{
  "id": 417803,
  "title": "11th Place Solution - Simplicity (Luck) Is All You Need",
  "content": "<p>First, my thanks to Kagglers for very informative and helpful discussions and notebooks. Also, thank you to Kaggle and GoDaddy for hosting the competition.</p>\n<p>My code is available at <a href=\"https://github.com/ywugwu/GoDaddy-MDF-Competition\" target=\"_blank\">github</a>, which is based on <a href=\"https://www.kaggle.com/code/vadimkamaev/microbusiness-density\" target=\"_blank\">VADIM KAMAEV's</a> public notebook. </p>\n<p>In this competition, my idea is to keep improving my CV and PB score without letting my model grow too complex.</p>\n<h1>Validation (CV)</h1>\n<p>I used the last 5 month data for validation. The model with the best validation score is used as one of my final submissions.</p>\n<h1>Model</h1>\n<p>Same with the public notebook, I used a Catboost Regressor to stack a LightGBM, an XGBoost, and a CatBoost model. Tuning a lot hyperparameters.</p>\n<h1>Feature Engineering</h1>\n<p>I didn't do too much novel feature engineering, only used techniques from public notebooks. </p>\n<p>But I change the hyperparameters a lot.</p>\n<h1>Blending Submissions</h1>\n<p>I used different hyperparameters of the model and different features (i.e., with different diff, max, mean, etc. settings) to conduct a large number of submissions. I select the best 4 (or maybe 3) submissions and use a weighted average to create a blended submission. The best (according to the public leaderboard) blended result is selected as one of my final submission. In the end, the private leaderboard score of blended submission is slightly better than the best CV model.</p>",
  "comments": [
    {
      "id": 2306619,
      "author_name": "CoreyJamesLevinson",
      "author_url": "",
      "post_date": "2023-06-17T12:46:20.477000",
      "content": "<p>Congrats, well done!</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 2309736,
      "author_name": "@kaggleqrdl",
      "author_url": "",
      "post_date": "2023-06-19T21:11:01.737000",
      "content": "<p>Hey Yu, congrats on your solo gold!   Did you ever get a chance to eval  your model against the march data separately?</p>",
      "votes": 0,
      "replies": [
        {
          "id": 2309749,
          "author_name": "Yu Wu",
          "author_url": "",
          "post_date": "2023-06-19T21:28:18.830000",
          "content": "<p>Ty, I hope it would be a gold for me😂.</p>\n<p>Do you mean using the data from <a href=\"https://www.kaggle.com/competitions/godaddy-microbusiness-density-forecasting/discussion/411428\" target=\"_blank\">here</a>?</p>\n<p>I got </p>\n<pre><code>. for single . for </code></pre>",
          "votes": 1,
          "replies": []
        }
      ]
    },
    {
      "id": 2306377,
      "author_name": "Machine Lying",
      "author_url": "",
      "post_date": "2023-06-17T08:41:40.820000",
      "content": "<p>Congratulations! But this github link is not working….</p>",
      "votes": 0,
      "replies": [
        {
          "id": 2306380,
          "author_name": "Yu Wu",
          "author_url": "",
          "post_date": "2023-06-17T08:46:26.880000",
          "content": "<p>thanks for your reminder, I added an extra <code>.</code> at the end of my URL 😭. Now, It should work. </p>",
          "votes": 1,
          "replies": [
            {
              "id": 2306386,
              "author_name": "Machine Lying",
              "author_url": "",
              "post_date": "2023-06-17T08:50:50.893000",
              "content": "<p>Thanks! Very detailed steps</p>",
              "votes": 0,
              "replies": []
            }
          ]
        }
      ]
    }
  ],
  "comment_count": 6,
  "messages": [
    {
      "id": 2306358,
      "postDate": "2023-06-17T08:29:14.087Z",
      "content": "<p>First, my thanks to Kagglers for very informative and helpful discussions and notebooks. Also, thank you to Kaggle and GoDaddy for hosting the competition.</p>\n<p>My code is available at <a href=\"https://github.com/ywugwu/GoDaddy-MDF-Competition\" target=\"_blank\">github</a>, which is based on <a href=\"https://www.kaggle.com/code/vadimkamaev/microbusiness-density\" target=\"_blank\">VADIM KAMAEV's</a> public notebook. </p>\n<p>In this competition, my idea is to keep improving my CV and PB score without letting my model grow too complex.</p>\n<h1>Validation (CV)</h1>\n<p>I used the last 5 month data for validation. The model with the best validation score is used as one of my final submissions.</p>\n<h1>Model</h1>\n<p>Same with the public notebook, I used a Catboost Regressor to stack a LightGBM, an XGBoost, and a CatBoost model. Tuning a lot hyperparameters.</p>\n<h1>Feature Engineering</h1>\n<p>I didn't do too much novel feature engineering, only used techniques from public notebooks. </p>\n<p>But I change the hyperparameters a lot.</p>\n<h1>Blending Submissions</h1>\n<p>I used different hyperparameters of the model and different features (i.e., with different diff, max, mean, etc. settings) to conduct a large number of submissions. I select the best 4 (or maybe 3) submissions and use a weighted average to create a blended submission. The best (according to the public leaderboard) blended result is selected as one of my final submission. In the end, the private leaderboard score of blended submission is slightly better than the best CV model.</p>",
