{
  "id": 171429,
  "title": "sample_submission.csv VS all samples in the test folder",
  "url": "/competitions/landmark-recognition-2020/discussion/171429",
  "author_name": "",
  "post_date": "2020-07-31T18:45:50.605491500Z",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>On generating submission file. should I loop through on all the ids in sample_submission.csv or all the files in the test folder? </p>\n\n<p>I see some notebooks are using the sample_submission.csv one. I don't understand why not just looping through all the files in the test folder? are they always the same (even testing in private data during submission) ? </p>",
  "messages": [
    {
      "id": "953394",
      "postDate": "07/31/2020 18:45:50",
      "content": "<p>On generating submission file. should I loop through on all the ids in sample_submission.csv or all the files in the test folder? </p>\n\n<p>I see some notebooks are using the sample_submission.csv one. I don't understand why not just looping through all the files in the test folder? are they always the same (even testing in private data during submission) ? </p>",
      "rawMarkdown": "On generating submission file. should I loop through on all the ids in sample_submission.csv or all the files in the test folder? \n\nI see some notebooks are using the sample_submission.csv one. I don't understand why not just looping through all the files in the test folder? are they always the same (even testing in private data during submission) ?",
      "votes": null
    },
    {
      "id": "953398",
      "postDate": "07/31/2020 18:51:21",
      "content": "<p>You can do either and they should always be identical. Typically the sample submission is the safest option, as that is what we directly use to setup scoring.</p>",
      "rawMarkdown": "You can do either and they should always be identical. Typically the sample submission is the safest option, as that is what we directly use to setup scoring.",
      "votes": null
    },
    {
      "id": "953433",
      "postDate": "07/31/2020 19:35:53",
      "content": "<p>Thanks for your prompt reply! </p>",
      "rawMarkdown": "Thanks for your prompt reply!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 953398,
      "author_name": "wcukierski",
      "author_url": "",
      "post_date": "07/31/2020 18:51:21",
      "content": "<p>You can do either and they should always be identical. Typically the sample submission is the safest option, as that is what we directly use to setup scoring.</p>",
      "votes": null,
      "replies": [
        {
          "id": 953433,
          "author_name": "yusc2020",
          "author_url": "",
          "post_date": "07/31/2020 19:35:53",
          "content": "<p>Thanks for your prompt reply! </p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "953394": "On generating submission file. should I loop through on all the ids in sample_submission.csv or all the files in the test folder? \n\nI see some notebooks are using the sample_submission.csv one. I don't understand why not just looping through all the files in the test folder? are they always the same (even testing in private data during submission) ?",
    "953398": "You can do either and they should always be identical. Typically the sample submission is the safest option, as that is what we directly use to setup scoring.",
    "953433": "Thanks for your prompt reply!"
  },
  "source": "meta"
}