{
  "id": 566595,
  "title": "36th place Solution",
  "content": "<p>I recently felt like I was going through a slump, and I didn't have high expectations, but fortunately, I managed to secure 36th place.</p>\n<p>Here are my three solutions:</p>\n<ol>\n<li>NN Pairwise Ranking</li>\n<li>Classifier + CAT + XGB</li>\n<li>XGBoost with Monotonicity</li>\n</ol>\n<p>You can check out the solutions in the following notebooks.</p>\n<p><a href=\"https://www.kaggle.com/code/junhanzangai/cibmtr-nn-lgb-cat-ensemble-of-solutions\" target=\"_blank\">https://www.kaggle.com/code/junhanzangai/cibmtr-nn-lgb-cat-ensemble-of-solutions</a></p>\n<p>Additionally, using MinMaxScaler to normalize each prediction to a 0–1 range and making predictions based on this normalization was a crucial factor. You can see this clearly in the following notebook.</p>\n<p>I believe my score could have improved further, if I incorporated DeepResidualNN into my approach.</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F13540117%2F41eb8f7f34d503acbf0718c4eb98fbc4%2F2025-03-06%20154008.png?generation=1741243271464164&amp;alt=media\" alt=\"\"></p>",
  "comments": [],
  "comment_count": 0,
  "messages": [
    {
      "id": 3142163,
      "postDate": "2025-03-06T06:37:49.720Z",
      "content": "<p>I recently felt like I was going through a slump, and I didn't have high expectations, but fortunately, I managed to secure 36th place.</p>\n<p>Here are my three solutions:</p>\n<ol>\n<li>NN Pairwise Ranking</li>\n<li>Classifier + CAT + XGB</li>\n<li>XGBoost with Monotonicity</li>\n</ol>\n<p>You can check out the solutions in the following notebooks.</p>\n<p><a href=\"https://www.kaggle.com/code/junhanzangai/cibmtr-nn-lgb-cat-ensemble-of-solutions\" target=\"_blank\">https://www.kaggle.com/code/junhanzangai/cibmtr-nn-lgb-cat-ensemble-of-solutions</a></p>\n<p>Additionally, using MinMaxScaler to normalize each prediction to a 0–1 range and making predictions based on this normalization was a crucial factor. You can see this clearly in the following notebook.</p>\n<p>I believe my score could have improved further, if I incorporated DeepResidualNN into my approach.</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F13540117%2F41eb8f7f34d503acbf0718c4eb98fbc4%2F2025-03-06%20154008.png?generation=1741243271464164&amp;alt=media\" alt=\"\"></p>",
      "rawMarkdown": "I recently felt like I was going through a slump, and I didn't have high expectations, but fortunately, I managed to secure 36th place.\n\nHere are my three solutions:\n\n1. NN Pairwise Ranking\n2. Classifier + CAT + XGB\n3. XGBoost with Monotonicity\n\nYou can check out the solutions in the following notebooks.\n\nhttps://www.kaggle.com/code/junhanzangai/cibmtr-nn-lgb-cat-ensemble-of-solutions\n\nAdditionally, using MinMaxScaler to normalize each prediction to a 0–1 range and making predictions based on this normalization was a crucial factor. You can see this clearly in the following notebook.\n\nI believe my score could have improved further, if I incorporated DeepResidualNN into my approach.\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F13540117%2F41eb8f7f34d503acbf0718c4eb98fbc4%2F2025-03-06%20154008.png?generation=1741243271464164&alt=media)",
      "votes": 5
    }
  ],
  "complete": true,
  "source": "live",
  "fetched_at": "2026-10-19T19:47:39Z",
  "competition_id": 70942,
  "competition_slug": "equity-post-HCT-survival-predictions",
  "team_id": 13191615,
  "solution_url": "/competitions/equity-post-HCT-survival-predictions/writeups/code-hacker-36th-place-solution"
}