      "rawMarkdown": "First, my thanks to Kagglers for very informative and helpful discussions and notebooks. Also, thank you to Kaggle and GoDaddy for hosting the competition.\n\nMy code is available at [github](https://github.com/ywugwu/GoDaddy-MDF-Competition), which is based on [VADIM KAMAEV's](https://www.kaggle.com/code/vadimkamaev/microbusiness-density) public notebook. \n\nIn this competition, my idea is to keep improving my CV and PB score without letting my model grow too complex.\n\n# Validation (CV)\nI used the last 5 month data for validation. The model with the best validation score is used as one of my final submissions.\n\n# Model\nSame with the public notebook, I used a Catboost Regressor to stack a LightGBM, an XGBoost, and a CatBoost model. Tuning a lot hyperparameters.\n\n# Feature Engineering\nI didn't do too much novel feature engineering, only used techniques from public notebooks. \n\nBut I change the hyperparameters a lot.\n\n# Blending Submissions\nI used different hyperparameters of the model and different features (i.e., with different diff, max, mean, etc. settings) to conduct a large number of submissions. I select the best 4 (or maybe 3) submissions and use a weighted average to create a blended submission. The best (according to the public leaderboard) blended result is selected as one of my final submission. In the end, the private leaderboard score of blended submission is slightly better than the best CV model.\n\n",
      "votes": 10
    },
    {
      "id": 2306619,
      "postDate": "2023-06-17T12:46:20.477Z",
      "content": "<p>Congrats, well done!</p>",
      "rawMarkdown": "Congrats, well done!",
      "votes": 1
    },
    {
      "id": 2309736,
      "postDate": "2023-06-19T21:11:01.737Z",
      "content": "<p>Hey Yu, congrats on your solo gold!   Did you ever get a chance to eval  your model against the march data separately?</p>",
      "rawMarkdown": "Hey Yu, congrats on your solo gold!   Did you ever get a chance to eval  your model against the march data separately?",
      "replies": [
        {
          "id": 2309749,
          "postDate": "2023-06-19T21:28:18.830Z",
          "content": "<p>Ty, I hope it would be a gold for me😂.</p>\n<p>Do you mean using the data from <a href=\"https://www.kaggle.com/competitions/godaddy-microbusiness-density-forecasting/discussion/411428\" target=\"_blank\">here</a>?</p>\n<p>I got </p>\n<pre><code>. for single . for </code></pre>",
          "rawMarkdown": "Ty, I hope it would be a gold for me😂.\n\nDo you mean using the data from [here](https://www.kaggle.com/competitions/godaddy-microbusiness-density-forecasting/discussion/411428)?\n\n\nI got \n\n```\n3.4473405812966584 for best single submission\n3.4440930640476974 for blended submission\n```\n",
          "votes": 1
        }
      ]
    },
    {
      "id": 2306377,
      "postDate": "2023-06-17T08:41:40.820Z",
      "content": "<p>Congratulations! But this github link is not working….</p>",
      "rawMarkdown": "Congratulations! But this github link is not working....",
      "replies": [
        {
          "id": 2306380,
          "postDate": "2023-06-17T08:46:26.880Z",
          "content": "<p>thanks for your reminder, I added an extra <code>.</code> at the end of my URL 😭. Now, It should work. </p>",
          "rawMarkdown": "thanks for your reminder, I added an extra `.` at the end of my URL 😭. Now, It should work. ",
          "votes": 1,
          "replies": [
            {
              "id": 2306386,
              "postDate": "2023-06-17T08:50:50.893Z",
              "content": "<p>Thanks! Very detailed steps</p>",
              "rawMarkdown": "Thanks! Very detailed steps"
            }
          ]
        }
      ]
    }
  ],
  "complete": true,
  "source": "live",
  "fetched_at": "2026-10-19T19:28:27Z",
  "competition_id": 41881,
  "competition_slug": "godaddy-microbusiness-density-forecasting",
  "team_id": 9672857,
  "solution_url": "/competitions/godaddy-microbusiness-density-forecasting/writeups/luck-is-all-you-need-11th-place-solution-simplicit"
}