{
  "id": 573167,
  "title": "Question about unzipping a file with two PO time series",
  "url": "/competitions/geolifeclef-2025/discussion/573167",
  "author_name": "",
  "post_date": "2025-04-14T04:03:28.897980500Z",
  "votes": 2,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I'm trying to train the model with PO data, but I'm running into a problem in that the PO data in zip form is very hard to unpack right now, and it took me seven days to unpack less than 40% of it, and especially after I got to 30% the unpacking slowed down significantly. I wonder if I should consider re-uploading it using tar.gz or tar format, the files will unzip faster?</p>",
  "messages": [
    {
      "id": "3178345",
      "postDate": "04/14/2025 04:03:28",
      "content": "<p>I'm trying to train the model with PO data, but I'm running into a problem in that the PO data in zip form is very hard to unpack right now, and it took me seven days to unpack less than 40% of it, and especially after I got to 30% the unpacking slowed down significantly. I wonder if I should consider re-uploading it using tar.gz or tar format, the files will unzip faster?</p>",
      "rawMarkdown": "I'm trying to train the model with PO data, but I'm running into a problem in that the PO data in zip form is very hard to unpack right now, and it took me seven days to unpack less than 40% of it, and especially after I got to 30% the unpacking slowed down significantly. I wonder if I should consider re-uploading it using tar.gz or tar format, the files will unzip faster?",
      "votes": null
    },
    {
      "id": "3178347",
      "postDate": "04/14/2025 04:07:38",
      "content": "<p>The unzip speed of PO train aka satellite imagery is OK, but the decompression speed of PO bioclimate and satellite time series in cube form is very slow, which I think may also be related to the internal directory structure. I sincerely hope that the organizer can notice this problem and re-upload a tar format zip file.</p>",
      "rawMarkdown": "The unzip speed of PO train aka satellite imagery is OK, but the decompression speed of PO bioclimate and satellite time series in cube form is very slow, which I think may also be related to the internal directory structure. I sincerely hope that the organizer can notice this problem and re-upload a tar format zip file.",
      "votes": null
    },
    {
      "id": "3178996",
      "postDate": "04/14/2025 21:36:08",
      "content": "<p>Dear <a href=\"https://www.kaggle.com/liuhaixu\" target=\"_blank\">@liuhaixu</a>,</p>\n<p>Thank you for raising this. It is most likely some limitation of your filesystem.<br>\nHowever, to help you and potentially others as well, we will restructure the files in the archives.</p>\n<p>At the moment, you can find the new Bioclim Time-series data on Seafile in an archive named <code>GLC25-PO-train-bioclimatic_monthly-new.zip</code>. The Landsat should be there in around 2 hours.</p>\n<p>Please let me know if it helped!</p>\n<p>Best,<br>\nLukas</p>",
      "rawMarkdown": "Dear @liuhaixu,\n\nThank you for raising this. It is most likely some limitation of your filesystem.\nHowever, to help you and potentially others as well, we will restructure the files in the archives.\n\nAt the moment, you can find the new Bioclim Time-series data on Seafile in an archive named `GLC25-PO-train-bioclimatic_monthly-new.zip`. The Landsat should be there in around 2 hours.\n\nPlease let me know if it helped!\n\nBest,\nLukas",
      "votes": null
    },
    {
      "id": "3179051",
      "postDate": "04/15/2025 01:02:51",
      "content": "<p>Dear <a href=\"https://www.kaggle.com/liuhaixu\" target=\"_blank\">@liuhaixu</a>,</p>\n<p>I'm back with good news. The second archive, i.e., <code>GLC25-PO-train-landsat_time_series.zip</code>, was just uploaded to SeaFile.<br>\nI also tested the extraction, and it took around 30 minutes on my local machine.</p>\n<p>Please let me know if it helped!</p>\n<p>Best,<br>\nLukas</p>",
      "rawMarkdown": "Dear @liuhaixu,\n\nI'm back with good news. The second archive, i.e., `GLC25-PO-train-landsat_time_series.zip`, was just uploaded to SeaFile.\nI also tested the extraction, and it took around 30 minutes on my local machine.\n\nPlease let me know if it helped!\n\nBest,\nLukas",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3178347,
      "author_name": "liuhaixu",
      "author_url": "",
      "post_date": "04/14/2025 04:07:38",
      "content": "<p>The unzip speed of PO train aka satellite imagery is OK, but the decompression speed of PO bioclimate and satellite time series in cube form is very slow, which I think may also be related to the internal directory structure. I sincerely hope that the organizer can notice this problem and re-upload a tar format zip file.</p>",
      "votes": null,
      "replies": [
        {
          "id": 3179051,
          "author_name": "picekl",
          "author_url": "",
          "post_date": "04/15/2025 01:02:51",
          "content": "<p>Dear <a href=\"https://www.kaggle.com/liuhaixu\" target=\"_blank\">@liuhaixu</a>,</p>\n<p>I'm back with good news. The second archive, i.e., <code>GLC25-PO-train-landsat_time_series.zip</code>, was just uploaded to SeaFile.<br>\nI also tested the extraction, and it took around 30 minutes on my local machine.</p>\n<p>Please let me know if it helped!</p>\n<p>Best,<br>\nLukas</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 3178996,
      "author_name": "picekl",
      "author_url": "",
      "post_date": "04/14/2025 21:36:08",
      "content": "<p>Dear <a href=\"https://www.kaggle.com/liuhaixu\" target=\"_blank\">@liuhaixu</a>,</p>\n<p>Thank you for raising this. It is most likely some limitation of your filesystem.<br>\nHowever, to help you and potentially others as well, we will restructure the files in the archives.</p>\n<p>At the moment, you can find the new Bioclim Time-series data on Seafile in an archive named <code>GLC25-PO-train-bioclimatic_monthly-new.zip</code>. The Landsat should be there in around 2 hours.</p>\n<p>Please let me know if it helped!</p>\n<p>Best,<br>\nLukas</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3178345": "I'm trying to train the model with PO data, but I'm running into a problem in that the PO data in zip form is very hard to unpack right now, and it took me seven days to unpack less than 40% of it, and especially after I got to 30% the unpacking slowed down significantly. I wonder if I should consider re-uploading it using tar.gz or tar format, the files will unzip faster?",
    "3178347": "The unzip speed of PO train aka satellite imagery is OK, but the decompression speed of PO bioclimate and satellite time series in cube form is very slow, which I think may also be related to the internal directory structure. I sincerely hope that the organizer can notice this problem and re-upload a tar format zip file.",
    "3178996": "Dear @liuhaixu,\n\nThank you for raising this. It is most likely some limitation of your filesystem.\nHowever, to help you and potentially others as well, we will restructure the files in the archives.\n\nAt the moment, you can find the new Bioclim Time-series data on Seafile in an archive named `GLC25-PO-train-bioclimatic_monthly-new.zip`. The Landsat should be there in around 2 hours.\n\nPlease let me know if it helped!\n\nBest,\nLukas",
    "3179051": "Dear @liuhaixu,\n\nI'm back with good news. The second archive, i.e., `GLC25-PO-train-landsat_time_series.zip`, was just uploaded to SeaFile.\nI also tested the extraction, and it took around 30 minutes on my local machine.\n\nPlease let me know if it helped!\n\nBest,\nLukas"
  },
  "source": "meta"
}