{
  "id": 463723,
  "title": "Where to start",
  "url": "/competitions/blood-vessel-segmentation/discussion/463723",
  "author_name": "",
  "post_date": "2023-12-26T19:36:15.708665200Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hi, I was wondering where I can start? I have not done a competition before based around images, and I was wondering what model I can start out with? Thank you</p>",
  "messages": [
    {
      "id": "2575393",
      "postDate": "12/26/2023 19:36:15",
      "content": "<p>Hi, I was wondering where I can start? I have not done a competition before based around images, and I was wondering what model I can start out with? Thank you</p>",
      "rawMarkdown": "Hi, I was wondering where I can start? I have not done a competition before based around images, and I was wondering what model I can start out with? Thank you",
      "votes": null
    },
    {
      "id": "2576172",
      "postDate": "12/27/2023 14:29:01",
      "content": "<p>You can start from segmentation models pytorch and UnetPlusPlus with simple resnet as a backbone. The important part is image preprocessing, it means image standardization or normalization, image and label resizing (check bilinear mode for image and nearest for label resizing).</p>",
      "rawMarkdown": "You can start from segmentation models pytorch and UnetPlusPlus with simple resnet as a backbone. The important part is image preprocessing, it means image standardization or normalization, image and label resizing (check bilinear mode for image and nearest for label resizing).",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2576172,
      "author_name": "jankowalski2000",
      "author_url": "",
      "post_date": "12/27/2023 14:29:01",
      "content": "<p>You can start from segmentation models pytorch and UnetPlusPlus with simple resnet as a backbone. The important part is image preprocessing, it means image standardization or normalization, image and label resizing (check bilinear mode for image and nearest for label resizing).</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2575393": "Hi, I was wondering where I can start? I have not done a competition before based around images, and I was wondering what model I can start out with? Thank you",
    "2576172": "You can start from segmentation models pytorch and UnetPlusPlus with simple resnet as a backbone. The important part is image preprocessing, it means image standardization or normalization, image and label resizing (check bilinear mode for image and nearest for label resizing)."
  },
  "source": "meta"
}