{
  "id": 542953,
  "title": "Discrepancy in integration time between columns in axis_info",
  "url": "/competitions/ariel-data-challenge-2024/discussion/542953",
  "author_name": "Maciej J Mikulski",
  "post_date": "2024-10-27T20:57:04.694000",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I analyzed <code>axis_info.parquet</code> especially for AIRS data found that:</p>\n<ul>\n<li><code>AIRS-CH0-integration_time</code> contains two values: 0.1 and 4.5, as expected - this is CDS method</li>\n<li>on the other hand, looking directly at <code>AIRS-CH0-axis0-h</code> suggest that integration times are 0.1 and 4.7 respecitvely:</li>\n</ul>\n<pre><code>a = df[]\n(a[]-a[]) *  \n(a[]-a[]) *  \n(a[]-a[]) *  \n(a[] - a[]) *  \n</code></pre>\n<p><a href=\"https://www.kaggle.com/lorenzomugnai\" target=\"_blank\">@lorenzomugnai</a> Could you help understanding  that? Shall we go with (4.5-0.1) or with (4.7 - 0.1) when removing dark current after CDS?</p>",
  "messages": [
    {
      "id": 3030240,
      "postDate": "2024-10-28T10:32:40.593Z",
      "content": "<p>Hello, <br>\nI think this discussion may help:<br>\n<a href=\"https://www.kaggle.com/code/gordonyip/update-calibrating-and-binning-astronomical-data/comments#2964529\" target=\"_blank\">https://www.kaggle.com/code/gordonyip/update-calibrating-and-binning-astronomical-data/comments#2964529</a></p>\n<p>Let me know if you have any other question</p>",
      "rawMarkdown": "Hello, \nI think this discussion may help:\nhttps://www.kaggle.com/code/gordonyip/update-calibrating-and-binning-astronomical-data/comments#2964529\n\nLet me know if you have any other question",
      "votes": 1,
      "replies": [
        {
          "id": 3030788,
          "postDate": "2024-10-28T22:10:07.697Z",
          "content": "<p>Yes, thank you, that was exactly it! I added also a comment there.</p>",
          "rawMarkdown": "Yes, thank you, that was exactly it! I added also a comment there."
        }
      ]
    },
    {
      "id": 3029868,
      "postDate": "2024-10-27T20:57:04.693Z",
      "content": "<p>I analyzed <code>axis_info.parquet</code> especially for AIRS data found that:</p>\n<ul>\n<li><code>AIRS-CH0-integration_time</code> contains two values: 0.1 and 4.5, as expected - this is CDS method</li>\n<li>on the other hand, looking directly at <code>AIRS-CH0-axis0-h</code> suggest that integration times are 0.1 and 4.7 respecitvely:</li>\n</ul>\n<pre><code>a = df[]\n(a[]-a[]) *  \n(a[]-a[]) *  \n(a[]-a[]) *  \n(a[] - a[]) *  \n</code></pre>\n<p><a href=\"https://www.kaggle.com/lorenzomugnai\" target=\"_blank\">@lorenzomugnai</a> Could you help understanding  that? Shall we go with (4.5-0.1) or with (4.7 - 0.1) when removing dark current after CDS?</p>",
      "rawMarkdown": "I analyzed `axis_info.parquet` especially for AIRS data found that:\n\n- `AIRS-CH0-integration_time` contains two values: 0.1 and 4.5, as expected - this is CDS method\n- on the other hand, looking directly at `AIRS-CH0-axis0-h` suggest that integration times are 0.1 and 4.7 respecitvely:\n\n```py\na = df['AIRS-CH0-axis0-h']\n(a[1]-a[0]) * 3600 # gives: np.float64(0.1)\n(a[4]-a[3]) * 3600 # gives: np.float64(4.7)\n(a[6]-a[5]) * 3600 # gives: np.float64(4.699999999999999)\n(a[10000] - a[9999]) * 3600 # gives np.float64(4.700000000000415)\n```\n\n@lorenzomugnai Could you help understanding  that? Shall we go with (4.5-0.1) or with (4.7 - 0.1) when removing dark current after CDS?",
      "votes": 1
    }
  ],
  "comments": [
    {
      "id": 3030240,
      "author_name": "Lorenzo Mugnai",
      "author_url": "",
      "post_date": "2024-10-28T10:32:40.593000",
      "content": "<p>Hello, <br>\nI think this discussion may help:<br>\n<a href=\"https://www.kaggle.com/code/gordonyip/update-calibrating-and-binning-astronomical-data/comments#2964529\" target=\"_blank\">https://www.kaggle.com/code/gordonyip/update-calibrating-and-binning-astronomical-data/comments#2964529</a></p>\n<p>Let me know if you have any other question</p>",
      "votes": 1,
      "replies": [
        {
          "id": 3030788,
          "author_name": "Maciej J Mikulski",
          "author_url": "",
          "post_date": "2024-10-28T22:10:07.697000",
          "content": "<p>Yes, thank you, that was exactly it! I added also a comment there.</p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3030240": "Hello, \nI think this discussion may help:\nhttps://www.kaggle.com/code/gordonyip/update-calibrating-and-binning-astronomical-data/comments#2964529\n\nLet me know if you have any other question",
    "3029868": "I analyzed `axis_info.parquet` especially for AIRS data found that:\n\n- `AIRS-CH0-integration_time` contains two values: 0.1 and 4.5, as expected - this is CDS method\n- on the other hand, looking directly at `AIRS-CH0-axis0-h` suggest that integration times are 0.1 and 4.7 respecitvely:\n\n```py\na = df['AIRS-CH0-axis0-h']\n(a[1]-a[0]) * 3600 # gives: np.float64(0.1)\n(a[4]-a[3]) * 3600 # gives: np.float64(4.7)\n(a[6]-a[5]) * 3600 # gives: np.float64(4.699999999999999)\n(a[10000] - a[9999]) * 3600 # gives np.float64(4.700000000000415)\n```\n\n@lorenzomugnai Could you help understanding  that? Shall we go with (4.5-0.1) or with (4.7 - 0.1) when removing dark current after CDS?"
  }
}