{
  "id": 548818,
  "title": "Did anyone just take the route of dropping columns",
  "url": "/competitions/brist1d/discussion/548818",
  "author_name": "",
  "post_date": "2024-11-29T02:48:47.210953700Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I know the competition is almost at an end I couldn't help but notice that a lot of columns were just noise in the DF, once I dropped columns that had 80% and above of empty columns my model performance jumped, but I wouldn't be able to use time series anymore</p>",
  "messages": [
    {
      "id": "3058088",
      "postDate": "11/29/2024 02:48:47",
      "content": "<p>I know the competition is almost at an end I couldn't help but notice that a lot of columns were just noise in the DF, once I dropped columns that had 80% and above of empty columns my model performance jumped, but I wouldn't be able to use time series anymore</p>",
      "rawMarkdown": "I know the competition is almost at an end I couldn't help but notice that a lot of columns were just noise in the DF, once I dropped columns that had 80% and above of empty columns my model performance jumped, but I wouldn't be able to use time series anymore",
      "votes": null
    },
    {
      "id": "3058394",
      "postDate": "11/29/2024 12:29:39",
      "content": "<p>You can drop most of the columns and still have 'time series'. It depends on which columns you are using. You can also make an ensemble of both models.</p>",
      "rawMarkdown": "You can drop most of the columns and still have 'time series'. It depends on which columns you are using. You can also make an ensemble of both models.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3058394,
      "author_name": "alrjandrorivasespa",
      "author_url": "",
      "post_date": "11/29/2024 12:29:39",
      "content": "<p>You can drop most of the columns and still have 'time series'. It depends on which columns you are using. You can also make an ensemble of both models.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3058088": "I know the competition is almost at an end I couldn't help but notice that a lot of columns were just noise in the DF, once I dropped columns that had 80% and above of empty columns my model performance jumped, but I wouldn't be able to use time series anymore",
    "3058394": "You can drop most of the columns and still have 'time series'. It depends on which columns you are using. You can also make an ensemble of both models."
  },
  "source": "meta"
}