{
  "id": 349741,
  "title": "3rd solution--simple is the best",
  "url": "/competitions/amex-default-prediction/discussion/349741",
  "author_name": "Alex",
  "post_date": "2022-09-02T14:24:05.328000",
  "votes": 80,
  "comment_count": 23,
  "views": 0,
  "content": "<h2>feature engineering is all you need</h2>\n<p>First of all,  Big thanks  to Amex and Kaggle for organizing a very interesting competition.I am too busy my work so that I haven't done this competition for about a month and a half. I came back a few days before the competition ended. I am very happy to rank 3rd in this competition. Thanks to <a href=\"https://www.kaggle.com/raddar\" target=\"_blank\">@raddar</a> denoise idea and <a href=\"https://www.kaggle.com/ragnar123\" target=\"_blank\">@ragnar123</a> public notebook.</p>\n<h2>my solution</h2>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4921130%2Fe26b7df663bf54181e721d156902c4d7%2FAmex_top3.png?generation=1662130516395369&amp;alt=media\" alt=\"\"></p>\n<p>I will share my code after winning call.Thanks for you reading!</p>",
  "messages": [
    {
      "id": 2904554,
      "postDate": "2024-07-04T12:48:27.520Z",
      "content": "<p>Hello Alex, Trust you are doing great</p>\n<p>Did you later drop the code?</p>",
      "rawMarkdown": "Hello Alex, Trust you are doing great\n\nDid you later drop the code?",
      "votes": 1
    },
    {
      "id": 1923778,
      "postDate": "2022-09-02T14:24:05.327Z",
      "content": "<h2>feature engineering is all you need</h2>\n<p>First of all,  Big thanks  to Amex and Kaggle for organizing a very interesting competition.I am too busy my work so that I haven't done this competition for about a month and a half. I came back a few days before the competition ended. I am very happy to rank 3rd in this competition. Thanks to <a href=\"https://www.kaggle.com/raddar\" target=\"_blank\">@raddar</a> denoise idea and <a href=\"https://www.kaggle.com/ragnar123\" target=\"_blank\">@ragnar123</a> public notebook.</p>\n<h2>my solution</h2>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4921130%2Fe26b7df663bf54181e721d156902c4d7%2FAmex_top3.png?generation=1662130516395369&amp;alt=media\" alt=\"\"></p>\n<p>I will share my code after winning call.Thanks for you reading!</p>",
      "rawMarkdown": "feature engineering is all you need\n---\nFirst of all,  Big thanks  to Amex and Kaggle for organizing a very interesting competition.I am too busy my work so that I haven't done this competition for about a month and a half. I came back a few days before the competition ended. I am very happy to rank 3rd in this competition. Thanks to @raddar denoise idea and @ragnar123 public notebook.\nmy solution\n---\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4921130%2Fe26b7df663bf54181e721d156902c4d7%2FAmex_top3.png?generation=1662130516395369&alt=media)\n\nI will share my code after winning call.Thanks for you reading!",
      "votes": 80
    },
    {
      "id": 1925546,
      "postDate": "2022-09-04T04:38:02.960Z",
      "content": "<p>Congratulations Alex. Amazing solo gold 3rd place Gold. You will be Kaggle competition Grandmaster soon!</p>\n<p>I like your simple solution. Looks like you made some strong features!</p>",
      "rawMarkdown": "Congratulations Alex. Amazing solo gold 3rd place Gold. You will be Kaggle competition Grandmaster soon!\n\nI like your simple solution. Looks like you made some strong features!",
      "votes": 6
    },
    {
      "id": 1936337,
      "postDate": "2022-09-12T17:07:45.837Z",
      "content": "<p>Hi Alex ( <a href=\"https://www.kaggle.com/lihuajing\" target=\"_blank\">@lihuajing</a>), congratulations on the solo 3rd place gold. Looking forward to learn from your code. </p>",
      "rawMarkdown": "Hi Alex ( @lihuajing), congratulations on the solo 3rd place gold. Looking forward to learn from your code. ",
      "votes": 1
    },
    {
      "id": 1929979,
      "postDate": "2022-09-07T13:15:59.477Z",
      "content": "<p>looking forward the code!</p>",
      "rawMarkdown": "looking forward the code!",
      "votes": 1
    },
    {
      "id": 1927747,
      "postDate": "2022-09-05T23:58:34.223Z",
      "content": "<p>Congratulations <a href=\"https://www.kaggle.com/lihuajing\" target=\"_blank\">@lihuajing</a> </p>",
      "rawMarkdown": "Congratulations @lihuajing ",
      "votes": 1
    },
    {
      "id": 1923826,
      "postDate": "2022-09-02T15:00:08.930Z",
      "content": "<p>Wow nice! Congrats<br>\nCan't wait for the code, this is insane! </p>",
      "rawMarkdown": "Wow nice! Congrats\nCan't wait for the code, this is insane! ",
      "votes": 2
    },
    {
      "id": 3068874,
      "postDate": "2024-12-10T18:43:46.673Z",
      "content": "<p>Thank you for sharing the diagram and detailing the simple solution!</p>",
      "rawMarkdown": "Thank you for sharing the diagram and detailing the simple solution!"
