{
  "id": 82048,
  "title": "Congratulations to everyone and many thanks to my teammates",
  "url": "/competitions/elo-merchant-category-recommendation/discussion/82048",
  "author_name": "",
  "post_date": "2019-02-27T01:40:43.572697200Z",
  "votes": 13,
  "comment_count": 11,
  "views": 0,
  "content": "<p>This is my first kaggle competition, I learned almost every thing from this competition and from my teammates(<a href=\"https://www.kaggle.com/codlife\">@Wang Jianfei</a>, <a href=\"https://www.kaggle.com/salamatburj\">@Salamat</a> and <a href=\"https://www.kaggle.com/palend\">@KimposiPungmudong</a>). We do not have much experience, so we learn a lot from previous winner's solution share. The most helpful rules I think is:</p>\n\n<ul>\n<li><p>Spend most of time on feature engineering and get a few good and diverse models;</p></li>\n<li><p>Get as many models as possible, weak but diverse models can also help stacking. In fact we have write a script to generate about ~1000 models, which turns out helpful.</p></li>\n<li><p>For stacking, it's really only important to have one or two models that score well individually. After that, it's all about diversity, and even very weak models can help. (thanks <a href=\"https://www.kaggle.com/c/two-sigma-connect-rental-listing-inquiries/discussion/31765\">Silogram's post</a>)</p></li>\n</ul>\n\n<p>Hope everyone have fun and learn a lot from this competition.</p>",
  "messages": [
    {
      "id": "479145",
      "postDate": "02/27/2019 01:40:43",
      "content": "<p>This is my first kaggle competition, I learned almost every thing from this competition and from my teammates(<a href=\"https://www.kaggle.com/codlife\">@Wang Jianfei</a>, <a href=\"https://www.kaggle.com/salamatburj\">@Salamat</a> and <a href=\"https://www.kaggle.com/palend\">@KimposiPungmudong</a>). We do not have much experience, so we learn a lot from previous winner's solution share. The most helpful rules I think is:</p>\n\n<ul>\n<li><p>Spend most of time on feature engineering and get a few good and diverse models;</p></li>\n<li><p>Get as many models as possible, weak but diverse models can also help stacking. In fact we have write a script to generate about ~1000 models, which turns out helpful.</p></li>\n<li><p>For stacking, it's really only important to have one or two models that score well individually. After that, it's all about diversity, and even very weak models can help. (thanks <a href=\"https://www.kaggle.com/c/two-sigma-connect-rental-listing-inquiries/discussion/31765\">Silogram's post</a>)</p></li>\n</ul>\n\n<p>Hope everyone have fun and learn a lot from this competition.</p>",
      "rawMarkdown": "This is my first kaggle competition, I learned almost every thing from this competition and from my teammates([@Wang Jianfei](https://www.kaggle.com/codlife), [@Salamat](https://www.kaggle.com/salamatburj) and [@KimposiPungmudong](https://www.kaggle.com/palend)). We do not have much experience, so we learn a lot from previous winner's solution share. The most helpful rules I think is:\n\n- Spend most of time on feature engineering and get a few good and diverse models;\n\n- Get as many models as possible, weak but diverse models can also help stacking. In fact we have write a script to generate about ~1000 models, which turns out helpful.\n\n- For stacking, it's really only important to have one or two models that score well individually. After that, it's all about diversity, and even very weak models can help. (thanks [Silogram's post](https://www.kaggle.com/c/two-sigma-connect-rental-listing-inquiries/discussion/31765))\n\nHope everyone have fun and learn a lot from this competition.",
      "votes": null
    },
    {
      "id": "479151",
      "postDate": "02/27/2019 01:47:25",
      "content": "<p>Congratulations</p>",
      "rawMarkdown": "Congratulations",
      "votes": null
    },
    {
      "id": "479163",
      "postDate": "02/27/2019 01:58:48",
      "content": "<p>Congratulations </p>",
      "rawMarkdown": "Congratulations",
      "votes": null
    },
    {
      "id": "479171",
      "postDate": "02/27/2019 02:06:14",
      "content": "<p>Congratulations</p>",
      "rawMarkdown": "Congratulations",
      "votes": null
    },
    {
      "id": "479198",
      "postDate": "02/27/2019 02:32:14",
      "content": "<p>Congratulations.</p>\n\n<p>Here is my post on feature selection techniques <a href=\"https://www.kaggle.com/c/elo-merchant-category-recommendation/discussion/82052\">https://www.kaggle.com/c/elo-merchant-category-recommendation/discussion/82052</a></p>",
      "rawMarkdown": "Congratulations.\n\nHere is my post on feature selection techniques https://www.kaggle.com/c/elo-merchant-category-recommendation/discussion/82052",
      "votes": null
    },
    {
      "id": "479232",
      "postDate": "02/27/2019 03:03:25",
      "content": "<p>Congratulations and thanks!</p>",
      "rawMarkdown": "Congratulations and thanks!",
      "votes": null
    },
    {
      "id": "479360",
      "postDate": "02/27/2019 05:02:19",
      "content": "<p>congrats！</p>",
      "rawMarkdown": "congrats！",
      "votes": null
    },
    {
      "id": "479385",
      "postDate": "02/27/2019 05:28:26",
      "content": "<p>Congratulations and Thanks for Sharing !!</p>",
      "rawMarkdown": "Congratulations and Thanks for Sharing !!",
      "votes": null
    },
    {
