{
  "id": 162624,
  "title": "Question about code requirements",
  "url": "/competitions/birdsong-recognition/discussion/162624",
  "author_name": "",
  "post_date": "2020-06-29T14:54:05.926621300Z",
  "votes": null,
  "comment_count": 2,
  "views": 0,
  "content": "<p>In the <a href=\"https://www.kaggle.com/c/birdsong-recognition/overview/code-requirements\">code requirements</a> section it states the GPU notebook runtime should be less than 2 hours. It does however also state TPU may be used, but only for training. Does the 2 hours only apply to notebooks which produce a prediction?</p>\n\n<p>Would I thus be allowed to use a TPU notebook which trains a model for 3 hours and use the obtained weights in a GPU notebook to make predictions within 2 hours?</p>",
  "messages": [
    {
      "id": "906821",
      "postDate": "06/29/2020 14:54:05",
      "content": "<p>In the <a href=\"https://www.kaggle.com/c/birdsong-recognition/overview/code-requirements\">code requirements</a> section it states the GPU notebook runtime should be less than 2 hours. It does however also state TPU may be used, but only for training. Does the 2 hours only apply to notebooks which produce a prediction?</p>\n\n<p>Would I thus be allowed to use a TPU notebook which trains a model for 3 hours and use the obtained weights in a GPU notebook to make predictions within 2 hours?</p>",
      "rawMarkdown": "In the [code requirements](https://www.kaggle.com/c/birdsong-recognition/overview/code-requirements) section it states the GPU notebook runtime should be less than 2 hours. It does however also state TPU may be used, but only for training. Does the 2 hours only apply to notebooks which produce a prediction?\n\nWould I thus be allowed to use a TPU notebook which trains a model for 3 hours and use the obtained weights in a GPU notebook to make predictions within 2 hours?",
      "votes": null
    },
    {
      "id": "906977",
      "postDate": "06/29/2020 16:36:15",
      "content": "<p>Correct, the restrictions only apply to the notebooks directly used for submissions.</p>",
      "rawMarkdown": "Correct, the restrictions only apply to the notebooks directly used for submissions.",
      "votes": null
    },
    {
      "id": "907024",
      "postDate": "06/29/2020 16:56:36",
      "content": "<p>Thanks for your reply!</p>",
      "rawMarkdown": "Thanks for your reply!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 906977,
      "author_name": "sohier",
      "author_url": "",
      "post_date": "06/29/2020 16:36:15",
      "content": "<p>Correct, the restrictions only apply to the notebooks directly used for submissions.</p>",
      "votes": null,
      "replies": [
        {
          "id": 907024,
          "author_name": "markwijkhuizen",
          "author_url": "",
          "post_date": "06/29/2020 16:56:36",
          "content": "<p>Thanks for your reply!</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "906821": "In the [code requirements](https://www.kaggle.com/c/birdsong-recognition/overview/code-requirements) section it states the GPU notebook runtime should be less than 2 hours. It does however also state TPU may be used, but only for training. Does the 2 hours only apply to notebooks which produce a prediction?\n\nWould I thus be allowed to use a TPU notebook which trains a model for 3 hours and use the obtained weights in a GPU notebook to make predictions within 2 hours?",
    "906977": "Correct, the restrictions only apply to the notebooks directly used for submissions.",
    "907024": "Thanks for your reply!"
  },
  "source": "meta"
}