{
  "id": 68853,
  "title": "Download interrupted, kaggle cli does not resume after all?",
  "url": "/competitions/airbus-ship-detection/discussion/68853",
  "author_name": "",
  "post_date": "2018-10-17T21:29:58.448936200Z",
  "votes": 3,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hi,</p>\n\n<p>I was downloading train_v2.zip, and my computer crashed around 7.3GB (this netbook never crashed before... I believe it is unrelated to downloading the dataset)</p>\n\n<p>To be safe I backed up the partial download, then tried downloading again, remembering I verified reading that the kaggle cli resumes downloads... except apparently it doesn't? it notices the previous file and instructs me to -force redownload...</p>\n\n<p>Then I remember reading about wget with exported cookies, but none of those methods seem to work anymore (the downloaded file is saved as train_v2.zip.1 and contains a kaggle html page filled with javascript...).</p>\n\n<p>Can the kaggle cli be fixed to resume downloads? Or sane instructions for downloading be provided? Or must I redownload the whole dataset and repeatedly pray no interruption occurs?</p>\n\n<p>Thanks in advane,\nLudwig</p>",
  "messages": [
    {
      "id": "405657",
      "postDate": "10/17/2018 21:29:58",
      "content": "<p>Hi,</p>\n\n<p>I was downloading train_v2.zip, and my computer crashed around 7.3GB (this netbook never crashed before... I believe it is unrelated to downloading the dataset)</p>\n\n<p>To be safe I backed up the partial download, then tried downloading again, remembering I verified reading that the kaggle cli resumes downloads... except apparently it doesn't? it notices the previous file and instructs me to -force redownload...</p>\n\n<p>Then I remember reading about wget with exported cookies, but none of those methods seem to work anymore (the downloaded file is saved as train_v2.zip.1 and contains a kaggle html page filled with javascript...).</p>\n\n<p>Can the kaggle cli be fixed to resume downloads? Or sane instructions for downloading be provided? Or must I redownload the whole dataset and repeatedly pray no interruption occurs?</p>\n\n<p>Thanks in advane,\nLudwig</p>",
      "rawMarkdown": "Hi,\n\nI was downloading train_v2.zip, and my computer crashed around 7.3GB (this netbook never crashed before... I believe it is unrelated to downloading the dataset)\n\nTo be safe I backed up the partial download, then tried downloading again, remembering I verified reading that the kaggle cli resumes downloads... except apparently it doesn't? it notices the previous file and instructs me to -force redownload...\n\nThen I remember reading about wget with exported cookies, but none of those methods seem to work anymore (the downloaded file is saved as train_v2.zip.1 and contains a kaggle html page filled with javascript...).\n\nCan the kaggle cli be fixed to resume downloads? Or sane instructions for downloading be provided? Or must I redownload the whole dataset and repeatedly pray no interruption occurs?\n\nThanks in advane,\nLudwig",
      "votes": null
    },
    {
      "id": "405676",
      "postDate": "10/17/2018 22:30:24",
      "content": "<p>An alternative that would certainly satisfy me is if the files were smaller (say 1GB) zips i.e. train_v2p1.zip, train_v2p2.zip ... train_v2p27.zip, then those with a fast and reliable connection are not substantially obstructed, while those with a home connection can decide to compete... the more the merrier?</p>",
      "rawMarkdown": "An alternative that would certainly satisfy me is if the files were smaller (say 1GB) zips i.e. train_v2p1.zip, train_v2p2.zip ... train_v2p27.zip, then those with a fast and reliable connection are not substantially obstructed, while those with a home connection can decide to compete... the more the merrier?",
      "votes": null
    },
    {
      "id": "405841",
      "postDate": "10/18/2018 07:44:53",
      "content": "<p>I finally got kaggle-cli to work, <a href=\"https://github.com/floydwch/kaggle-cli\">https://github.com/floydwch/kaggle-cli</a> after (re) installing python-yaml, cliff, colorama</p>\n\n<p>Now it is resuming from 26%</p>",
      "rawMarkdown": "I finally got kaggle-cli to work, https://github.com/floydwch/kaggle-cli after (re) installing python-yaml, cliff, colorama\n\nNow it is resuming from 26%",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 405676,
      "author_name": "ludwigmaes",
      "author_url": "",
      "post_date": "10/17/2018 22:30:24",
      "content": "<p>An alternative that would certainly satisfy me is if the files were smaller (say 1GB) zips i.e. train_v2p1.zip, train_v2p2.zip ... train_v2p27.zip, then those with a fast and reliable connection are not substantially obstructed, while those with a home connection can decide to compete... the more the merrier?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 405841,
      "author_name": "ludwigmaes",
      "author_url": "",
      "post_date": "10/18/2018 07:44:53",
      "content": "<p>I finally got kaggle-cli to work, <a href=\"https://github.com/floydwch/kaggle-cli\">https://github.com/floydwch/kaggle-cli</a> after (re) installing python-yaml, cliff, colorama</p>\n\n<p>Now it is resuming from 26%</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "405657": "Hi,\n\nI was downloading train_v2.zip, and my computer crashed around 7.3GB (this netbook never crashed before... I believe it is unrelated to downloading the dataset)\n\nTo be safe I backed up the partial download, then tried downloading again, remembering I verified reading that the kaggle cli resumes downloads... except apparently it doesn't? it notices the previous file and instructs me to -force redownload...\n\nThen I remember reading about wget with exported cookies, but none of those methods seem to work anymore (the downloaded file is saved as train_v2.zip.1 and contains a kaggle html page filled with javascript...).\n\nCan the kaggle cli be fixed to resume downloads? Or sane instructions for downloading be provided? Or must I redownload the whole dataset and repeatedly pray no interruption occurs?\n\nThanks in advane,\nLudwig",
    "405676": "An alternative that would certainly satisfy me is if the files were smaller (say 1GB) zips i.e. train_v2p1.zip, train_v2p2.zip ... train_v2p27.zip, then those with a fast and reliable connection are not substantially obstructed, while those with a home connection can decide to compete... the more the merrier?",
    "405841": "I finally got kaggle-cli to work, https://github.com/floydwch/kaggle-cli after (re) installing python-yaml, cliff, colorama\n\nNow it is resuming from 26%"
  },
  "source": "meta"
}