{
  "id": 447332,
  "title": "Is it okay to use bidirectional model.",
  "url": "/competitions/child-mind-institute-detect-sleep-states/discussion/447332",
  "author_name": "",
  "post_date": "2023-10-15T11:38:28.864376200Z",
  "votes": 1,
  "comment_count": 4,
  "views": 0,
  "content": "<p>I am thinking about this dataset. I am not able to convince myself as a bidirectional model actually will be using future data point to detect the event. Such model might limit the applicability of the model or even can lead to data leakage. </p>",
  "messages": [
    {
      "id": "2482939",
      "postDate": "10/15/2023 11:38:28",
      "content": "<p>I am thinking about this dataset. I am not able to convince myself as a bidirectional model actually will be using future data point to detect the event. Such model might limit the applicability of the model or even can lead to data leakage. </p>",
      "rawMarkdown": "I am thinking about this dataset. I am not able to convince myself as a bidirectional model actually will be using future data point to detect the event. Such model might limit the applicability of the model or even can lead to data leakage.",
      "votes": null
    },
    {
      "id": "2483170",
      "postDate": "10/15/2023 14:25:28",
      "content": "<p><a href=\"https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/discussion/437534#2427761\" target=\"_blank\">https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/discussion/437534#2427761</a><br>\nFeel free to use the future data, as confirmed by the competition host linked above. <br>\nThis competition does not aim to detect the sleep in real time, but just do a segmentation after obtaining the full data.</p>",
      "rawMarkdown": "https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/discussion/437534#2427761\nFeel free to use the future data, as confirmed by the competition host linked above. \nThis competition does not aim to detect the sleep in real time, but just do a segmentation after obtaining the full data.",
      "votes": null
    },
    {
      "id": "2483189",
      "postDate": "10/15/2023 14:33:42",
      "content": "<p>Thanks for sharing that I really appreciate it.</p>",
      "rawMarkdown": "Thanks for sharing that I really appreciate it.",
      "votes": null
    },
    {
      "id": "2483330",
      "postDate": "10/15/2023 16:28:48",
      "content": "<p>Thanks for sharing!</p>",
      "rawMarkdown": "Thanks for sharing!",
      "votes": null
    },
    {
      "id": "2488200",
      "postDate": "10/19/2023 05:50:59",
      "content": "<p>BiLSTM first comes into my mind. </p>",
      "rawMarkdown": "BiLSTM first comes into my mind.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2483170,
      "author_name": "buumoo",
      "author_url": "",
      "post_date": "10/15/2023 14:25:28",
      "content": "<p><a href=\"https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/discussion/437534#2427761\" target=\"_blank\">https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/discussion/437534#2427761</a><br>\nFeel free to use the future data, as confirmed by the competition host linked above. <br>\nThis competition does not aim to detect the sleep in real time, but just do a segmentation after obtaining the full data.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2483189,
          "author_name": "riyajm",
          "author_url": "",
          "post_date": "10/15/2023 14:33:42",
          "content": "<p>Thanks for sharing that I really appreciate it.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 2483330,
          "author_name": "haoqinhong",
          "author_url": "",
          "post_date": "10/15/2023 16:28:48",
          "content": "<p>Thanks for sharing!</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 2488200,
      "author_name": "liushundawang",
      "author_url": "",
      "post_date": "10/19/2023 05:50:59",
      "content": "<p>BiLSTM first comes into my mind. </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2482939": "I am thinking about this dataset. I am not able to convince myself as a bidirectional model actually will be using future data point to detect the event. Such model might limit the applicability of the model or even can lead to data leakage.",
    "2483170": "https://www.kaggle.com/competitions/child-mind-institute-detect-sleep-states/discussion/437534#2427761\nFeel free to use the future data, as confirmed by the competition host linked above. \nThis competition does not aim to detect the sleep in real time, but just do a segmentation after obtaining the full data.",
    "2483189": "Thanks for sharing that I really appreciate it.",
    "2483330": "Thanks for sharing!",
    "2488200": "BiLSTM first comes into my mind."
  },
  "source": "meta"
}