{
  "id": 122649,
  "title": "3th place solution - I CAn'T believe I won",
  "content": "<p><strong>Model</strong>\nLogistic regression\nC=0.095, class_weight={0: 1, 1: 1.4}, tol=0.00001,\nsolver='liblinear', penalty='l2'</p>\n\n<p><strong>Encoding</strong>\nbinary: 0 and 1\nnominal, month and day: one hot encoding\nordinal: ordinal encoding</p>\n\n<p><strong>Unseen value handling (only for nom_7, nom_8 and nom_9)</strong>\nThe features which are not found in both training and testing data will be grouped into one class, called \"other\". For nom_9, there are many sparse features (counts of the features are small). To prevent overfitting, we also group these sparse features into the \"other\" class describe above. A featurea in nom_9 with count &lt; 3 (a parameter needs to be tuned) will be treated as sparse features</p>",
  "comments": [
    {
      "id": 701267,
      "author_name": "aldparis",
      "author_url": "",
      "post_date": "2019-12-23T09:53:22.447000",
      "content": "<p>Congratulations and thank you for sharing !</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 700358,
      "author_name": "Youkyoung",
      "author_url": "",
      "post_date": "2019-12-21T20:52:51.313000",
      "content": "<p>Congratulations!!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 722763,
      "author_name": "Vetelus",
      "author_url": "",
      "post_date": "2020-01-19T03:49:20.927000",
      "content": "<p>Thank you!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 710967,
      "author_name": "Akahachi",
      "author_url": "",
      "post_date": "2020-01-05T13:43:30.510000",
      "content": "<p>Thank you for sharing🙏 </p>",
      "votes": 0,
      "replies": []
    }
  ],
  "comment_count": 4,
  "messages": [
    {
      "id": 700272,
      "postDate": "2019-12-21T18:03:35.120Z",
      "content": "<p><strong>Model</strong>\nLogistic regression\nC=0.095, class_weight={0: 1, 1: 1.4}, tol=0.00001,\nsolver='liblinear', penalty='l2'</p>\n\n<p><strong>Encoding</strong>\nbinary: 0 and 1\nnominal, month and day: one hot encoding\nordinal: ordinal encoding</p>\n\n<p><strong>Unseen value handling (only for nom_7, nom_8 and nom_9)</strong>\nThe features which are not found in both training and testing data will be grouped into one class, called \"other\". For nom_9, there are many sparse features (counts of the features are small). To prevent overfitting, we also group these sparse features into the \"other\" class describe above. A featurea in nom_9 with count &lt; 3 (a parameter needs to be tuned) will be treated as sparse features</p>",
      "rawMarkdown": "**Model**\nLogistic regression\nC=0.095, class_weight={0: 1, 1: 1.4}, tol=0.00001,\nsolver='liblinear', penalty='l2'\n\n\n**Encoding**\nbinary: 0 and 1\nnominal, month and day: one hot encoding\nordinal: ordinal encoding\n\n\n**Unseen value handling (only for nom_7, nom_8 and nom_9)**\nThe features which are not found in both training and testing data will be grouped into one class, called \"other\". For nom_9, there are many sparse features (counts of the features are small). To prevent overfitting, we also group these sparse features into the \"other\" class describe above. A featurea in nom_9 with count &lt; 3 (a parameter needs to be tuned) will be treated as sparse features\n",
      "votes": 7
    },
    {
      "id": 701267,
      "postDate": "2019-12-23T09:53:22.447Z",
      "content": "<p>Congratulations and thank you for sharing !</p>",
      "rawMarkdown": "Congratulations and thank you for sharing !"
    },
    {
      "id": 700358,
      "postDate": "2019-12-21T20:52:51.313Z",
      "content": "<p>Congratulations!!</p>",
      "rawMarkdown": "Congratulations!!"
    },
    {
      "id": 722763,
      "postDate": "2020-01-19T03:49:20.927Z",
      "content": "<p>Thank you!</p>",
      "rawMarkdown": "Thank you!"
    },
    {
      "id": 710967,
      "postDate": "2020-01-05T13:43:30.510Z",
      "content": "<p>Thank you for sharing🙏 </p>",
      "rawMarkdown": "Thank you for sharing🙏 "
    }
  ],
  "complete": true,
  "source": "live",
  "fetched_at": "2026-10-19T17:33:28Z",
  "competition_id": 14999,
  "competition_slug": "cat-in-the-dat",
  "team_id": 3628937,
  "solution_url": "/competitions/cat-in-the-dat/writeups/hardworking-kaggler-3th-place-solution-i-can-t-bel"
}