{
  "id": 72578,
  "title": "Separate script files",
  "url": "/competitions/quora-insincere-questions-classification/discussion/72578",
  "author_name": "",
  "post_date": "2018-11-24T20:04:59.977532300Z",
  "votes": 3,
  "comment_count": 6,
  "views": 0,
  "content": "<p>Can I store my code in separate .py files and use them in the kernel or does that count as external data?</p>",
  "messages": [
    {
      "id": "427178",
      "postDate": "11/24/2018 20:04:59",
      "content": "<p>Can I store my code in separate .py files and use them in the kernel or does that count as external data?</p>",
      "rawMarkdown": "Can I store my code in separate .py files and use them in the kernel or does that count as external data?",
      "votes": null
    },
    {
      "id": "427253",
      "postDate": "11/25/2018 00:41:12",
      "content": "<p>Pretty sure it counts as external data.\nJust copy/paste the code in your kernel and you'll be fine.</p>",
      "rawMarkdown": "Pretty sure it counts as external data.\nJust copy/paste the code in your kernel and you'll be fine.",
      "votes": null
    },
    {
      "id": "427620",
      "postDate": "11/25/2018 21:29:42",
      "content": "<p>Anything that you upload to Kaggle counts as external data. </p>",
      "rawMarkdown": "Anything that you upload to Kaggle counts as external data.",
      "votes": null
    },
    {
      "id": "427659",
      "postDate": "11/25/2018 23:16:50",
      "content": "<p>That's what I thought. Thanks, guys!</p>",
      "rawMarkdown": "That's what I thought. Thanks, guys!",
      "votes": null
    },
    {
      "id": "427698",
      "postDate": "11/26/2018 01:02:52",
      "content": "<p><a href=\"/hhstrand\">@hhstrand</a>, nice to see you here. It has been a long time since toxic comments competition :-)</p>\n\n<p>The challenge with this type of Kaggle only competition is the time limit. This is my 2nd one of this type and I failed miserably in the 1st one i.e. \"<a href=\"https://www.kaggle.com/c/mercari-price-suggestion-challenge\">Mercari Price Suggestion Challenge</a>\" competition . In Mercari, both my models failed the time test. So time is very important here, you may look at <a href=\"https://www.kaggle.com/c/mercari-price-suggestion-challenge/discussion\">Mercari solutions</a> for inspiration.</p>",
      "rawMarkdown": "hhstrand, nice to see you here. It has been a long time since toxic comments competition :-)\n\nThe challenge with this type of Kaggle only competition is the time limit. This is my 2nd one of this type and I failed miserably in the 1st one i.e. \"[Mercari Price Suggestion Challenge][1]\" competition . In Mercari, both my models failed the time test. So time is very important here, you may look at [Mercari solutions][2] for inspiration.\n\n\n  [1]: https://www.kaggle.com/c/mercari-price-suggestion-challenge\n  [2]: https://www.kaggle.com/c/mercari-price-suggestion-challenge/discussion",
      "votes": null
    },
    {
      "id": "428182",
      "postDate": "11/26/2018 22:05:08",
      "content": "<p>Thanks! I’m hoping to find some time for this one, being a Quora addict and all.</p>\n\n<p>I like the idea of kernels only. It’s very democratic.</p>\n\n<p>Good luck to you in the competition! I see you’ve gone off to a good start!</p>",
      "rawMarkdown": "Thanks! I’m hoping to find some time for this one, being a Quora addict and all.\n\nI like the idea of kernels only. It’s very democratic.\n\nGood luck to you in the competition! I see you’ve gone off to a good start!",
      "votes": null
    },
    {
      "id": "428203",
      "postDate": "11/26/2018 23:05:55",
      "content": "<p>I agree, it democratic as the HW resource variance is eliminated. Thanks, I am doing okay for now but we have 2 months left. So far I am only using a tweaked  versions of my models from Toxic comment.</p>\n\n<p>The problem of inconsistencies in results is really a big one here. In fact there is inconsistencies in time taken with the exact same model. I just ran another one of my models twice, where I tried a few tricks of fixing the seeds. The 1st run finished within 2 hrs whilst the 2nd run took 128 minutes with no changes. i really do not understand what is happening here.</p>",
      "rawMarkdown": "I agree, it democratic as the HW resource variance is eliminated. Thanks, I am doing okay for now but we have 2 months left. So far I am only using a tweaked  versions of my models from Toxic comment.\n\nThe problem of inconsistencies in results is really a big one here. In fact there is inconsistencies in time taken with the exact same model. I just ran another one of my models twice, where I tried a few tricks of fixing the seeds. The 1st run finished within 2 hrs whilst the 2nd run took 128 minutes with no changes. i really do not understand what is happening here.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 427253,