    },
    {
      "id": 2281218,
      "postDate": "2023-05-30T17:28:08.463Z",
      "content": "<p>Hello Alex, where can I find your code? I love to learn it!</p>",
      "rawMarkdown": "Hello Alex, where can I find your code? I love to learn it!"
    },
    {
      "id": 2009503,
      "postDate": "2022-10-30T04:50:44.243Z",
      "content": "<p>Amazing ! Thanks! Eagerly waiting for your code…</p>",
      "rawMarkdown": "Amazing ! Thanks! Eagerly waiting for your code..."
    },
    {
      "id": 2007049,
      "postDate": "2022-10-28T01:34:31.717Z",
      "content": "<p>Wow! Congratulations Alex. The feature engineering part is the killer. Looking to hear more of your solution. Thanks.</p>",
      "rawMarkdown": "Wow! Congratulations Alex. The feature engineering part is the killer. Looking to hear more of your solution. Thanks."
    },
    {
      "id": 2007046,
      "postDate": "2022-10-28T01:29:44.557Z",
      "content": "<p>congratulations on the solo 3rd place gold. Looking forward to learn from your code.</p>",
      "rawMarkdown": "congratulations on the solo 3rd place gold. Looking forward to learn from your code."
    },
    {
      "id": 1999347,
      "postDate": "2022-10-22T09:05:10.020Z",
      "content": "<p>Congratulations！ Looking forward to learning from your code.</p>",
      "rawMarkdown": "Congratulations！ Looking forward to learning from your code."
    },
    {
      "id": 1940188,
      "postDate": "2022-09-15T08:09:29.443Z",
      "content": "<p>How do you select the initial num_features and cat_features to use? <br>\nAlso, love to know the magic parts =)</p>",
      "rawMarkdown": "How do you select the initial num_features and cat_features to use? \nAlso, love to know the magic parts =)"
    },
    {
      "id": 1936610,
      "postDate": "2022-09-12T22:51:19.497Z",
      "content": "<p>Amazing Alex, Congratulations </p>",
      "rawMarkdown": "Amazing Alex, Congratulations "
    },
    {
      "id": 1936574,
      "postDate": "2022-09-12T21:16:15.963Z",
      "content": "<p>Wow! Congratulations Alex. The feature engineering part is the killer. Looking to hear more of your solution. Thanks.</p>",
      "rawMarkdown": "Wow! Congratulations Alex. The feature engineering part is the killer. Looking to hear more of your solution. Thanks."
    },
    {
      "id": 1936427,
      "postDate": "2022-09-12T18:35:47.193Z",
      "content": "<p>can you share code, for  for 1st place already code shared ?</p>",
      "rawMarkdown": "can you share code, for  for 1st place already code shared ?"
    },
    {
      "id": 1927766,
      "postDate": "2022-09-06T01:00:37.500Z",
      "content": "<p>awesome! Awaiting your code to be trained.</p>",
      "rawMarkdown": "awesome! Awaiting your code to be trained."
    },
    {
      "id": 1926594,
      "postDate": "2022-09-05T01:24:36.523Z",
      "content": "<p>can't wait for your code to learn!</p>",
      "rawMarkdown": "can't wait for your code to learn!"
    },
    {
      "id": 1924022,
      "postDate": "2022-09-02T17:05:05.430Z",
      "content": "<p>Great result emanating from a baroque approach! Keep up the great work, wishing you the best and hearty congratulations <a href=\"https://www.kaggle.com/lihuajing\" target=\"_blank\">@lihuajing</a>!!</p>",
      "rawMarkdown": "Great result emanating from a baroque approach! Keep up the great work, wishing you the best and hearty congratulations @lihuajing!!"
