{
  "id": 450768,
  "title": "Does predicting for missing labels or unlabeled data impact the Leaderboard (LB) score?",
  "url": "/competitions/child-mind-institute-detect-sleep-states/discussion/450768",
  "author_name": "",
  "post_date": "2023-10-25T14:38:34.013547400Z",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Does predicting for missing labels or unlabeled data impact the Leaderboard (LB) score? If so, could you recommend potential strategies for addressing this issue? Thank you.  <img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4583131%2Fd48fb54311496cbd60dbb318daec5344%2Fdownload.png?generation=1698244385716701&amp;alt=media\" alt=\"\"></p>",
  "messages": [
    {
      "id": "2498828",
      "postDate": "10/25/2023 14:38:34",
      "content": "<p>Does predicting for missing labels or unlabeled data impact the Leaderboard (LB) score? If so, could you recommend potential strategies for addressing this issue? Thank you.  <img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4583131%2Fd48fb54311496cbd60dbb318daec5344%2Fdownload.png?generation=1698244385716701&amp;alt=media\" alt=\"\"></p>",
      "rawMarkdown": "Does predicting for missing labels or unlabeled data impact the Leaderboard (LB) score? If so, could you recommend potential strategies for addressing this issue? Thank you.  ![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4583131%2Fd48fb54311496cbd60dbb318daec5344%2Fdownload.png?generation=1698244385716701&alt=media)",
      "votes": null
    },
    {
      "id": "2498966",
      "postDate": "10/25/2023 16:21:38",
      "content": "<p>Yes it does, it counts as a false positive</p>",
      "rawMarkdown": "Yes it does, it counts as a false positive",
      "votes": null
    },
    {
      "id": "2500353",
      "postDate": "10/26/2023 15:41:43",
      "content": "<p>Below is a quote from the <code>Scoring</code> section from here: <a href=\"https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/overview\" target=\"_blank\">https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/overview</a></p>\n<blockquote>\n  <p>In this calculation, matched predictions over the threshold are scored as TP and unmatched predictions as FP. Unmatched ground-truths are scored as FN.</p>\n</blockquote>",
      "rawMarkdown": "Below is a quote from the `Scoring` section from here: https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/overview\n\n> In this calculation, matched predictions over the threshold are scored as TP and unmatched predictions as FP. Unmatched ground-truths are scored as FN.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2498966,
      "author_name": "fahmiayari",
      "author_url": "",
      "post_date": "10/25/2023 16:21:38",
      "content": "<p>Yes it does, it counts as a false positive</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2500353,
      "author_name": "bkowshik",
      "author_url": "",
      "post_date": "10/26/2023 15:41:43",
      "content": "<p>Below is a quote from the <code>Scoring</code> section from here: <a href=\"https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/overview\" target=\"_blank\">https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/overview</a></p>\n<blockquote>\n  <p>In this calculation, matched predictions over the threshold are scored as TP and unmatched predictions as FP. Unmatched ground-truths are scored as FN.</p>\n</blockquote>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2498828": "Does predicting for missing labels or unlabeled data impact the Leaderboard (LB) score? If so, could you recommend potential strategies for addressing this issue? Thank you.  ![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4583131%2Fd48fb54311496cbd60dbb318daec5344%2Fdownload.png?generation=1698244385716701&alt=media)",
    "2498966": "Yes it does, it counts as a false positive",
    "2500353": "Below is a quote from the `Scoring` section from here: https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/overview\n\n> In this calculation, matched predictions over the threshold are scored as TP and unmatched predictions as FP. Unmatched ground-truths are scored as FN."
  },
  "source": "meta"
}