{
  "id": 417497,
  "title": "Data description vs Preprint",
  "url": "/competitions/google-research-identify-contrails-reduce-global-warming/discussion/417497",
  "author_name": "ticadumi",
  "post_date": "2023-06-16T00:35:05.017000",
  "votes": 17,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Thank you for organizing this competition and the detail description of the dataset!<br>\nI have two clarifying questions about the dataset, crosschecking the information in the provided <a href=\"https://arxiv.org/pdf/2304.02122.pdf\" target=\"_blank\">preprint</a> with that in the Kaggle data section (and theassociated <a href=\"https://www.kaggle.com/code/inversion/visualizing-contrails\" target=\"_blank\">notebook</a>):</p>\n<p>Q1. What is the correct value for \"n_times_before\"?</p>\n<ul>\n<li>Preprint reads: \"we show labelers <strong>5 images (50 minutes) before and 2\nimages (20 minutes) after</strong> the image being labeled\"</li>\n<li>Kaggle data section reads: \"all examples have <strong>n_times_before=4 and n_times_after=3</strong>\"</li>\n</ul>\n<p>Q2. What is the correct false-color formula?</p>\n<ul>\n<li>Preprint reads: \" The <strong>red, blue and green</strong> channels are represented by the <strong>12µm, difference between 12µm and 11µm, and difference between 11µm and 8µm respectively</strong>.\"</li>\n<li>Kaggle data section notebook seems to have <strong>red as difference between 12µm and 11µm, green as difference between 11µm and 8µm, and blue as 11µm</strong>:</li>\n</ul>\n<pre><code>r = normalize_range(band15 - band14, _TDIFF_BOUNDS)\ng = normalize_range(band14 - band11, _CLOUD_TOP_TDIFF_BOUNDS)\nb = normalize_range(band14, _T11_BOUNDS)\n</code></pre>\n<p>In other words:</p>\n<ul>\n<li>kaggle red matches preprint blue</li>\n<li>kaggle green is preprint green (ok)</li>\n<li>kaggle blue matches preprint red as self-contained value, except that value is 11µm in kaggle vs 12µm in preprint</li>\n</ul>\n<p>Thank you!</p>",
  "messages": [
    {
      "id": 2304376,
      "postDate": "2023-06-16T00:35:05.017Z",
      "content": "<p>Thank you for organizing this competition and the detail description of the dataset!<br>\nI have two clarifying questions about the dataset, crosschecking the information in the provided <a href=\"https://arxiv.org/pdf/2304.02122.pdf\" target=\"_blank\">preprint</a> with that in the Kaggle data section (and theassociated <a href=\"https://www.kaggle.com/code/inversion/visualizing-contrails\" target=\"_blank\">notebook</a>):</p>\n<p>Q1. What is the correct value for \"n_times_before\"?</p>\n<ul>\n<li>Preprint reads: \"we show labelers <strong>5 images (50 minutes) before and 2\nimages (20 minutes) after</strong> the image being labeled\"</li>\n<li>Kaggle data section reads: \"all examples have <strong>n_times_before=4 and n_times_after=3</strong>\"</li>\n</ul>\n<p>Q2. What is the correct false-color formula?</p>\n<ul>\n<li>Preprint reads: \" The <strong>red, blue and green</strong> channels are represented by the <strong>12µm, difference between 12µm and 11µm, and difference between 11µm and 8µm respectively</strong>.\"</li>\n<li>Kaggle data section notebook seems to have <strong>red as difference between 12µm and 11µm, green as difference between 11µm and 8µm, and blue as 11µm</strong>:</li>\n</ul>\n<pre><code>r = normalize_range(band15 - band14, _TDIFF_BOUNDS)\ng = normalize_range(band14 - band11, _CLOUD_TOP_TDIFF_BOUNDS)\nb = normalize_range(band14, _T11_BOUNDS)\n</code></pre>\n<p>In other words:</p>\n<ul>\n<li>kaggle red matches preprint blue</li>\n<li>kaggle green is preprint green (ok)</li>\n<li>kaggle blue matches preprint red as self-contained value, except that value is 11µm in kaggle vs 12µm in preprint</li>\n</ul>\n<p>Thank you!</p>",
      "rawMarkdown": "Thank you for organizing this competition and the detail description of the dataset!\nI have two clarifying questions about the dataset, crosschecking the information in the provided [preprint](https://arxiv.org/pdf/2304.02122.pdf) with that in the Kaggle data section (and theassociated [notebook](https://www.kaggle.com/code/inversion/visualizing-contrails)):\n\nQ1. What is the correct value for \"n_times_before\"?\n- Preprint reads: \"we show labelers **5 images (50 minutes) before and 2\nimages (20 minutes) after** the image being labeled\"\n- Kaggle data section reads: \"all examples have **n_times_before=4 and n_times_after=3**\"\n\nQ2. What is the correct false-color formula?\n- Preprint reads: \" The **red, blue and green** channels are represented by the **12µm, difference between 12µm and 11µm, and difference between 11µm and 8µm respectively**.\"\n- Kaggle data section notebook seems to have **red as difference between 12µm and 11µm, green as difference between 11µm and 8µm, and blue as 11µm**:\n```python\nr = normalize_range(band15 - band14, _TDIFF_BOUNDS)\ng = normalize_range(band14 - band11, _CLOUD_TOP_TDIFF_BOUNDS)\nb = normalize_range(band14, _T11_BOUNDS)\n```\nIn other words:\n* kaggle red matches preprint blue\n* kaggle green is preprint green (ok)\n* kaggle blue matches preprint red as self-contained value, except that value is 11µm in kaggle vs 12µm in preprint\n\nThank you!",
