{
  "id": 188886,
  "title": "How long does it take to Scoring the Submission file?",
  "url": "/competitions/rsna-str-pulmonary-embolism-detection/discussion/188886",
  "author_name": "Ahmed Hamada",
  "post_date": "2020-10-05T18:52:07.736000",
  "votes": 0,
  "comment_count": 2,
  "views": 0,
  "content": "<p>How long does it take to Scoring the Submission file?<br>\nIt take 8 hours and it still \"submission pending\" !!<br>\nWhat's wrong ?!</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4157774%2Fae9b137d2538d11136c89564457e9ef6%2F2020-10-05_204620.png?generation=1601923788889174&amp;alt=media\" alt=\".\"></p>",
  "messages": [
    {
      "id": 1041220,
      "postDate": "2020-10-07T16:12:24.010Z",
      "content": "<p>From data description:</p>\n<blockquote>\n  <p>Note also that the private test set is approximately 3x larger than the public test set (230GB vs. 70GB), so ensure that your kernels have enough time to finish their re-run. The training set includes 7279 studies, the public set 650, and the private set has 1517.</p>\n</blockquote>",
      "rawMarkdown": "From data description:\n> Note also that the private test set is approximately 3x larger than the public test set (230GB vs. 70GB), so ensure that your kernels have enough time to finish their re-run. The training set includes 7279 studies, the public set 650, and the private set has 1517.",
      "votes": 1
    },
    {
      "id": 1038558,
      "postDate": "2020-10-05T21:31:31.823Z",
      "content": "<p>Indeed, one of the rules of the competition is to have less than 9h runtime. Try GPU, it may take less time. You may try other tricks, such as running training on less data or running a less complex model -- it worked for me. This discussion may be a clue also : <br>\n<a href=\"https://www.kaggle.com/c/rsna-str-pulmonary-embolism-detection/discussion/186982\" target=\"_blank\">https://www.kaggle.com/c/rsna-str-pulmonary-embolism-detection/discussion/186982</a></p>",
      "rawMarkdown": "Indeed, one of the rules of the competition is to have less than 9h runtime. Try GPU, it may take less time. You may try other tricks, such as running training on less data or running a less complex model -- it worked for me. This discussion may be a clue also : \nhttps://www.kaggle.com/c/rsna-str-pulmonary-embolism-detection/discussion/186982",
      "votes": 1
    },
    {
      "id": 1038369,
      "postDate": "2020-10-05T18:52:07.737Z",
      "content": "<p>How long does it take to Scoring the Submission file?<br>\nIt take 8 hours and it still \"submission pending\" !!<br>\nWhat's wrong ?!</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4157774%2Fae9b137d2538d11136c89564457e9ef6%2F2020-10-05_204620.png?generation=1601923788889174&amp;alt=media\" alt=\".\"></p>",
      "rawMarkdown": "How long does it take to Scoring the Submission file?\nIt take 8 hours and it still \"submission pending\" !!\nWhat's wrong ?!\n\n![.](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4157774%2Fae9b137d2538d11136c89564457e9ef6%2F2020-10-05_204620.png?generation=1601923788889174&alt=media)"
    }
  ],
  "comments": [
    {
      "id": 1041220,
      "author_name": "Araik Tamazian",
      "author_url": "",
      "post_date": "2020-10-07T16:12:24.010000",
      "content": "<p>From data description:</p>\n<blockquote>\n  <p>Note also that the private test set is approximately 3x larger than the public test set (230GB vs. 70GB), so ensure that your kernels have enough time to finish their re-run. The training set includes 7279 studies, the public set 650, and the private set has 1517.</p>\n</blockquote>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 1038558,
      "author_name": "Catadanna",
      "author_url": "",
      "post_date": "2020-10-05T21:31:31.823000",
      "content": "<p>Indeed, one of the rules of the competition is to have less than 9h runtime. Try GPU, it may take less time. You may try other tricks, such as running training on less data or running a less complex model -- it worked for me. This discussion may be a clue also : <br>\n<a href=\"https://www.kaggle.com/c/rsna-str-pulmonary-embolism-detection/discussion/186982\" target=\"_blank\">https://www.kaggle.com/c/rsna-str-pulmonary-embolism-detection/discussion/186982</a></p>",
      "votes": 1,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1041220": "From data description:\n> Note also that the private test set is approximately 3x larger than the public test set (230GB vs. 70GB), so ensure that your kernels have enough time to finish their re-run. The training set includes 7279 studies, the public set 650, and the private set has 1517.",
    "1038558": "Indeed, one of the rules of the competition is to have less than 9h runtime. Try GPU, it may take less time. You may try other tricks, such as running training on less data or running a less complex model -- it worked for me. This discussion may be a clue also : \nhttps://www.kaggle.com/c/rsna-str-pulmonary-embolism-detection/discussion/186982",
    "1038369": "How long does it take to Scoring the Submission file?\nIt take 8 hours and it still \"submission pending\" !!\nWhat's wrong ?!\n\n![.](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4157774%2Fae9b137d2538d11136c89564457e9ef6%2F2020-10-05_204620.png?generation=1601923788889174&alt=media)"
  }
}