{
  "id": 202367,
  "title": "More encoders available with segmentation models",
  "url": "/competitions/hubmap-kidney-segmentation/discussion/202367",
  "author_name": "",
  "post_date": "2020-12-09T16:52:41.771189200Z",
  "votes": 2,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hi all,</p>\n<p>I've noticed that some people are using <a href=\"https://www.kaggle.com/pavel92\" target=\"_blank\">@pavel92</a>  repo <a href=\"https://github.com/qubvel/segmentation_models.pytorch\" target=\"_blank\">https://github.com/qubvel/segmentation_models.pytorch</a> in this competition.<br>\nRecently, some additional encoders were added. I am running some test, but preliminary results (on two different datasets) shows that resnest encoders are the best :)<br>\nHope it helps! </p>",
  "messages": [
    {
      "id": "1107390",
      "postDate": "12/09/2020 16:52:41",
      "content": "<p>Hi all,</p>\n<p>I've noticed that some people are using <a href=\"https://www.kaggle.com/pavel92\" target=\"_blank\">@pavel92</a>  repo <a href=\"https://github.com/qubvel/segmentation_models.pytorch\" target=\"_blank\">https://github.com/qubvel/segmentation_models.pytorch</a> in this competition.<br>\nRecently, some additional encoders were added. I am running some test, but preliminary results (on two different datasets) shows that resnest encoders are the best :)<br>\nHope it helps! </p>",
      "rawMarkdown": "Hi all,\n\nI've noticed that some people are using @pavel92  repo https://github.com/qubvel/segmentation_models.pytorch in this competition.\nRecently, some additional encoders were added. I am running some test, but preliminary results (on two different datasets) shows that resnest encoders are the best :)\nHope it helps!",
      "votes": null
    },
    {
      "id": "1117912",
      "postDate": "12/18/2020 15:25:27",
      "content": "<p>did you compare them with efficientnet and google net encoder?<br>\nand also does using Tensorflow instead of pytorch version of qubvel segmentation models have any disadvantage/advantage?<br>\nI am curious about the loss function you are using</p>\n<p>Thanks</p>",
      "rawMarkdown": "did you compare them with efficientnet and google net encoder?\nand also does using Tensorflow instead of pytorch version of qubvel segmentation models have any disadvantage/advantage?\nI am curious about the loss function you are using\n\nThanks",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1117912,
      "author_name": "sagarkarki136",
      "author_url": "",
      "post_date": "12/18/2020 15:25:27",
      "content": "<p>did you compare them with efficientnet and google net encoder?<br>\nand also does using Tensorflow instead of pytorch version of qubvel segmentation models have any disadvantage/advantage?<br>\nI am curious about the loss function you are using</p>\n<p>Thanks</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1107390": "Hi all,\n\nI've noticed that some people are using @pavel92  repo https://github.com/qubvel/segmentation_models.pytorch in this competition.\nRecently, some additional encoders were added. I am running some test, but preliminary results (on two different datasets) shows that resnest encoders are the best :)\nHope it helps!",
    "1117912": "did you compare them with efficientnet and google net encoder?\nand also does using Tensorflow instead of pytorch version of qubvel segmentation models have any disadvantage/advantage?\nI am curious about the loss function you are using\n\nThanks"
  },
  "source": "meta"
}