{
  "id": 122810,
  "title": "Is the 9hr on P100 applicable on Private Test Set as well?",
  "url": "/competitions/deepfake-detection-challenge/discussion/122810",
  "author_name": "hirviö",
  "post_date": "2019-12-23T02:52:01.871000",
  "votes": 3,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Nothing makes it clear that the 9hr limit is applicable on the final Private Test Set? If so, is there any idea of the relative size of the Private Test when compared to Public Test or Public Val? This part from Getting Started might suggest that the limit is only for Public Test on the Leaderboard(?)</p>\n\n<p><a href=\"https://www.kaggle.com/c/deepfake-detection-challenge/overview/getting-started\">https://www.kaggle.com/c/deepfake-detection-challenge/overview/getting-started</a></p>\n\n<blockquote>\n  <p>You are limited to 9 hours of GPU compute time. This constraint is also imposed on the Public Test Set re-run. You should anticipate the Public Test set to be 10 times the size of the Public Validation Set and budget accordingly. The 9 hour compute constraint will be applied to the submission in Kaggle, when evaluated against the Public Test Set. So if you are able to make a successful submission that posts a score to the public leaderboard without an error, then it has met the 9 hour constraint on the Public Test Set. Review the full set of Code Requirements.</p>\n</blockquote>",
  "messages": [
    {
      "id": 701048,
      "postDate": "2019-12-23T02:52:01.870Z",
      "content": "<p>Nothing makes it clear that the 9hr limit is applicable on the final Private Test Set? If so, is there any idea of the relative size of the Private Test when compared to Public Test or Public Val? This part from Getting Started might suggest that the limit is only for Public Test on the Leaderboard(?)</p>\n\n<p><a href=\"https://www.kaggle.com/c/deepfake-detection-challenge/overview/getting-started\">https://www.kaggle.com/c/deepfake-detection-challenge/overview/getting-started</a></p>\n\n<blockquote>\n  <p>You are limited to 9 hours of GPU compute time. This constraint is also imposed on the Public Test Set re-run. You should anticipate the Public Test set to be 10 times the size of the Public Validation Set and budget accordingly. The 9 hour compute constraint will be applied to the submission in Kaggle, when evaluated against the Public Test Set. So if you are able to make a successful submission that posts a score to the public leaderboard without an error, then it has met the 9 hour constraint on the Public Test Set. Review the full set of Code Requirements.</p>\n</blockquote>",
      "rawMarkdown": "Nothing makes it clear that the 9hr limit is applicable on the final Private Test Set? If so, is there any idea of the relative size of the Private Test when compared to Public Test or Public Val? This part from Getting Started might suggest that the limit is only for Public Test on the Leaderboard(?)\n\nhttps://www.kaggle.com/c/deepfake-detection-challenge/overview/getting-started\n\n&gt; You are limited to 9 hours of GPU compute time. This constraint is also imposed on the Public Test Set re-run. You should anticipate the Public Test set to be 10 times the size of the Public Validation Set and budget accordingly. The 9 hour compute constraint will be applied to the submission in Kaggle, when evaluated against the Public Test Set. So if you are able to make a successful submission that posts a score to the public leaderboard without an error, then it has met the 9 hour constraint on the Public Test Set. Review the full set of Code Requirements.",
      "votes": 3
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "701048": "Nothing makes it clear that the 9hr limit is applicable on the final Private Test Set? If so, is there any idea of the relative size of the Private Test when compared to Public Test or Public Val? This part from Getting Started might suggest that the limit is only for Public Test on the Leaderboard(?)\n\nhttps://www.kaggle.com/c/deepfake-detection-challenge/overview/getting-started\n\n&gt; You are limited to 9 hours of GPU compute time. This constraint is also imposed on the Public Test Set re-run. You should anticipate the Public Test set to be 10 times the size of the Public Validation Set and budget accordingly. The 9 hour compute constraint will be applied to the submission in Kaggle, when evaluated against the Public Test Set. So if you are able to make a successful submission that posts a score to the public leaderboard without an error, then it has met the 9 hour constraint on the Public Test Set. Review the full set of Code Requirements."
  }
}