{
  "id": 552556,
  "title": "Simple model that I didn't submitted (private 0.477)",
  "url": "/competitions/child-mind-institute-problematic-internet-use/discussion/552556",
  "author_name": "",
  "post_date": "2024-12-20T09:04:20.266752Z",
  "votes": null,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Congratulations to everyone!<br>\nThis is a very interesting competition although I've dropped out in the 2nd half since I needed to focus on an important interview (and luckily I nailed it!).</p>\n<p>I found a dropped notebook that scored 0.477 (public 0.463, my 9th highest notebook). Of course I didn't pick it!<br>\nAll the code was borrowed from public notebook. I just did a little bit of feature engineering.<br>\n<a href=\"https://www.kaggle.com/code/tomyuen/clean-of-cmi-single-lgbm-f8db91?scriptVersionId=199839476\" target=\"_blank\">notebook</a></p>\n<p>Here's what has been done:</p>\n<ol>\n<li>Merge PAQ_A &amp; PAQ_C</li>\n<li>Drop id 2065 3205 </li>\n<li>Add new weight% = \"Difference compared with US average weight, according to age group\"</li>\n<li>Clip Heartrate/ BMC/ BMR to remove the outliners</li>\n<li>In both train set/ test set, fill blank BP, Heartrate, BMI, Height, Weight according to average of these in train set based on age + sex.</li>\n</ol>\n<p>And then 1 single LGBM, SKF=10.</p>\n<p>My best wishes to you all.</p>",
  "messages": [
    {
      "id": "3076784",
      "postDate": "12/20/2024 09:04:20",
      "content": "<p>Congratulations to everyone!<br>\nThis is a very interesting competition although I've dropped out in the 2nd half since I needed to focus on an important interview (and luckily I nailed it!).</p>\n<p>I found a dropped notebook that scored 0.477 (public 0.463, my 9th highest notebook). Of course I didn't pick it!<br>\nAll the code was borrowed from public notebook. I just did a little bit of feature engineering.<br>\n<a href=\"https://www.kaggle.com/code/tomyuen/clean-of-cmi-single-lgbm-f8db91?scriptVersionId=199839476\" target=\"_blank\">notebook</a></p>\n<p>Here's what has been done:</p>\n<ol>\n<li>Merge PAQ_A &amp; PAQ_C</li>\n<li>Drop id 2065 3205 </li>\n<li>Add new weight% = \"Difference compared with US average weight, according to age group\"</li>\n<li>Clip Heartrate/ BMC/ BMR to remove the outliners</li>\n<li>In both train set/ test set, fill blank BP, Heartrate, BMI, Height, Weight according to average of these in train set based on age + sex.</li>\n</ol>\n<p>And then 1 single LGBM, SKF=10.</p>\n<p>My best wishes to you all.</p>",
      "rawMarkdown": "Congratulations to everyone!\nThis is a very interesting competition although I've dropped out in the 2nd half since I needed to focus on an important interview (and luckily I nailed it!).\n\nI found a dropped notebook that scored 0.477 (public 0.463, my 9th highest notebook). Of course I didn't pick it!\nAll the code was borrowed from public notebook. I just did a little bit of feature engineering.\n[notebook](https://www.kaggle.com/code/tomyuen/clean-of-cmi-single-lgbm-f8db91?scriptVersionId=199839476)\n\nHere's what has been done:\n\n1. Merge PAQ_A & PAQ_C\n2. Drop id 2065 3205 \n3. Add new weight% = \"Difference compared with US average weight, according to age group\"\n4. Clip Heartrate/ BMC/ BMR to remove the outliners\n5. In both train set/ test set, fill blank BP, Heartrate, BMI, Height, Weight according to average of these in train set based on age + sex.\n\nAnd then 1 single LGBM, SKF=10.\n\nMy best wishes to you all.",
      "votes": null
    },
    {
      "id": "3076838",
      "postDate": "12/20/2024 10:08:40",
      "content": "<p>me too I have a 5 simple submission with lighGBM (silver medal but didn't chose anyone of them 😭) </p>",
      "rawMarkdown": "me too I have a 5 simple submission with lighGBM (silver medal but didn't chose anyone of them 😭)",
      "votes": null
    },
    {
      "id": "3077449",
      "postDate": "12/21/2024 01:02:40",
      "content": "<p>Ah, yes, there is a world outside of kaggle 🙃  Congrats on nailing your interview!</p>",
      "rawMarkdown": "Ah, yes, there is a world outside of kaggle 🙃  Congrats on nailing your interview!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3076838,
      "author_name": "hamzaghanmi",
      "author_url": "",
      "post_date": "12/20/2024 10:08:40",
      "content": "<p>me too I have a 5 simple submission with lighGBM (silver medal but didn't chose anyone of them 😭) </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 3077449,
      "author_name": "dan3dewey",
      "author_url": "",
      "post_date": "12/21/2024 01:02:40",
      "content": "<p>Ah, yes, there is a world outside of kaggle 🙃  Congrats on nailing your interview!</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3076784": "Congratulations to everyone!\nThis is a very interesting competition although I've dropped out in the 2nd half since I needed to focus on an important interview (and luckily I nailed it!).\n\nI found a dropped notebook that scored 0.477 (public 0.463, my 9th highest notebook). Of course I didn't pick it!\nAll the code was borrowed from public notebook. I just did a little bit of feature engineering.\n[notebook](https://www.kaggle.com/code/tomyuen/clean-of-cmi-single-lgbm-f8db91?scriptVersionId=199839476)\n\nHere's what has been done:\n\n1. Merge PAQ_A & PAQ_C\n2. Drop id 2065 3205 \n3. Add new weight% = \"Difference compared with US average weight, according to age group\"\n4. Clip Heartrate/ BMC/ BMR to remove the outliners\n5. In both train set/ test set, fill blank BP, Heartrate, BMI, Height, Weight according to average of these in train set based on age + sex.\n\nAnd then 1 single LGBM, SKF=10.\n\nMy best wishes to you all.",
    "3076838": "me too I have a 5 simple submission with lighGBM (silver medal but didn't chose anyone of them 😭)",
    "3077449": "Ah, yes, there is a world outside of kaggle 🙃  Congrats on nailing your interview!"
  },
  "source": "meta"
}