{
  "id": 276257,
  "title": "Is it allowed to separate the training and Submission notebooks?",
  "url": "/competitions/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/276257",
  "author_name": "",
  "post_date": "2021-10-03T21:03:20.685010Z",
  "votes": 2,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Hello everyone, I have a simple question. </p>\n<p>Currently I am training my model in a notebook, then save the model and load it in a new notebook to predict the output for test dataset. Because every notebook has to run in less then 9 hours. Is it allowed to proceed like this for the final score? </p>",
  "messages": [
    {
      "id": "1533280",
      "postDate": "10/03/2021 21:03:20",
      "content": "<p>Hello everyone, I have a simple question. </p>\n<p>Currently I am training my model in a notebook, then save the model and load it in a new notebook to predict the output for test dataset. Because every notebook has to run in less then 9 hours. Is it allowed to proceed like this for the final score? </p>",
      "rawMarkdown": "Hello everyone, I have a simple question. \n\nCurrently I am training my model in a notebook, then save the model and load it in a new notebook to predict the output for test dataset. Because every notebook has to run in less then 9 hours. Is it allowed to proceed like this for the final score?",
      "votes": null
    },
    {
      "id": "1533343",
      "postDate": "10/03/2021 22:38:42",
      "content": "<p>Yes, we can have one notebook for training and another for inference.</p>",
      "rawMarkdown": "Yes, we can have one notebook for training and another for inference.",
      "votes": null
    },
    {
      "id": "1533352",
      "postDate": "10/03/2021 23:02:30",
      "content": "<p>Thank you!</p>",
      "rawMarkdown": "Thank you!",
      "votes": null
    },
    {
      "id": "1541810",
      "postDate": "10/11/2021 21:13:52",
      "content": "<p>and even more training notebooks and one final notebook for ensemble</p>",
      "rawMarkdown": "and even more training notebooks and one final notebook for ensemble",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1533343,
      "author_name": "arunamenon",
      "author_url": "",
      "post_date": "10/03/2021 22:38:42",
      "content": "<p>Yes, we can have one notebook for training and another for inference.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1533352,
          "author_name": "niloofarrahmani",
          "author_url": "",
          "post_date": "10/03/2021 23:02:30",
          "content": "<p>Thank you!</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1541810,
          "author_name": "tegzes",
          "author_url": "",
          "post_date": "10/11/2021 21:13:52",
          "content": "<p>and even more training notebooks and one final notebook for ensemble</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1533280": "Hello everyone, I have a simple question. \n\nCurrently I am training my model in a notebook, then save the model and load it in a new notebook to predict the output for test dataset. Because every notebook has to run in less then 9 hours. Is it allowed to proceed like this for the final score?",
    "1533343": "Yes, we can have one notebook for training and another for inference.",
    "1533352": "Thank you!",
    "1541810": "and even more training notebooks and one final notebook for ensemble"
  },
  "source": "meta"
}