly with Pandas.",
    "357902": ""
  }
}