      "author_name": "theoviel",
      "author_url": "",
      "post_date": "11/25/2018 00:41:12",
      "content": "<p>Pretty sure it counts as external data.\nJust copy/paste the code in your kernel and you'll be fine.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 427620,
      "author_name": "tunguz",
      "author_url": "",
      "post_date": "11/25/2018 21:29:42",
      "content": "<p>Anything that you upload to Kaggle counts as external data. </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 427659,
      "author_name": "hhstrand",
      "author_url": "",
      "post_date": "11/25/2018 23:16:50",
      "content": "<p>That's what I thought. Thanks, guys!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 427698,
      "author_name": "sheriytm",
      "author_url": "",
      "post_date": "11/26/2018 01:02:52",
      "content": "<p><a href=\"/hhstrand\">@hhstrand</a>, nice to see you here. It has been a long time since toxic comments competition :-)</p>\n\n<p>The challenge with this type of Kaggle only competition is the time limit. This is my 2nd one of this type and I failed miserably in the 1st one i.e. \"<a href=\"https://www.kaggle.com/c/mercari-price-suggestion-challenge\">Mercari Price Suggestion Challenge</a>\" competition . In Mercari, both my models failed the time test. So time is very important here, you may look at <a href=\"https://www.kaggle.com/c/mercari-price-suggestion-challenge/discussion\">Mercari solutions</a> for inspiration.</p>",
      "votes": null,
      "replies": [
        {
          "id": 428182,
          "author_name": "hhstrand",
          "author_url": "",
          "post_date": "11/26/2018 22:05:08",
          "content": "<p>Thanks! I’m hoping to find some time for this one, being a Quora addict and all.</p>\n\n<p>I like the idea of kernels only. It’s very democratic.</p>\n\n<p>Good luck to you in the competition! I see you’ve gone off to a good start!</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 428203,
          "author_name": "sheriytm",
          "author_url": "",
          "post_date": "11/26/2018 23:05:55",
          "content": "<p>I agree, it democratic as the HW resource variance is eliminated. Thanks, I am doing okay for now but we have 2 months left. So far I am only using a tweaked  versions of my models from Toxic comment.</p>\n\n<p>The problem of inconsistencies in results is really a big one here. In fact there is inconsistencies in time taken with the exact same model. I just ran another one of my models twice, where I tried a few tricks of fixing the seeds. The 1st run finished within 2 hrs whilst the 2nd run took 128 minutes with no changes. i really do not understand what is happening here.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "427178": "Can I store my code in separate .py files and use them in the kernel or does that count as external data?",
    "427253": "Pretty sure it counts as external data.\nJust copy/paste the code in your kernel and you'll be fine.",
    "427620": "Anything that you upload to Kaggle counts as external data.",
    "427659": "That's what I thought. Thanks, guys!",
    "427698": "hhstrand, nice to see you here. It has been a long time since toxic comments competition :-)\n\nThe challenge with this type of Kaggle only competition is the time limit. This is my 2nd one of this type and I failed miserably in the 1st one i.e. \"[Mercari Price Suggestion Challenge][1]\" competition . In Mercari, both my models failed the time test. So time is very important here, you may look at [Mercari solutions][2] for inspiration.\n\n\n  [1]: https://www.kaggle.com/c/mercari-price-suggestion-challenge\n  [2]: https://www.kaggle.com/c/mercari-price-suggestion-challenge/discussion",
    "428182": "Thanks! I’m hoping to find some time for this one, being a Quora addict and all.\n\nI like the idea of kernels only. It’s very democratic.\n\nGood luck to you in the competition! I see you’ve gone off to a good start!",
    "428203": "I agree, it democratic as the HW resource variance is eliminated. Thanks, I am doing okay for now but we have 2 months left. So far I am only using a tweaked  versions of my models from Toxic comment.\n\nThe problem of inconsistencies in results is really a big one here. In fact there is inconsistencies in time taken with the exact same model. I just ran another one of my models twice, where I tried a few tricks of fixing the seeds. The 1st run finished within 2 hrs whilst the 2nd run took 128 minutes with no changes. i really do not understand what is happening here."
  },
  "source": "meta"
}