    },
    {
      "id": 1923932,
      "postDate": "2022-09-02T16:04:22.587Z",
      "content": "<p>Hello <a href=\"https://www.kaggle.com/lihuajing\" target=\"_blank\">@lihuajing</a> thanks for sharing your solution</p>",
      "rawMarkdown": "Hello @lihuajing thanks for sharing your solution"
    },
    {
      "id": 1923812,
      "postDate": "2022-09-02T14:47:00.030Z",
      "content": "<p>Congratulations! Very impressive approach! I'm very interested in your LGB model and I would like to accelerate the inference part using NVIDIA Triton. Please keep me posted when the code is available. Thank you! </p>",
      "rawMarkdown": "Congratulations! Very impressive approach! I'm very interested in your LGB model and I would like to accelerate the inference part using NVIDIA Triton. Please keep me posted when the code is available. Thank you! ",
      "replies": [
        {
          "id": 1924551,
          "postDate": "2022-09-03T06:47:19.873Z",
          "content": "<p>Keep me too!</p>",
          "rawMarkdown": "Keep me too!"
        }
      ]
    },
    {
      "id": 1928052,
      "postDate": "2022-09-06T07:55:55Z",
      "rawMarkdown": "",
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 2904554,
      "author_name": "Fagbemi Tunji",
      "author_url": "",
      "post_date": "2024-07-04T12:48:27.520000",
      "content": "<p>Hello Alex, Trust you are doing great</p>\n<p>Did you later drop the code?</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 1925546,
      "author_name": "Chris Deotte",
      "author_url": "",
      "post_date": "2022-09-04T04:38:02.960000",
      "content": "<p>Congratulations Alex. Amazing solo gold 3rd place Gold. You will be Kaggle competition Grandmaster soon!</p>\n<p>I like your simple solution. Looks like you made some strong features!</p>",
      "votes": 6,
      "replies": []
    },
    {
      "id": 1936337,
      "author_name": "Data student",
      "author_url": "",
      "post_date": "2022-09-12T17:07:45.837000",
      "content": "<p>Hi Alex ( <a href=\"https://www.kaggle.com/lihuajing\" target=\"_blank\">@lihuajing</a>), congratulations on the solo 3rd place gold. Looking forward to learn from your code. </p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 1929979,
      "author_name": "Quân Trần Đình Đại",
      "author_url": "",
      "post_date": "2022-09-07T13:15:59.477000",
      "content": "<p>looking forward the code!</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 1927747,
      "author_name": "Muhammad Anas Mahmood",
      "author_url": "",
      "post_date": "2022-09-05T23:58:34.223000",
      "content": "<p>Congratulations <a href=\"https://www.kaggle.com/lihuajing\" target=\"_blank\">@lihuajing</a> </p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 1923826,
      "author_name": "The Devastator",
      "author_url": "",
      "post_date": "2022-09-02T15:00:08.930000",
      "content": "<p>Wow nice! Congrats<br>\nCan't wait for the code, this is insane! </p>",
      "votes": 2,
      "replies": []
    },
    {
      "id": 3068874,
      "author_name": "Anthony Bernardi",
      "author_url": "",
      "post_date": "2024-12-10T18:43:46.673000",
      "content": "<p>Thank you for sharing the diagram and detailing the simple solution!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2281218,
      "author_name": "Just a Duck",
      "author_url": "",
      "post_date": "2023-05-30T17:28:08.463000",
      "content": "<p>Hello Alex, where can I find your code? I love to learn it!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2009503,
      "author_name": "Johnny Torres",
      "author_url": "",
      "post_date": "2022-10-30T04:50:44.243000",
      "content": "<p>Amazing ! Thanks! Eagerly waiting for your code…</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2007049,
      "author_name": "nianliu107",
      "author_url": "",
      "post_date": "2022-10-28T01:34:31.717000",
      "content": "<p>Wow! Congratulations Alex. The feature engineering part is the killer. Looking to hear more of your solution. Thanks.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2007046,
      "author_name": "nianliu107",
      "author_url": "",
      "post_date": "2022-10-28T01:29:44.557000",
      "content": "<p>congratulations on the solo 3rd place gold. Looking forward to learn from your code.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1999347,
      "author_name": "Liming",
      "author_url": "",
      "post_date": "2022-10-22T09:05:10.020000",
      "content": "<p>Congratulations！ Looking forward to learning from your code.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1940188,
      "author_name": "skw1990",
      "author_url": "",
      "post_date": "2022-09-15T08:09:29.443000",
      "content": "<p>How do you select the initial num_features and cat_features to use? <br>\nAlso, love to know the magic parts =)</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1936610,
      "author_name": "Ezzaldin Mamdouh",
      "author_url": "",
      "post_date": "2022-09-12T22:51:19.497000",
      "content": "<p>Amazing Alex, Congratulations </p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1936574,
      "author_name": "raccoon",
      "author_url": "",
