{
  "id": 427986,
  "title": "Ensemble learning in segmentation",
  "url": "/competitions/google-research-identify-contrails-reduce-global-warming/discussion/427986",
  "author_name": "Monkey D Donut",
  "post_date": "2023-07-30T14:46:52.670000",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>hi :)<br>\ni tried to ensembel unet and unet++ (both models are trained on same data). there was no siginificant improvement . which ensembel methods works best in case of segmentation tasks generally ? is ensembel learning common in segmentation ?</p>\n<p>thank you :)</p>",
  "messages": [
    {
      "id": 2366334,
      "postDate": "2023-07-30T23:23:45.127Z",
      "content": "<p>Hello, I suggest you experiement with different encoders, different fold seeds and different hyperparams in general.</p>",
      "rawMarkdown": "Hello, I suggest you experiement with different encoders, different fold seeds and different hyperparams in general.",
      "votes": 1,
      "replies": [
        {
          "id": 2366602,
          "postDate": "2023-07-31T05:31:35.983Z",
          "content": "<p>thank you , i will try those :)</p>",
          "rawMarkdown": "thank you , i will try those :)"
        }
      ]
    },
    {
      "id": 2365904,
      "postDate": "2023-07-30T14:46:52.670Z",
      "content": "<p>hi :)<br>\ni tried to ensembel unet and unet++ (both models are trained on same data). there was no siginificant improvement . which ensembel methods works best in case of segmentation tasks generally ? is ensembel learning common in segmentation ?</p>\n<p>thank you :)</p>",
      "rawMarkdown": "hi :)\ni tried to ensembel unet and unet++ (both models are trained on same data). there was no siginificant improvement . which ensembel methods works best in case of segmentation tasks generally ? is ensembel learning common in segmentation ?\n\nthank you :)",
      "votes": 1
    }
  ],
  "comments": [
    {
      "id": 2366334,
      "author_name": "JEANMPIA",
      "author_url": "",
      "post_date": "2023-07-30T23:23:45.127000",
      "content": "<p>Hello, I suggest you experiement with different encoders, different fold seeds and different hyperparams in general.</p>",
      "votes": 1,
      "replies": [
        {
          "id": 2366602,
          "author_name": "Monkey D Donut",
          "author_url": "",
          "post_date": "2023-07-31T05:31:35.983000",
          "content": "<p>thank you , i will try those :)</p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2366334": "Hello, I suggest you experiement with different encoders, different fold seeds and different hyperparams in general.",
    "2365904": "hi :)\ni tried to ensembel unet and unet++ (both models are trained on same data). there was no siginificant improvement . which ensembel methods works best in case of segmentation tasks generally ? is ensembel learning common in segmentation ?\n\nthank you :)"
  }
}