{
  "id": 501309,
  "title": "Welcome to this new edition of the tabular benchmark competition",
  "url": "/competitions/bench-tab-v2/discussion/501309",
  "author_name": "Lucas Morin",
  "post_date": "2024-05-08T18:36:48.966000",
  "votes": 6,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Welcome to this new edition of the tabular benchmark competition</p>\n<p>Hello everyone and welcome to this custom competition. I aim is to provide a long and diverse, tabular data competition, that could be both a learning place for newcomers and a benchmark to try and evaluate more advanced models.</p>\n<p>The competition is actually a 59-in-1 as the source for the datasets is the benchmark of the original paper: \"Why do tree-based models still outperform deep learning on tabular data?\". This should provide a diverse ensemble of tasks so that we can discuss general findings about tabular ML.</p>\n<p>I'd like to thanks <a href=\"https://www.kaggle.com/carlmcbrideellis\" target=\"_blank\">@carlmcbrideellis</a> for beta-testing the first version of the competition and <a href=\"https://www.kaggle.com/leogri\" target=\"_blank\">@leogri</a> for discussions about the original paper. I would also like to thank the deepai image generator for the somewhat creepy illustration.</p>\n<p>I hope you all enjoy this competition. Let's build, learn, and share new stuff about tabular ML!</p>",
  "messages": [
    {
      "id": 2801804,
      "postDate": "2024-05-08T18:36:48.967Z",
      "content": "<p>Welcome to this new edition of the tabular benchmark competition</p>\n<p>Hello everyone and welcome to this custom competition. I aim is to provide a long and diverse, tabular data competition, that could be both a learning place for newcomers and a benchmark to try and evaluate more advanced models.</p>\n<p>The competition is actually a 59-in-1 as the source for the datasets is the benchmark of the original paper: \"Why do tree-based models still outperform deep learning on tabular data?\". This should provide a diverse ensemble of tasks so that we can discuss general findings about tabular ML.</p>\n<p>I'd like to thanks <a href=\"https://www.kaggle.com/carlmcbrideellis\" target=\"_blank\">@carlmcbrideellis</a> for beta-testing the first version of the competition and <a href=\"https://www.kaggle.com/leogri\" target=\"_blank\">@leogri</a> for discussions about the original paper. I would also like to thank the deepai image generator for the somewhat creepy illustration.</p>\n<p>I hope you all enjoy this competition. Let's build, learn, and share new stuff about tabular ML!</p>",
      "rawMarkdown": "Welcome to this new edition of the tabular benchmark competition\n\nHello everyone and welcome to this custom competition. I aim is to provide a long and diverse, tabular data competition, that could be both a learning place for newcomers and a benchmark to try and evaluate more advanced models.\n\nThe competition is actually a 59-in-1 as the source for the datasets is the benchmark of the original paper: \"Why do tree-based models still outperform deep learning on tabular data?\". This should provide a diverse ensemble of tasks so that we can discuss general findings about tabular ML.\n\nI'd like to thanks @carlmcbrideellis for beta-testing the first version of the competition and @leogri for discussions about the original paper. I would also like to thank the deepai image generator for the somewhat creepy illustration.\n\nI hope you all enjoy this competition. Let's build, learn, and share new stuff about tabular ML!",
      "votes": 6
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "2801804": "Welcome to this new edition of the tabular benchmark competition\n\nHello everyone and welcome to this custom competition. I aim is to provide a long and diverse, tabular data competition, that could be both a learning place for newcomers and a benchmark to try and evaluate more advanced models.\n\nThe competition is actually a 59-in-1 as the source for the datasets is the benchmark of the original paper: \"Why do tree-based models still outperform deep learning on tabular data?\". This should provide a diverse ensemble of tasks so that we can discuss general findings about tabular ML.\n\nI'd like to thanks @carlmcbrideellis for beta-testing the first version of the competition and @leogri for discussions about the original paper. I would also like to thank the deepai image generator for the somewhat creepy illustration.\n\nI hope you all enjoy this competition. Let's build, learn, and share new stuff about tabular ML!"
  }
}