      "post_date": "2022-09-12T21:16:15.963000",
      "content": "<p>Wow! Congratulations Alex. The feature engineering part is the killer. Looking to hear more of your solution. Thanks.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1936427,
      "author_name": "Sandy",
      "author_url": "",
      "post_date": "2022-09-12T18:35:47.193000",
      "content": "<p>can you share code, for  for 1st place already code shared ?</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1927766,
      "author_name": "Anubhav Kundu",
      "author_url": "",
      "post_date": "2022-09-06T01:00:37.500000",
      "content": "<p>awesome! Awaiting your code to be trained.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1926594,
      "author_name": "backpack",
      "author_url": "",
      "post_date": "2022-09-05T01:24:36.523000",
      "content": "<p>can't wait for your code to learn!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1924022,
      "author_name": "Ravi Ramakrishnan",
      "author_url": "",
      "post_date": "2022-09-02T17:05:05.430000",
      "content": "<p>Great result emanating from a baroque approach! Keep up the great work, wishing you the best and hearty congratulations <a href=\"https://www.kaggle.com/lihuajing\" target=\"_blank\">@lihuajing</a>!!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1923932,
      "author_name": "C4rl05/V",
      "author_url": "",
      "post_date": "2022-09-02T16:04:22.587000",
      "content": "<p>Hello <a href=\"https://www.kaggle.com/lihuajing\" target=\"_blank\">@lihuajing</a> thanks for sharing your solution</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 1923812,
      "author_name": "Jiwei Liu",
      "author_url": "",
      "post_date": "2022-09-02T14:47:00.030000",
      "content": "<p>Congratulations! Very impressive approach! I'm very interested in your LGB model and I would like to accelerate the inference part using NVIDIA Triton. Please keep me posted when the code is available. Thank you! </p>",
      "votes": 0,
      "replies": [
        {
          "id": 1924551,
          "author_name": "Jackson You",
          "author_url": "",
          "post_date": "2022-09-03T06:47:19.873000",
          "content": "<p>Keep me too!</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 1928052,
      "author_name": "",
      "author_url": "",
      "post_date": "2022-09-06T07:55:55",
      "content": "",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2904554": "Hello Alex, Trust you are doing great\n\nDid you later drop the code?",
    "1923778": "feature engineering is all you need\n---\nFirst of all,  Big thanks  to Amex and Kaggle for organizing a very interesting competition.I am too busy my work so that I haven't done this competition for about a month and a half. I came back a few days before the competition ended. I am very happy to rank 3rd in this competition. Thanks to @raddar denoise idea and @ragnar123 public notebook.\nmy solution\n---\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F4921130%2Fe26b7df663bf54181e721d156902c4d7%2FAmex_top3.png?generation=1662130516395369&alt=media)\n\nI will share my code after winning call.Thanks for you reading!",
    "1925546": "Congratulations Alex. Amazing solo gold 3rd place Gold. You will be Kaggle competition Grandmaster soon!\n\nI like your simple solution. Looks like you made some strong features!",
    "1936337": "Hi Alex ( @lihuajing), congratulations on the solo 3rd place gold. Looking forward to learn from your code. ",
    "1929979": "looking forward the code!",
    "1927747": "Congratulations @lihuajing ",
    "1923826": "Wow nice! Congrats\nCan't wait for the code, this is insane! ",
    "3068874": "Thank you for sharing the diagram and detailing the simple solution!",
    "2281218": "Hello Alex, where can I find your code? I love to learn it!",
    "2009503": "Amazing ! Thanks! Eagerly waiting for your code...",
    "2007049": "Wow! Congratulations Alex. The feature engineering part is the killer. Looking to hear more of your solution. Thanks.",
    "2007046": "congratulations on the solo 3rd place gold. Looking forward to learn from your code.",
    "1999347": "Congratulations！ Looking forward to learning from your code.",
    "1940188": "How do you select the initial num_features and cat_features to use? \nAlso, love to know the magic parts =)",
    "1936610": "Amazing Alex, Congratulations ",
    "1936574": "Wow! Congratulations Alex. The feature engineering part is the killer. Looking to hear more of your solution. Thanks.",
    "1936427": "can you share code, for  for 1st place already code shared ?",
    "1927766": "awesome! Awaiting your code to be trained.",
    "1926594": "can't wait for your code to learn!",
    "1924022": "Great result emanating from a baroque approach! Keep up the great work, wishing you the best and hearty congratulations @lihuajing!!",
    "1923932": "Hello @lihuajing thanks for sharing your solution",
    "1923812": "Congratulations! Very impressive approach! I'm very interested in your LGB model and I would like to accelerate the inference part using NVIDIA Triton. Please keep me posted when the code is available. Thank you! ",
    "1928052": ""
  }
}