{
  "id": 536683,
  "title": "Dropping accelometer data",
  "url": "/competitions/child-mind-institute-problematic-internet-use/discussion/536683",
  "author_name": "Manoj kumar koppurouthu",
  "post_date": "2024-09-29T07:10:45.608000",
  "votes": 1,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hello all,this is my first kaggle competition and i got a score of 0.350.</p>\n<p>I would like to know whether my approach is correct or wrong.<br>\nI have dropped the accelometer data columns(.parquet) i.e i didnt use them,because around 2900 values in each of those columns were missing and my score improved from 0.321 to 0.350.</p>\n<p>Also,I have filled missing Sii values by training with remaining data and filled missing sii values.<br>\nand i have filled remaining numerical columns with median.<br>\ni Know that there is lot to be explored in data and above approches arent good,i am here for the suggestions for my above approcahes,<br>\nI would like to accept Suggestions,Thankyou</p>",
  "messages": [
    {
      "id": 3001713,
      "postDate": "2024-09-29T07:10:45.610Z",
      "content": "<p>Hello all,this is my first kaggle competition and i got a score of 0.350.</p>\n<p>I would like to know whether my approach is correct or wrong.<br>\nI have dropped the accelometer data columns(.parquet) i.e i didnt use them,because around 2900 values in each of those columns were missing and my score improved from 0.321 to 0.350.</p>\n<p>Also,I have filled missing Sii values by training with remaining data and filled missing sii values.<br>\nand i have filled remaining numerical columns with median.<br>\ni Know that there is lot to be explored in data and above approches arent good,i am here for the suggestions for my above approcahes,<br>\nI would like to accept Suggestions,Thankyou</p>",
      "rawMarkdown": "Hello all,this is my first kaggle competition and i got a score of 0.350.\n\nI would like to know whether my approach is correct or wrong.\nI have dropped the accelometer data columns(.parquet) i.e i didnt use them,because around 2900 values in each of those columns were missing and my score improved from 0.321 to 0.350.\n\nAlso,I have filled missing Sii values by training with remaining data and filled missing sii values.\nand i have filled remaining numerical columns with median.\ni Know that there is lot to be explored in data and above approches arent good,i am here for the suggestions for my above approcahes,\nI would like to accept Suggestions,Thankyou\n",
      "votes": 1
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "3001713": "Hello all,this is my first kaggle competition and i got a score of 0.350.\n\nI would like to know whether my approach is correct or wrong.\nI have dropped the accelometer data columns(.parquet) i.e i didnt use them,because around 2900 values in each of those columns were missing and my score improved from 0.321 to 0.350.\n\nAlso,I have filled missing Sii values by training with remaining data and filled missing sii values.\nand i have filled remaining numerical columns with median.\ni Know that there is lot to be explored in data and above approches arent good,i am here for the suggestions for my above approcahes,\nI would like to accept Suggestions,Thankyou\n"
  }
}