      "votes": 16
    },
    {
      "id": 2305255,
      "postDate": "2023-06-16T14:34:07.540Z",
      "content": "<p>Thanks for catching these!</p>\n<p>For Q2. I believe that the description in the data section of this competition is correct and the preprint is incorrect. We'll fix this in the next draft of the preprint.</p>\n<p>For Q1, I'll have to confirm with coauthors, but it is possible that both are correct. Since the labelers only labeled a single frame, it's possible that we showed them 5 frames before and 2 after, but for this competition we provide 4 frames before and 3 after. I don't know what would have driven them to be different, but they could both be correct.</p>",
      "rawMarkdown": "Thanks for catching these!\n\nFor Q2. I believe that the description in the data section of this competition is correct and the preprint is incorrect. We'll fix this in the next draft of the preprint.\n\nFor Q1, I'll have to confirm with coauthors, but it is possible that both are correct. Since the labelers only labeled a single frame, it's possible that we showed them 5 frames before and 2 after, but for this competition we provide 4 frames before and 3 after. I don't know what would have driven them to be different, but they could both be correct.",
      "votes": 10,
      "replies": [
        {
          "id": 2305349,
          "postDate": "2023-06-16T16:06:59.830Z",
          "content": "<p>Thank you.</p>",
          "rawMarkdown": "Thank you."
        }
      ]
    }
  ],
  "comments": [
    {
      "id": 2305255,
      "author_name": "Aaron Sarna",
      "author_url": "",
      "post_date": "2023-06-16T14:34:07.540000",
      "content": "<p>Thanks for catching these!</p>\n<p>For Q2. I believe that the description in the data section of this competition is correct and the preprint is incorrect. We'll fix this in the next draft of the preprint.</p>\n<p>For Q1, I'll have to confirm with coauthors, but it is possible that both are correct. Since the labelers only labeled a single frame, it's possible that we showed them 5 frames before and 2 after, but for this competition we provide 4 frames before and 3 after. I don't know what would have driven them to be different, but they could both be correct.</p>",
      "votes": 10,
      "replies": [
        {
          "id": 2305349,
          "author_name": "ticadumi",
          "author_url": "",
          "post_date": "2023-06-16T16:06:59.830000",
          "content": "<p>Thank you.</p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2304376": "Thank you for organizing this competition and the detail description of the dataset!\nI have two clarifying questions about the dataset, crosschecking the information in the provided [preprint](https://arxiv.org/pdf/2304.02122.pdf) with that in the Kaggle data section (and theassociated [notebook](https://www.kaggle.com/code/inversion/visualizing-contrails)):\n\nQ1. What is the correct value for \"n_times_before\"?\n- Preprint reads: \"we show labelers **5 images (50 minutes) before and 2\nimages (20 minutes) after** the image being labeled\"\n- Kaggle data section reads: \"all examples have **n_times_before=4 and n_times_after=3**\"\n\nQ2. What is the correct false-color formula?\n- Preprint reads: \" The **red, blue and green** channels are represented by the **12µm, difference between 12µm and 11µm, and difference between 11µm and 8µm respectively**.\"\n- Kaggle data section notebook seems to have **red as difference between 12µm and 11µm, green as difference between 11µm and 8µm, and blue as 11µm**:\n```python\nr = normalize_range(band15 - band14, _TDIFF_BOUNDS)\ng = normalize_range(band14 - band11, _CLOUD_TOP_TDIFF_BOUNDS)\nb = normalize_range(band14, _T11_BOUNDS)\n```\nIn other words:\n* kaggle red matches preprint blue\n* kaggle green is preprint green (ok)\n* kaggle blue matches preprint red as self-contained value, except that value is 11µm in kaggle vs 12µm in preprint\n\nThank you!",
    "2305255": "Thanks for catching these!\n\nFor Q2. I believe that the description in the data section of this competition is correct and the preprint is incorrect. We'll fix this in the next draft of the preprint.\n\nFor Q1, I'll have to confirm with coauthors, but it is possible that both are correct. Since the labelers only labeled a single frame, it's possible that we showed them 5 frames before and 2 after, but for this competition we provide 4 frames before and 3 after. I don't know what would have driven them to be different, but they could both be correct."
  }
}