{
  "id": 602640,
  "title": "Do I need to finish the entire test set to get a submission score?",
  "url": "/competitions/rsna-intracranial-aneurysm-detection/discussion/602640",
  "author_name": "Chia Chuan",
  "post_date": "2025-08-29T01:07:46.101000",
  "votes": 3,
  "comment_count": 1,
  "views": 0,
  "content": "<p>This is my first time participating in a Kaggle competition, and I’m wondering about the submission process. Do we need to complete inference on the entire test set in order to receive a score?</p>\n<p>The time limit is 12 hours for around 2,500 cases, which works out to only about 17 seconds per case (including DICOM loading, slice sorting, and basic preprocessing).</p>\n<p>With this constraint, it seems challenging to use more complex models. Has anyone else run into the same issue, and how did you handle it?</p>\n<p>Thanks!</p>",
  "messages": [
    {
      "id": 3277946,
      "postDate": "2025-08-29T01:07:46.103Z",
      "content": "<p>This is my first time participating in a Kaggle competition, and I’m wondering about the submission process. Do we need to complete inference on the entire test set in order to receive a score?</p>\n<p>The time limit is 12 hours for around 2,500 cases, which works out to only about 17 seconds per case (including DICOM loading, slice sorting, and basic preprocessing).</p>\n<p>With this constraint, it seems challenging to use more complex models. Has anyone else run into the same issue, and how did you handle it?</p>\n<p>Thanks!</p>",
      "rawMarkdown": "This is my first time participating in a Kaggle competition, and I’m wondering about the submission process. Do we need to complete inference on the entire test set in order to receive a score?\n\nThe time limit is 12 hours for around 2,500 cases, which works out to only about 17 seconds per case (including DICOM loading, slice sorting, and basic preprocessing).\n\nWith this constraint, it seems challenging to use more complex models. Has anyone else run into the same issue, and how did you handle it?\n\nThanks!",
      "votes": 3
    },
    {
      "id": 3280280,
      "postDate": "2025-09-02T12:29:37.323Z",
      "content": "<p>Thanks for this question. Hoping <a href=\"https://www.kaggle.com/ryanholbrook\" target=\"_blank\">@ryanholbrook</a> can help with this one.</p>",
      "rawMarkdown": "Thanks for this question. Hoping @ryanholbrook can help with this one."
    }
  ],
  "comments": [
    {
      "id": 3280280,
      "author_name": "Evan Calabrese",
      "author_url": "",
      "post_date": "2025-09-02T12:29:37.323000",
      "content": "<p>Thanks for this question. Hoping <a href=\"https://www.kaggle.com/ryanholbrook\" target=\"_blank\">@ryanholbrook</a> can help with this one.</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3277946": "This is my first time participating in a Kaggle competition, and I’m wondering about the submission process. Do we need to complete inference on the entire test set in order to receive a score?\n\nThe time limit is 12 hours for around 2,500 cases, which works out to only about 17 seconds per case (including DICOM loading, slice sorting, and basic preprocessing).\n\nWith this constraint, it seems challenging to use more complex models. Has anyone else run into the same issue, and how did you handle it?\n\nThanks!",
    "3280280": "Thanks for this question. Hoping @ryanholbrook can help with this one."
  }
}