{
  "id": 290411,
  "title": "Generating additional labeling data",
  "url": "/competitions/tensorflow-great-barrier-reef/discussion/290411",
  "author_name": "Wonjun Park",
  "post_date": "2021-11-24T14:23:14.086000",
  "votes": 15,
  "comment_count": 2,
  "views": 0,
  "content": "<p>As a result of EDA, 80% of the learning data is not labeled.<br>\nBut I think we can get much more labeling data.</p>\n<p>Watch the video on the next notebook!<br>\n<a href=\"https://www.kaggle.com/diegoalejogm/great-barrier-reefs-eda-with-animations\" target=\"_blank\">https://www.kaggle.com/diegoalejogm/great-barrier-reefs-eda-with-animations</a></p>\n<p>In the continuous video, we can be seen that labeling occurs at some point.<br>\nWouldn't it be possible to add labeling based on this point in time?</p>",
  "messages": [
    {
      "id": 1594082,
      "postDate": "2021-11-24T14:23:14.087Z",
      "content": "<p>As a result of EDA, 80% of the learning data is not labeled.<br>\nBut I think we can get much more labeling data.</p>\n<p>Watch the video on the next notebook!<br>\n<a href=\"https://www.kaggle.com/diegoalejogm/great-barrier-reefs-eda-with-animations\" target=\"_blank\">https://www.kaggle.com/diegoalejogm/great-barrier-reefs-eda-with-animations</a></p>\n<p>In the continuous video, we can be seen that labeling occurs at some point.<br>\nWouldn't it be possible to add labeling based on this point in time?</p>",
      "rawMarkdown": "As a result of EDA, 80% of the learning data is not labeled.\nBut I think we can get much more labeling data.\n\nWatch the video on the next notebook!\nhttps://www.kaggle.com/diegoalejogm/great-barrier-reefs-eda-with-animations\n\nIn the continuous video, we can be seen that labeling occurs at some point.\nWouldn't it be possible to add labeling based on this point in time?",
      "votes": 15
    },
    {
      "id": 1594146,
      "postDate": "2021-11-24T15:16:11.570Z",
      "content": "<p>In that video from the notebook, the perspective view is changing. It is hard to say when it starts to be a starfish. It would be harmful if starfish is annotated only for a few pixel visibility.</p>",
      "rawMarkdown": "In that video from the notebook, the perspective view is changing. It is hard to say when it starts to be a starfish. It would be harmful if starfish is annotated only for a few pixel visibility.",
      "votes": 3
    },
    {
      "id": 1600692,
      "postDate": "2021-11-30T16:25:11.260Z",
      "rawMarkdown": "",
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 1594146,
      "author_name": "Mykola",
      "author_url": "",
      "post_date": "2021-11-24T15:16:11.570000",
      "content": "<p>In that video from the notebook, the perspective view is changing. It is hard to say when it starts to be a starfish. It would be harmful if starfish is annotated only for a few pixel visibility.</p>",
      "votes": 3,
      "replies": []
    },
    {
      "id": 1600692,
      "author_name": "",
      "author_url": "",
      "post_date": "2021-11-30T16:25:11.260000",
      "content": "",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1594082": "As a result of EDA, 80% of the learning data is not labeled.\nBut I think we can get much more labeling data.\n\nWatch the video on the next notebook!\nhttps://www.kaggle.com/diegoalejogm/great-barrier-reefs-eda-with-animations\n\nIn the continuous video, we can be seen that labeling occurs at some point.\nWouldn't it be possible to add labeling based on this point in time?",
    "1594146": "In that video from the notebook, the perspective view is changing. It is hard to say when it starts to be a starfish. It would be harmful if starfish is annotated only for a few pixel visibility.",
    "1600692": ""
  }
}