{
  "id": 585239,
  "title": "IMU-only data",
  "url": "/competitions/cmi-detect-behavior-with-sensor-data/discussion/585239",
  "author_name": "",
  "post_date": "2025-06-19T02:47:38.651509600Z",
  "votes": 1,
  "comment_count": 3,
  "views": 0,
  "content": "<p>In reference to the following on the <a href=\"https://www.kaggle.com/competitions/cmi-detect-behavior-with-sensor-data/data\" target=\"_blank\">Data page</a>:</p>\n<blockquote>\n  <p>Expect approximately 3,500 sequences in the hidden test set.<br>\n  Half of the hidden-test sequences are recorded with IMU only; the thermopile (thm_) and time-of-flight (tof__v*) columns are still present but contain null values for those sequences. </p>\n</blockquote>\n<p>Has anyone got an idea if this actually means null or -1? It is unfortunate that the test.csv data does not have a sample with missing thm and tof data (both sample sequences have complete data recorded in one or more of the thm and tof variables). -1 is used throughout tof for \"no sensor response\" but I think this is different to the IMU-only data. </p>\n<p>I'd like to have a conditional on the test sequences with IMU-only data, but I'm not entirely sure what data to expect. </p>",
  "messages": [
    {
      "id": "3227441",
      "postDate": "06/19/2025 02:47:38",
      "content": "<p>In reference to the following on the <a href=\"https://www.kaggle.com/competitions/cmi-detect-behavior-with-sensor-data/data\" target=\"_blank\">Data page</a>:</p>\n<blockquote>\n  <p>Expect approximately 3,500 sequences in the hidden test set.<br>\n  Half of the hidden-test sequences are recorded with IMU only; the thermopile (thm_) and time-of-flight (tof__v*) columns are still present but contain null values for those sequences. </p>\n</blockquote>\n<p>Has anyone got an idea if this actually means null or -1? It is unfortunate that the test.csv data does not have a sample with missing thm and tof data (both sample sequences have complete data recorded in one or more of the thm and tof variables). -1 is used throughout tof for \"no sensor response\" but I think this is different to the IMU-only data. </p>\n<p>I'd like to have a conditional on the test sequences with IMU-only data, but I'm not entirely sure what data to expect. </p>",
      "rawMarkdown": "In reference to the following on the [Data page](https://www.kaggle.com/competitions/cmi-detect-behavior-with-sensor-data/data):\n\n>Expect approximately 3,500 sequences in the hidden test set.\nHalf of the hidden-test sequences are recorded with IMU only; the thermopile (thm_) and time-of-flight (tof__v*) columns are still present but contain null values for those sequences. \n\nHas anyone got an idea if this actually means null or -1? It is unfortunate that the test.csv data does not have a sample with missing thm and tof data (both sample sequences have complete data recorded in one or more of the thm and tof variables). -1 is used throughout tof for \"no sensor response\" but I think this is different to the IMU-only data. \n\nI'd like to have a conditional on the test sequences with IMU-only data, but I'm not entirely sure what data to expect.",
      "votes": null
    },
    {
      "id": "3227469",
      "postDate": "06/19/2025 03:05:30",
      "content": "<p>I think I answered my own question. There are several sequences in train.csv with null values in all of the thm and tof columns, such as SEQ_025240. I think that is probably the template for IMU-only data.</p>",
      "rawMarkdown": "I think I answered my own question. There are several sequences in train.csv with null values in all of the thm and tof columns, such as SEQ_025240. I think that is probably the template for IMU-only data.",
      "votes": null
    },
    {
      "id": "3227791",
      "postDate": "06/19/2025 09:28:06",
      "content": "<p>it's null. You can check the train set some of the columns also contain null values.<br>\nBut I an unsure of for a single sequence will values for <code>th_m</code> be null or part of them be null</p>",
      "rawMarkdown": "it's null. You can check the train set some of the columns also contain null values.\nBut I an unsure of for a single sequence will values for `th_m` be null or part of them be null",
      "votes": null
    },
    {
      "id": "3232195",
      "postDate": "06/25/2025 13:28:13",
      "content": "<p>You can detect the NaN's and choose a model accordingly.</p>",
      "rawMarkdown": "You can detect the NaN's and choose a model accordingly.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3227469,
      "author_name": "nnjjpp",
      "author_url": "",
      "post_date": "06/19/2025 03:05:30",
      "content": "<p>I think I answered my own question. There are several sequences in train.csv with null values in all of the thm and tof columns, such as SEQ_025240. I think that is probably the template for IMU-only data.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 3227791,
      "author_name": "zhangyue199",
      "author_url": "",
      "post_date": "06/19/2025 09:28:06",
      "content": "<p>it's null. You can check the train set some of the columns also contain null values.<br>\nBut I an unsure of for a single sequence will values for <code>th_m</code> be null or part of them be null</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 3232195,
      "author_name": "stefanoclss",
      "author_url": "",
      "post_date": "06/25/2025 13:28:13",
      "content": "<p>You can detect the NaN's and choose a model accordingly.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3227441": "In reference to the following on the [Data page](https://www.kaggle.com/competitions/cmi-detect-behavior-with-sensor-data/data):\n\n>Expect approximately 3,500 sequences in the hidden test set.\nHalf of the hidden-test sequences are recorded with IMU only; the thermopile (thm_) and time-of-flight (tof__v*) columns are still present but contain null values for those sequences. \n\nHas anyone got an idea if this actually means null or -1? It is unfortunate that the test.csv data does not have a sample with missing thm and tof data (both sample sequences have complete data recorded in one or more of the thm and tof variables). -1 is used throughout tof for \"no sensor response\" but I think this is different to the IMU-only data. \n\nI'd like to have a conditional on the test sequences with IMU-only data, but I'm not entirely sure what data to expect.",
    "3227469": "I think I answered my own question. There are several sequences in train.csv with null values in all of the thm and tof columns, such as SEQ_025240. I think that is probably the template for IMU-only data.",
    "3227791": "it's null. You can check the train set some of the columns also contain null values.\nBut I an unsure of for a single sequence will values for `th_m` be null or part of them be null",
    "3232195": "You can detect the NaN's and choose a model accordingly."
  },
  "source": "meta"
}