{
  "id": 182283,
  "title": "Downloading dataset using Kaggle API",
  "url": "/competitions/osic-pulmonary-fibrosis-progression/discussion/182283",
  "author_name": "",
  "post_date": "2020-09-12T04:15:43.735114400Z",
  "votes": 1,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I am not sure if am the only one experiencing this, but it seems that when I download using the Kaggle API, I can only get  handful of participants (about 20+) and not the full list (176 participants). Attached is the list as returned by  <code>kaggle competitions files</code> command in the API.</p>\n<p>Any help with this is appreciated! I'm planning to migrate some of my work in colab for pre-training.</p>",
  "messages": [
    {
      "id": "1007307",
      "postDate": "09/12/2020 04:15:43",
      "content": "<p>I am not sure if am the only one experiencing this, but it seems that when I download using the Kaggle API, I can only get  handful of participants (about 20+) and not the full list (176 participants). Attached is the list as returned by  <code>kaggle competitions files</code> command in the API.</p>\n<p>Any help with this is appreciated! I'm planning to migrate some of my work in colab for pre-training.</p>",
      "rawMarkdown": "I am not sure if am the only one experiencing this, but it seems that when I download using the Kaggle API, I can only get  handful of participants (about 20+) and not the full list (176 participants). Attached is the list as returned by  `kaggle competitions files` command in the API.\n\nAny help with this is appreciated! I'm planning to migrate some of my work in colab for pre-training.",
      "votes": null
    },
    {
      "id": "1007462",
      "postDate": "09/12/2020 07:20:54",
      "content": "<p>I am not sure of list of participants but if you want to shift to colab and do the training( pre-training or whatever), download your json file from your kaggle account.</p>\n<ul>\n<li>create a .kaggle directory like this:<br>\n<code>\n! mkdir ~/.kaggle\n</code></li>\n<li>Store your json file in some directory in drive. It reduces the pain of uploading the json file every time.  <strong>Copy</strong> (don't move using <em>mv</em> command) the json file inside the newly created folder as below:<br>\n<code>\n! cp /content/drive/'My Drive'/kaggle/kaggle.json ~/.kaggle/\n</code></li>\n<li>Allow permission for kaggle api :<br>\n<code>! chmod 600 ~/.kaggle/kaggle.json</code></li>\n</ul>\n<p>The api is set and ready to go. <br>\nHope this was helpful</p>",
      "rawMarkdown": "I am not sure of list of participants but if you want to shift to colab and do the training( pre-training or whatever), download your json file from your kaggle account.\n- create a .kaggle directory like this:\n`\n! mkdir ~/.kaggle\n`\n- Store your json file in some directory in drive. It reduces the pain of uploading the json file every time.  **Copy** (don't move using *mv* command) the json file inside the newly created folder as below:\n`\n! cp /content/drive/'My Drive'/kaggle/kaggle.json ~/.kaggle/\n`\n- Allow permission for kaggle api :\n`! chmod 600 ~/.kaggle/kaggle.json`\n\nThe api is set and ready to go. \nHope this was helpful",
      "votes": null
    },
    {
      "id": "1007500",
      "postDate": "09/12/2020 08:02:11",
      "content": "<p>Thanks for the reply! I have some knowledge about the procedures, but I'm struggling to get the full extent of data. I can only get a fraction of it through the API.</p>",
      "rawMarkdown": "Thanks for the reply! I have some knowledge about the procedures, but I'm struggling to get the full extent of data. I can only get a fraction of it through the API.",
      "votes": null
    },
    {
      "id": "1007707",
      "postDate": "09/12/2020 12:09:36",
      "content": "<p>I faced a similar issue, first check that you have the latest version of kaggle API installed. I had some problems with colab but didn't face any issue on was, also you can use python API which is useful. I have a stupid hack that I use:</p>\n<pre><code>import os\nos.system(\"kaggle command\")\n</code></pre>",
      "rawMarkdown": "I faced a similar issue, first check that you have the latest version of kaggle API installed. I had some problems with colab but didn't face any issue on was, also you can use python API which is useful. I have a stupid hack that I use:\n```\nimport os\nos.system(\"kaggle command\")\n```",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1007462,
      "author_name": "databeginner",
      "author_url": "",
      "post_date": "09/12/2020 07:20:54",
      "content": "<p>I am not sure of list of participants but if you want to shift to colab and do the training( pre-training or whatever), download your json file from your kaggle account.</p>\n<ul>\n<li>create a .kaggle directory like this:<br>\n<code>\n! mkdir ~/.kaggle\n</code></li>\n<li>Store your json file in some directory in drive. It reduces the pain of uploading the json file every time.  <strong>Copy</strong> (don't move using <em>mv</em> command) the json file inside the newly created folder as below:<br>\n<code>\n! cp /content/drive/'My Drive'/kaggle/kaggle.json ~/.kaggle/\n</code></li>\n<li>Allow permission for kaggle api :<br>\n<code>! chmod 600 ~/.kaggle/kaggle.json</code></li>\n</ul>\n<p>The api is set and ready to go. <br>\nHope this was helpful</p>",
      "votes": null,
      "replies": [
        {
          "id": 1007500,
          "author_name": "kitangladmatu",
          "author_url": "",
          "post_date": "09/12/2020 08:02:11",
          "content": "<p>Thanks for the reply! I have some knowledge about the procedures, but I'm struggling to get the full extent of data. I can only get a fraction of it through the API.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1007707,
      "author_name": "furcifer",
      "author_url": "",
      "post_date": "09/12/2020 12:09:36",
      "content": "<p>I faced a similar issue, first check that you have the latest version of kaggle API installed. I had some problems with colab but didn't face any issue on was, also you can use python API which is useful. I have a stupid hack that I use:</p>\n<pre><code>import os\nos.system(\"kaggle command\")\n</code></pre>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1007307": "I am not sure if am the only one experiencing this, but it seems that when I download using the Kaggle API, I can only get  handful of participants (about 20+) and not the full list (176 participants). Attached is the list as returned by  `kaggle competitions files` command in the API.\n\nAny help with this is appreciated! I'm planning to migrate some of my work in colab for pre-training.",
    "1007462": "I am not sure of list of participants but if you want to shift to colab and do the training( pre-training or whatever), download your json file from your kaggle account.\n- create a .kaggle directory like this:\n`\n! mkdir ~/.kaggle\n`\n- Store your json file in some directory in drive. It reduces the pain of uploading the json file every time.  **Copy** (don't move using *mv* command) the json file inside the newly created folder as below:\n`\n! cp /content/drive/'My Drive'/kaggle/kaggle.json ~/.kaggle/\n`\n- Allow permission for kaggle api :\n`! chmod 600 ~/.kaggle/kaggle.json`\n\nThe api is set and ready to go. \nHope this was helpful",
    "1007500": "Thanks for the reply! I have some knowledge about the procedures, but I'm struggling to get the full extent of data. I can only get a fraction of it through the API.",
    "1007707": "I faced a similar issue, first check that you have the latest version of kaggle API installed. I had some problems with colab but didn't face any issue on was, also you can use python API which is useful. I have a stupid hack that I use:\n```\nimport os\nos.system(\"kaggle command\")\n```"
  },
  "source": "meta"
}