{
  "id": 89555,
  "title": "Step profile in time to failure",
  "url": "/competitions/LANL-Earthquake-Prediction/discussion/89555",
  "author_name": "",
  "post_date": "2019-04-15T16:35:17.066451600Z",
  "votes": 3,
  "comment_count": 2,
  "views": 0,
  "content": "<p>As mentioned in <a href=\"https://www.kaggle.com/allunia/shaking-earth\">this</a> great kernel, the <code>time_to_failure</code> curve has a step profile.</p>\n\n<p>Bertrand said:</p>\n\n<blockquote>\n  <p>Time to failure is based on a measure of fault strength (shear stress, not part of the data for the competition). When a labquake occurs this stress drops unambiguously.</p>\n</blockquote>\n\n<p>I wonder if the gauge used to measure the shear stress has a lower sampling rate relative to the piezoceramic sensor that records the <code>acoustic_data</code>, and that could be causing the step profile?</p>",
  "messages": [
    {
      "id": "517166",
      "postDate": "04/15/2019 16:35:17",
      "content": "<p>As mentioned in <a href=\"https://www.kaggle.com/allunia/shaking-earth\">this</a> great kernel, the <code>time_to_failure</code> curve has a step profile.</p>\n\n<p>Bertrand said:</p>\n\n<blockquote>\n  <p>Time to failure is based on a measure of fault strength (shear stress, not part of the data for the competition). When a labquake occurs this stress drops unambiguously.</p>\n</blockquote>\n\n<p>I wonder if the gauge used to measure the shear stress has a lower sampling rate relative to the piezoceramic sensor that records the <code>acoustic_data</code>, and that could be causing the step profile?</p>",
      "rawMarkdown": "As mentioned in [this](https://www.kaggle.com/allunia/shaking-earth) great kernel, the `time_to_failure` curve has a step profile.\n\nBertrand said:\n&gt; Time to failure is based on a measure of fault strength (shear stress, not part of the data for the competition). When a labquake occurs this stress drops unambiguously.\n\nI wonder if the gauge used to measure the shear stress has a lower sampling rate relative to the piezoceramic sensor that records the `acoustic_data`, and that could be causing the step profile?",
      "votes": null
    },
    {
      "id": "517313",
      "postDate": "04/15/2019 21:52:29",
      "content": "<p>My guess is it has to do with the equipment used to measure, as you have suggested. </p>",
      "rawMarkdown": "My guess is it has to do with the equipment used to measure, as you have suggested.",
      "votes": null
    },
    {
      "id": "518991",
      "postDate": "04/18/2019 07:52:51",
      "content": "<blockquote>\n  <p>The data is recorded in bins of 4096 samples. Withing those bins seismic data is recorded at 4MHz, but there is a 12 microseconds gap between each bin, an artifact of the recording device. </p>\n</blockquote>\n\n<p>There is a step every 4096 observations...  The step is the 12 microseconds gap.</p>",
      "rawMarkdown": "&gt; The data is recorded in bins of 4096 samples. Withing those bins seismic data is recorded at 4MHz, but there is a 12 microseconds gap between each bin, an artifact of the recording device. \n\nThere is a step every 4096 observations...  The step is the 12 microseconds gap.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 517313,
      "author_name": "halldalton94",
      "author_url": "",
      "post_date": "04/15/2019 21:52:29",
      "content": "<p>My guess is it has to do with the equipment used to measure, as you have suggested. </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 518991,
      "author_name": "cpmpml",
      "author_url": "",
      "post_date": "04/18/2019 07:52:51",
      "content": "<blockquote>\n  <p>The data is recorded in bins of 4096 samples. Withing those bins seismic data is recorded at 4MHz, but there is a 12 microseconds gap between each bin, an artifact of the recording device. </p>\n</blockquote>\n\n<p>There is a step every 4096 observations...  The step is the 12 microseconds gap.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "517166": "As mentioned in [this](https://www.kaggle.com/allunia/shaking-earth) great kernel, the `time_to_failure` curve has a step profile.\n\nBertrand said:\n&gt; Time to failure is based on a measure of fault strength (shear stress, not part of the data for the competition). When a labquake occurs this stress drops unambiguously.\n\nI wonder if the gauge used to measure the shear stress has a lower sampling rate relative to the piezoceramic sensor that records the `acoustic_data`, and that could be causing the step profile?",
    "517313": "My guess is it has to do with the equipment used to measure, as you have suggested.",
    "518991": "&gt; The data is recorded in bins of 4096 samples. Withing those bins seismic data is recorded at 4MHz, but there is a 12 microseconds gap between each bin, an artifact of the recording device. \n\nThere is a step every 4096 observations...  The step is the 12 microseconds gap."
  },
  "source": "meta"
}