{
  "id": 384885,
  "title": "Auxiliary information biased to certain sensors",
  "url": "/competitions/icecube-neutrinos-in-deep-ice/discussion/384885",
  "author_name": "",
  "post_date": "2023-02-10T03:03:10.158419400Z",
  "votes": 3,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hi!</p>\n<p>I've recently explored the <a href=\"https://www.kaggle.com/code/suhancho/eda-inspection-of-noisy-sensors/notebook\" target=\"_blank\">number of Auxiliary=False and True signals</a> across all batches grouped by sensor ids.<br>\nAnd I figured out that there's a certain bias of numbers between the signals from DeepCore, and those not from DeepCore.</p>\n<p>DeepCore tend to have more Auxiliary=True signals, so i was wondering if there's any reason for that.</p>\n<p>And if that's so, are there normalization or bias-correcting methods for that?</p>",
  "messages": [
    {
      "id": "2137458",
      "postDate": "02/10/2023 03:03:10",
      "content": "<p>Hi!</p>\n<p>I've recently explored the <a href=\"https://www.kaggle.com/code/suhancho/eda-inspection-of-noisy-sensors/notebook\" target=\"_blank\">number of Auxiliary=False and True signals</a> across all batches grouped by sensor ids.<br>\nAnd I figured out that there's a certain bias of numbers between the signals from DeepCore, and those not from DeepCore.</p>\n<p>DeepCore tend to have more Auxiliary=True signals, so i was wondering if there's any reason for that.</p>\n<p>And if that's so, are there normalization or bias-correcting methods for that?</p>",
      "rawMarkdown": "Hi!\n\nI've recently explored the [number of Auxiliary=False and True signals](https://www.kaggle.com/code/suhancho/eda-inspection-of-noisy-sensors/notebook) across all batches grouped by sensor ids.\nAnd I figured out that there's a certain bias of numbers between the signals from DeepCore, and those not from DeepCore.\n\nDeepCore tend to have more Auxiliary=True signals, so i was wondering if there's any reason for that.\n\nAnd if that's so, are there normalization or bias-correcting methods for that?",
      "votes": null
    },
    {
      "id": "2137620",
      "postDate": "02/10/2023 07:28:54",
      "content": "<p>The DeepCore sensors have PMTs with a higher quantum efficiency, so they will in general pick up more hits. I think this explains what you are reporting. </p>",
      "rawMarkdown": "The DeepCore sensors have PMTs with a higher quantum efficiency, so they will in general pick up more hits. I think this explains what you are reporting.",
      "votes": null
    },
    {
      "id": "2139798",
      "postDate": "02/11/2023 06:59:09",
      "content": "<p><a href=\"https://www.kaggle.com/pellerphys\" target=\"_blank\">@pellerphys</a> Thanks!<br>\n But when analyzing signals/charges from multiple sensors, do signals from DeepCore sensors undergo certain processes?</p>",
      "rawMarkdown": "pellerphys Thanks!\n But when analyzing signals/charges from multiple sensors, do signals from DeepCore sensors undergo certain processes?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2137620,
      "author_name": "pellerphys",
      "author_url": "",
      "post_date": "02/10/2023 07:28:54",
      "content": "<p>The DeepCore sensors have PMTs with a higher quantum efficiency, so they will in general pick up more hits. I think this explains what you are reporting. </p>",
      "votes": null,
      "replies": [
        {
          "id": 2139798,
          "author_name": "suhancho",
          "author_url": "",
          "post_date": "02/11/2023 06:59:09",
          "content": "<p><a href=\"https://www.kaggle.com/pellerphys\" target=\"_blank\">@pellerphys</a> Thanks!<br>\n But when analyzing signals/charges from multiple sensors, do signals from DeepCore sensors undergo certain processes?</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2137458": "Hi!\n\nI've recently explored the [number of Auxiliary=False and True signals](https://www.kaggle.com/code/suhancho/eda-inspection-of-noisy-sensors/notebook) across all batches grouped by sensor ids.\nAnd I figured out that there's a certain bias of numbers between the signals from DeepCore, and those not from DeepCore.\n\nDeepCore tend to have more Auxiliary=True signals, so i was wondering if there's any reason for that.\n\nAnd if that's so, are there normalization or bias-correcting methods for that?",
    "2137620": "The DeepCore sensors have PMTs with a higher quantum efficiency, so they will in general pick up more hits. I think this explains what you are reporting.",
    "2139798": "pellerphys Thanks!\n But when analyzing signals/charges from multiple sensors, do signals from DeepCore sensors undergo certain processes?"
  },
  "source": "meta"
}