{
  "id": 421546,
  "title": "Best way to ensemble models from different folds?",
  "url": "/competitions/hubmap-hacking-the-human-vasculature/discussion/421546",
  "author_name": "",
  "post_date": "2023-07-05T18:57:33.620841100Z",
  "votes": 5,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Since the model ensemble could be a way to boost LB, may I ask if someone can share some ideas or strategies for ensembling models, like 5 models from 5 different fold training? Any ideas or opinions are much appreciated!</p>",
  "messages": [
    {
      "id": "2331768",
      "postDate": "07/05/2023 18:57:33",
      "content": "<p>Since the model ensemble could be a way to boost LB, may I ask if someone can share some ideas or strategies for ensembling models, like 5 models from 5 different fold training? Any ideas or opinions are much appreciated!</p>",
      "rawMarkdown": "Since the model ensemble could be a way to boost LB, may I ask if someone can share some ideas or strategies for ensembling models, like 5 models from 5 different fold training? Any ideas or opinions are much appreciated!",
      "votes": null
    },
    {
      "id": "2332056",
      "postDate": "07/06/2023 00:00:05",
      "content": "<p>WBF (<a href=\"https://www.kaggle.com/code/mistag/sartorius-tta-with-weighted-segments-fusion\" target=\"_blank\">https://www.kaggle.com/code/mistag/sartorius-tta-with-weighted-segments-fusion</a>) soft nms (<a href=\"https://arxiv.org/abs/1704.04503\" target=\"_blank\">https://arxiv.org/abs/1704.04503</a>)</p>",
      "rawMarkdown": "WBF (https://www.kaggle.com/code/mistag/sartorius-tta-with-weighted-segments-fusion) soft nms (https://arxiv.org/abs/1704.04503)",
      "votes": null
    },
    {
      "id": "2332062",
      "postDate": "07/06/2023 00:07:00",
      "content": "<p>Thank you so much! I'll take a look at them!</p>",
      "rawMarkdown": "Thank you so much! I'll take a look at them!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2332056,
      "author_name": "maksimovka",
      "author_url": "",
      "post_date": "07/06/2023 00:00:05",
      "content": "<p>WBF (<a href=\"https://www.kaggle.com/code/mistag/sartorius-tta-with-weighted-segments-fusion\" target=\"_blank\">https://www.kaggle.com/code/mistag/sartorius-tta-with-weighted-segments-fusion</a>) soft nms (<a href=\"https://arxiv.org/abs/1704.04503\" target=\"_blank\">https://arxiv.org/abs/1704.04503</a>)</p>",
      "votes": null,
      "replies": [
        {
          "id": 2332062,
          "author_name": "xstargate",
          "author_url": "",
          "post_date": "07/06/2023 00:07:00",
          "content": "<p>Thank you so much! I'll take a look at them!</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2331768": "Since the model ensemble could be a way to boost LB, may I ask if someone can share some ideas or strategies for ensembling models, like 5 models from 5 different fold training? Any ideas or opinions are much appreciated!",
    "2332056": "WBF (https://www.kaggle.com/code/mistag/sartorius-tta-with-weighted-segments-fusion) soft nms (https://arxiv.org/abs/1704.04503)",
    "2332062": "Thank you so much! I'll take a look at them!"
  },
  "source": "meta"
}