{
  "id": 425919,
  "title": "Question on 'submission.tsv'",
  "url": "/competitions/cafa-5-protein-function-prediction/discussion/425919",
  "author_name": "",
  "post_date": "2023-07-20T23:30:54.933660100Z",
  "votes": -1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Am I doing couple of experiments offline (as they are not optimized yet on memory management, kaggle seems to run out of memory during execution!!!)</p>\n<p>My question is </p>\n<p>can I create 'submission.tsv' offline, create a dataset on kaggle and submit it to competition thru an 'inference' notebook ?</p>",
  "messages": [
    {
      "id": "2352409",
      "postDate": "07/20/2023 23:30:54",
      "content": "<p>Am I doing couple of experiments offline (as they are not optimized yet on memory management, kaggle seems to run out of memory during execution!!!)</p>\n<p>My question is </p>\n<p>can I create 'submission.tsv' offline, create a dataset on kaggle and submit it to competition thru an 'inference' notebook ?</p>",
      "rawMarkdown": "Am I doing couple of experiments offline (as they are not optimized yet on memory management, kaggle seems to run out of memory during execution!!!)\n\nMy question is \n\ncan I create 'submission.tsv' offline, create a dataset on kaggle and submit it to competition thru an 'inference' notebook ?",
      "votes": null
    },
    {
      "id": "2353353",
      "postDate": "07/21/2023 15:46:42",
      "content": "<p>Yes you can.</p>",
      "rawMarkdown": "Yes you can.",
      "votes": null
    },
    {
      "id": "2353541",
      "postDate": "07/21/2023 18:29:28",
      "content": "<p>Yes, you can do that. It's easy and convenient to contrive a notebook to do so. One such approach is that the submission.tsv file is read in from a dataset and then written out again as the notebook's output. </p>\n<p>However, you can't upload the predictions for scoring directly outside the context of a notebook. In a very pedantic and formal sense, this is a code competition.</p>",
      "rawMarkdown": "Yes, you can do that. It's easy and convenient to contrive a notebook to do so. One such approach is that the submission.tsv file is read in from a dataset and then written out again as the notebook's output. \n\nHowever, you can't upload the predictions for scoring directly outside the context of a notebook. In a very pedantic and formal sense, this is a code competition.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2353353,
      "author_name": "simonveitner",
      "author_url": "",
      "post_date": "07/21/2023 15:46:42",
      "content": "<p>Yes you can.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2353541,
      "author_name": "jbomitchell",
      "author_url": "",
      "post_date": "07/21/2023 18:29:28",
      "content": "<p>Yes, you can do that. It's easy and convenient to contrive a notebook to do so. One such approach is that the submission.tsv file is read in from a dataset and then written out again as the notebook's output. </p>\n<p>However, you can't upload the predictions for scoring directly outside the context of a notebook. In a very pedantic and formal sense, this is a code competition.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2352409": "Am I doing couple of experiments offline (as they are not optimized yet on memory management, kaggle seems to run out of memory during execution!!!)\n\nMy question is \n\ncan I create 'submission.tsv' offline, create a dataset on kaggle and submit it to competition thru an 'inference' notebook ?",
    "2353353": "Yes you can.",
    "2353541": "Yes, you can do that. It's easy and convenient to contrive a notebook to do so. One such approach is that the submission.tsv file is read in from a dataset and then written out again as the notebook's output. \n\nHowever, you can't upload the predictions for scoring directly outside the context of a notebook. In a very pedantic and formal sense, this is a code competition."
  },
  "source": "meta"
}