      "id": "479492",
      "postDate": "02/27/2019 07:35:21",
      "content": "<p>Congratulations <a href=\"/jiazhuang\">@jiazhuang</a> for getting your first gold medal.</p>",
      "rawMarkdown": "Congratulations @jiazhuang for getting your first gold medal.",
      "votes": null
    },
    {
      "id": "479792",
      "postDate": "02/27/2019 12:33:03",
      "content": "<p>Congratulations</p>",
      "rawMarkdown": "Congratulations",
      "votes": null
    },
    {
      "id": "479964",
      "postDate": "02/27/2019 16:19:09",
      "content": "<p>Wow, that's a lot of models! Could you tell us how many were involved in your final submissions, before stacking?</p>",
      "rawMarkdown": "Wow, that's a lot of models! Could you tell us how many were involved in your final submissions, before stacking?",
      "votes": null
    },
    {
      "id": "480363",
      "postDate": "02/28/2019 05:12:33",
      "content": "<p>Congratulations</p>",
      "rawMarkdown": "Congratulations",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 479151,
      "author_name": "kangzhang",
      "author_url": "",
      "post_date": "02/27/2019 01:47:25",
      "content": "<p>Congratulations</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 479163,
      "author_name": "longyin2",
      "author_url": "",
      "post_date": "02/27/2019 01:58:48",
      "content": "<p>Congratulations </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 479171,
      "author_name": "chizhu2018",
      "author_url": "",
      "post_date": "02/27/2019 02:06:14",
      "content": "<p>Congratulations</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 479198,
      "author_name": "sheriytm",
      "author_url": "",
      "post_date": "02/27/2019 02:32:14",
      "content": "<p>Congratulations.</p>\n\n<p>Here is my post on feature selection techniques <a href=\"https://www.kaggle.com/c/elo-merchant-category-recommendation/discussion/82052\">https://www.kaggle.com/c/elo-merchant-category-recommendation/discussion/82052</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 479232,
          "author_name": "jiazhuang",
          "author_url": "",
          "post_date": "02/27/2019 03:03:25",
          "content": "<p>Congratulations and thanks!</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 479360,
      "author_name": "process531",
      "author_url": "",
      "post_date": "02/27/2019 05:02:19",
      "content": "<p>congrats！</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 479385,
      "author_name": "viswanathravindran",
      "author_url": "",
      "post_date": "02/27/2019 05:28:26",
      "content": "<p>Congratulations and Thanks for Sharing !!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 479492,
      "author_name": "karthik7395",
      "author_url": "",
      "post_date": "02/27/2019 07:35:21",
      "content": "<p>Congratulations <a href=\"/jiazhuang\">@jiazhuang</a> for getting your first gold medal.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 479792,
      "author_name": "fuengwul5",
      "author_url": "",
      "post_date": "02/27/2019 12:33:03",
      "content": "<p>Congratulations</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 479964,
      "author_name": "brunogc",
      "author_url": "",
      "post_date": "02/27/2019 16:19:09",
      "content": "<p>Wow, that's a lot of models! Could you tell us how many were involved in your final submissions, before stacking?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 480363,
      "author_name": "zjucor",
      "author_url": "",
      "post_date": "02/28/2019 05:12:33",
      "content": "<p>Congratulations</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "479145": "This is my first kaggle competition, I learned almost every thing from this competition and from my teammates([@Wang Jianfei](https://www.kaggle.com/codlife), [@Salamat](https://www.kaggle.com/salamatburj) and [@KimposiPungmudong](https://www.kaggle.com/palend)). We do not have much experience, so we learn a lot from previous winner's solution share. The most helpful rules I think is:\n\n- Spend most of time on feature engineering and get a few good and diverse models;\n\n- Get as many models as possible, weak but diverse models can also help stacking. In fact we have write a script to generate about ~1000 models, which turns out helpful.\n\n- For stacking, it's really only important to have one or two models that score well individually. After that, it's all about diversity, and even very weak models can help. (thanks [Silogram's post](https://www.kaggle.com/c/two-sigma-connect-rental-listing-inquiries/discussion/31765))\n\nHope everyone have fun and learn a lot from this competition.",
    "479151": "Congratulations",
    "479163": "Congratulations",
    "479171": "Congratulations",
    "479198": "Congratulations.\n\nHere is my post on feature selection techniques https://www.kaggle.com/c/elo-merchant-category-recommendation/discussion/82052",
    "479232": "Congratulations and thanks!",
    "479360": "congrats！",
    "479385": "Congratulations and Thanks for Sharing !!",
    "479492": "Congratulations @jiazhuang for getting your first gold medal.",
    "479792": "Congratulations",
    "479964": "Wow, that's a lot of models! Could you tell us how many were involved in your final submissions, before stacking?",
    "480363": "Congratulations"
  },
  "source": "meta"
}