{
  "id": 103272,
  "title": "When I submit the submission.csv, the kernel will re-run again?",
  "url": "/competitions/aptos2019-blindness-detection/discussion/103272",
  "author_name": "",
  "post_date": "2019-08-08T08:00:50.622690800Z",
  "votes": null,
  "comment_count": 3,
  "views": 0,
  "content": "<p>It took almost one hour to calculate score. So the score calculation will re-run the kernel? I think it should only need submission.csv.</p>",
  "messages": [
    {
      "id": "594610",
      "postDate": "08/08/2019 08:00:50",
      "content": "<p>It took almost one hour to calculate score. So the score calculation will re-run the kernel? I think it should only need submission.csv.</p>",
      "rawMarkdown": "It took almost one hour to calculate score. So the score calculation will re-run the kernel? I think it should only need submission.csv.",
      "votes": null
    },
    {
      "id": "594615",
      "postDate": "08/08/2019 08:05:46",
      "content": "<p>Yes, it does. Once you submit, the kernel will re-run using the whole test.csv (public + private)</p>",
      "rawMarkdown": "Yes, it does. Once you submit, the kernel will re-run using the whole test.csv (public + private)",
      "votes": null
    },
    {
      "id": "594629",
      "postDate": "08/08/2019 08:25:19",
      "content": "<p>Thanks for your reply. I got the answer but wonder why do not just use the submission.csv for score calculation as the kernel has got run when commit.  Never mind :)</p>",
      "rawMarkdown": "Thanks for your reply. I got the answer but wonder why do not just use the submission.csv for score calculation as the kernel has got run when commit.  Never mind :)",
      "votes": null
    },
    {
      "id": "594820",
      "postDate": "08/08/2019 13:37:54",
      "content": "<p>This is a <a href=\"http://blog.kaggle.com/2019/05/17/synchronous-kernels-only-competitions-real-ml-in-real-time/\">Synchronous Kernels-only Competition</a> - after you submit, both public and private LB are scored(but only private is displayed so far).\nBut if you only want to check public LB quickly, you can use <a href=\"https://www.kaggle.com/c/aptos2019-blindness-detection/discussion/98879\">that tip</a>.\nI do it slightly differently - save submission.csv in a dataset and in the separate submission kernel just merge it with the sample submission and output as submission.csv. Then submit that file - it takes few seconds to check LB that way.\nDon't use it as a final submission of course.  </p>",
      "rawMarkdown": "This is a [Synchronous Kernels-only Competition](http://blog.kaggle.com/2019/05/17/synchronous-kernels-only-competitions-real-ml-in-real-time/) - after you submit, both public and private LB are scored(but only private is displayed so far).\nBut if you only want to check public LB quickly, you can use [that tip](https://www.kaggle.com/c/aptos2019-blindness-detection/discussion/98879).\nI do it slightly differently - save submission.csv in a dataset and in the separate submission kernel just merge it with the sample submission and output as submission.csv. Then submit that file - it takes few seconds to check LB that way.\nDon't use it as a final submission of course.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 594615,
      "author_name": "raimonds1993",
      "author_url": "",
      "post_date": "08/08/2019 08:05:46",
      "content": "<p>Yes, it does. Once you submit, the kernel will re-run using the whole test.csv (public + private)</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 594629,
      "author_name": "tabsun",
      "author_url": "",
      "post_date": "08/08/2019 08:25:19",
      "content": "<p>Thanks for your reply. I got the answer but wonder why do not just use the submission.csv for score calculation as the kernel has got run when commit.  Never mind :)</p>",
      "votes": null,
      "replies": [
        {
          "id": 594820,
          "author_name": "vzaguskin",
          "author_url": "",
          "post_date": "08/08/2019 13:37:54",
          "content": "<p>This is a <a href=\"http://blog.kaggle.com/2019/05/17/synchronous-kernels-only-competitions-real-ml-in-real-time/\">Synchronous Kernels-only Competition</a> - after you submit, both public and private LB are scored(but only private is displayed so far).\nBut if you only want to check public LB quickly, you can use <a href=\"https://www.kaggle.com/c/aptos2019-blindness-detection/discussion/98879\">that tip</a>.\nI do it slightly differently - save submission.csv in a dataset and in the separate submission kernel just merge it with the sample submission and output as submission.csv. Then submit that file - it takes few seconds to check LB that way.\nDon't use it as a final submission of course.  </p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "594610": "It took almost one hour to calculate score. So the score calculation will re-run the kernel? I think it should only need submission.csv.",
    "594615": "Yes, it does. Once you submit, the kernel will re-run using the whole test.csv (public + private)",
    "594629": "Thanks for your reply. I got the answer but wonder why do not just use the submission.csv for score calculation as the kernel has got run when commit.  Never mind :)",
    "594820": "This is a [Synchronous Kernels-only Competition](http://blog.kaggle.com/2019/05/17/synchronous-kernels-only-competitions-real-ml-in-real-time/) - after you submit, both public and private LB are scored(but only private is displayed so far).\nBut if you only want to check public LB quickly, you can use [that tip](https://www.kaggle.com/c/aptos2019-blindness-detection/discussion/98879).\nI do it slightly differently - save submission.csv in a dataset and in the separate submission kernel just merge it with the sample submission and output as submission.csv. Then submit that file - it takes few seconds to check LB that way.\nDon't use it as a final submission of course."
  },
  "source": "meta"
}