{
  "id": 191292,
  "title": "Are we allowed to use pre-processed train data in our submissions?",
  "url": "/competitions/riiid-test-answer-prediction/discussion/191292",
  "author_name": "",
  "post_date": "2020-10-15T17:44:49.074361800Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I just want to pre-process the train data and then store it in a Kaggle dataset so that I don't have to waste memory on re-doing it for every run. The competition rules don't really say anything about this so I was just wondering if anybody knows. </p>\n<p>Also, if allowed, should this dataset be public or private?</p>",
  "messages": [
    {
      "id": "1050764",
      "postDate": "10/15/2020 17:44:49",
      "content": "<p>I just want to pre-process the train data and then store it in a Kaggle dataset so that I don't have to waste memory on re-doing it for every run. The competition rules don't really say anything about this so I was just wondering if anybody knows. </p>\n<p>Also, if allowed, should this dataset be public or private?</p>",
      "rawMarkdown": "I just want to pre-process the train data and then store it in a Kaggle dataset so that I don't have to waste memory on re-doing it for every run. The competition rules don't really say anything about this so I was just wondering if anybody knows. \n\nAlso, if allowed, should this dataset be public or private?",
      "votes": null
    },
    {
      "id": "1050849",
      "postDate": "10/15/2020 19:39:13",
      "content": "<p>Yes you can use preprocessed data, and it can be private. I believe that as long as the dataset you're deriving your data from is publicly available (e.g. the raw training data and supplementary tables from the competition data), you're fine to make it private. </p>\n<p>I asked the same question recently -- <a href=\"https://www.kaggle.com/c/riiid-test-answer-prediction/discussion/189555\" target=\"_blank\">https://www.kaggle.com/c/riiid-test-answer-prediction/discussion/189555</a></p>",
      "rawMarkdown": "Yes you can use preprocessed data, and it can be private. I believe that as long as the dataset you're deriving your data from is publicly available (e.g. the raw training data and supplementary tables from the competition data), you're fine to make it private. \n\nI asked the same question recently -- https://www.kaggle.com/c/riiid-test-answer-prediction/discussion/189555",
      "votes": null
    },
    {
      "id": "1050949",
      "postDate": "10/16/2020 00:01:14",
      "content": "<p>Cheers, Thanks!</p>",
      "rawMarkdown": "Cheers, Thanks!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1050849,
      "author_name": "aquatic",
      "author_url": "",
      "post_date": "10/15/2020 19:39:13",
      "content": "<p>Yes you can use preprocessed data, and it can be private. I believe that as long as the dataset you're deriving your data from is publicly available (e.g. the raw training data and supplementary tables from the competition data), you're fine to make it private. </p>\n<p>I asked the same question recently -- <a href=\"https://www.kaggle.com/c/riiid-test-answer-prediction/discussion/189555\" target=\"_blank\">https://www.kaggle.com/c/riiid-test-answer-prediction/discussion/189555</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 1050949,
          "author_name": "abhaykatoch",
          "author_url": "",
          "post_date": "10/16/2020 00:01:14",
          "content": "<p>Cheers, Thanks!</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1050764": "I just want to pre-process the train data and then store it in a Kaggle dataset so that I don't have to waste memory on re-doing it for every run. The competition rules don't really say anything about this so I was just wondering if anybody knows. \n\nAlso, if allowed, should this dataset be public or private?",
    "1050849": "Yes you can use preprocessed data, and it can be private. I believe that as long as the dataset you're deriving your data from is publicly available (e.g. the raw training data and supplementary tables from the competition data), you're fine to make it private. \n\nI asked the same question recently -- https://www.kaggle.com/c/riiid-test-answer-prediction/discussion/189555",
    "1050949": "Cheers, Thanks!"
  },
  "source": "meta"
}