{
  "id": 77678,
  "title": "How to perform model ensembling?",
  "url": "/competitions/quora-insincere-questions-classification/discussion/77678",
  "author_name": "",
  "post_date": "2019-01-15T13:33:22.631445400Z",
  "votes": 3,
  "comment_count": 5,
  "views": 0,
  "content": "<p>How to perform model ensembling on a single kernel, given the time constraints? Any ides would be appreciated.</p>",
  "messages": [
    {
      "id": "456282",
      "postDate": "01/15/2019 13:33:22",
      "content": "<p>How to perform model ensembling on a single kernel, given the time constraints? Any ides would be appreciated.</p>",
      "rawMarkdown": "How to perform model ensembling on a single kernel, given the time constraints? Any ides would be appreciated.",
      "votes": null
    },
    {
      "id": "456327",
      "postDate": "01/15/2019 15:19:29",
      "content": "<p>No magic that I know of. Just train smaller models and ensemble them within the two hours. Plenty of examples in the Kernels</p>",
      "rawMarkdown": "No magic that I know of. Just train smaller models and ensemble them within the two hours. Plenty of examples in the Kernels",
      "votes": null
    },
    {
      "id": "456388",
      "postDate": "01/15/2019 18:07:14",
      "content": "<p>@RatulGhosh. Have a look at a similar competition <a href=\"https://www.kaggle.com/abhishek/approaching-almost-any-nlp-problem-on-kaggle\">https://www.kaggle.com/abhishek/approaching-almost-any-nlp-problem-on-kaggle</a></p>",
      "rawMarkdown": "RatulGhosh. Have a look at a similar competition https://www.kaggle.com/abhishek/approaching-almost-any-nlp-problem-on-kaggle",
      "votes": null
    },
    {
      "id": "456871",
      "postDate": "01/16/2019 17:20:01",
      "content": "<p>@Ratul Ghosh. Have a look at this kernel. You will get to know how to apply\n<a href=\"https://www.kaggle.com/yassineghouzam/titanic-top-4-with-ensemble-modeling\">https://www.kaggle.com/yassineghouzam/titanic-top-4-with-ensemble-modeling</a>\nYou can also watch this video to learn model ensembling.\n<a href=\"https://www.youtube.com/playlist?list=PLvFFyDr-FYmkyik6qssrXFpwbz_LJF_wN\">https://www.youtube.com/playlist?list=PLvFFyDr-FYmkyik6qssrXFpwbz_LJF_wN</a>\nI hope this will help you.</p>",
      "rawMarkdown": "Ratul Ghosh. Have a look at this kernel. You will get to know how to apply\nhttps://www.kaggle.com/yassineghouzam/titanic-top-4-with-ensemble-modeling\nYou can also watch this video to learn model ensembling.\nhttps://www.youtube.com/playlist?list=PLvFFyDr-FYmkyik6qssrXFpwbz_LJF_wN\nI hope this will help you.",
      "votes": null
    },
    {
      "id": "460176",
      "postDate": "01/23/2019 05:59:42",
      "content": "<p>I am contemplating a different way of ensembling, using ranges of confidence. I did not implement it yet and honestly am not too wishful about it but worth a shot.</p>",
      "rawMarkdown": "I am contemplating a different way of ensembling, using ranges of confidence. I did not implement it yet and honestly am not too wishful about it but worth a shot.",
      "votes": null
    },
    {
      "id": "462043",
      "postDate": "01/27/2019 14:39:10",
      "content": "<p>@Ratul ,\nPlease can you check this kernel : </p>\n\n<p><a href=\"https://www.kaggle.com/ashishpatel26/ensemble-learning-for-beginner-to-advance\">https://www.kaggle.com/ashishpatel26/ensemble-learning-for-beginner-to-advance</a></p>\n\n<p>So please let me know if have any query ?\nThanks </p>",
      "rawMarkdown": "Ratul ,\nPlease can you check this kernel : \n\nhttps://www.kaggle.com/ashishpatel26/ensemble-learning-for-beginner-to-advance\n\nSo please let me know if have any query ?\nThanks",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 456327,
      "author_name": "mschumacher",
      "author_url": "",
      "post_date": "01/15/2019 15:19:29",
      "content": "<p>No magic that I know of. Just train smaller models and ensemble them within the two hours. Plenty of examples in the Kernels</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 456388,
      "author_name": "cyberia",
      "author_url": "",
      "post_date": "01/15/2019 18:07:14",
      "content": "<p>@RatulGhosh. Have a look at a similar competition <a href=\"https://www.kaggle.com/abhishek/approaching-almost-any-nlp-problem-on-kaggle\">https://www.kaggle.com/abhishek/approaching-almost-any-nlp-problem-on-kaggle</a></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 456871,
      "author_name": "alokpratap",
      "author_url": "",
      "post_date": "01/16/2019 17:20:01",
      "content": "<p>@Ratul Ghosh. Have a look at this kernel. You will get to know how to apply\n<a href=\"https://www.kaggle.com/yassineghouzam/titanic-top-4-with-ensemble-modeling\">https://www.kaggle.com/yassineghouzam/titanic-top-4-with-ensemble-modeling</a>\nYou can also watch this video to learn model ensembling.\n<a href=\"https://www.youtube.com/playlist?list=PLvFFyDr-FYmkyik6qssrXFpwbz_LJF_wN\">https://www.youtube.com/playlist?list=PLvFFyDr-FYmkyik6qssrXFpwbz_LJF_wN</a>\nI hope this will help you.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 460176,
      "author_name": "isikkuntay",
      "author_url": "",
      "post_date": "01/23/2019 05:59:42",
      "content": "<p>I am contemplating a different way of ensembling, using ranges of confidence. I did not implement it yet and honestly am not too wishful about it but worth a shot.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 462043,
      "author_name": "sunilcube",
      "author_url": "",
      "post_date": "01/27/2019 14:39:10",
      "content": "<p>@Ratul ,\nPlease can you check this kernel : </p>\n\n<p><a href=\"https://www.kaggle.com/ashishpatel26/ensemble-learning-for-beginner-to-advance\">https://www.kaggle.com/ashishpatel26/ensemble-learning-for-beginner-to-advance</a></p>\n\n<p>So please let me know if have any query ?\nThanks </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "456282": "How to perform model ensembling on a single kernel, given the time constraints? Any ides would be appreciated.",
    "456327": "No magic that I know of. Just train smaller models and ensemble them within the two hours. Plenty of examples in the Kernels",
    "456388": "RatulGhosh. Have a look at a similar competition https://www.kaggle.com/abhishek/approaching-almost-any-nlp-problem-on-kaggle",
    "456871": "Ratul Ghosh. Have a look at this kernel. You will get to know how to apply\nhttps://www.kaggle.com/yassineghouzam/titanic-top-4-with-ensemble-modeling\nYou can also watch this video to learn model ensembling.\nhttps://www.youtube.com/playlist?list=PLvFFyDr-FYmkyik6qssrXFpwbz_LJF_wN\nI hope this will help you.",
    "460176": "I am contemplating a different way of ensembling, using ranges of confidence. I did not implement it yet and honestly am not too wishful about it but worth a shot.",
    "462043": "Ratul ,\nPlease can you check this kernel : \n\nhttps://www.kaggle.com/ashishpatel26/ensemble-learning-for-beginner-to-advance\n\nSo please let me know if have any query ?\nThanks"
  },
  "source": "meta"
}