{
  "id": 511850,
  "title": "What to do with coordinates",
  "url": "/competitions/rsna-2024-lumbar-spine-degenerative-classification/discussion/511850",
  "author_name": "",
  "post_date": "2024-06-12T11:11:46.933324400Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>How do I use coordinates in the training?<br>\nWhat I am doing right now is to use pretrained model for the classification of 3 classes. But there, I just input the images, apply the transformation and get the inference for the submission. Can anyone pls provide me pseudo process of using those coordinates in this problem? Do I need to get the ROI and use it somehow for in the classification process? Please share your thoughts. Thanks</p>",
  "messages": [
    {
      "id": "2868414",
      "postDate": "06/12/2024 11:11:46",
      "content": "<p>How do I use coordinates in the training?<br>\nWhat I am doing right now is to use pretrained model for the classification of 3 classes. But there, I just input the images, apply the transformation and get the inference for the submission. Can anyone pls provide me pseudo process of using those coordinates in this problem? Do I need to get the ROI and use it somehow for in the classification process? Please share your thoughts. Thanks</p>",
      "rawMarkdown": "How do I use coordinates in the training?\nWhat I am doing right now is to use pretrained model for the classification of 3 classes. But there, I just input the images, apply the transformation and get the inference for the submission. Can anyone pls provide me pseudo process of using those coordinates in this problem? Do I need to get the ROI and use it somehow for in the classification process? Please share your thoughts. Thanks",
      "votes": null
    },
    {
      "id": "2868666",
      "postDate": "06/12/2024 14:49:00",
      "content": "<p>You can use them as two values in output to support classification [class1, class2, class3, X,Y] but if you resize the image make sure to multiple height and width by factor.</p>",
      "rawMarkdown": "You can use them as two values in output to support classification [class1, class2, class3, X,Y] but if you resize the image make sure to multiple height and width by factor.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2868666,
      "author_name": "manaidu",
      "author_url": "",
      "post_date": "06/12/2024 14:49:00",
      "content": "<p>You can use them as two values in output to support classification [class1, class2, class3, X,Y] but if you resize the image make sure to multiple height and width by factor.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2868414": "How do I use coordinates in the training?\nWhat I am doing right now is to use pretrained model for the classification of 3 classes. But there, I just input the images, apply the transformation and get the inference for the submission. Can anyone pls provide me pseudo process of using those coordinates in this problem? Do I need to get the ROI and use it somehow for in the classification process? Please share your thoughts. Thanks",
    "2868666": "You can use them as two values in output to support classification [class1, class2, class3, X,Y] but if you resize the image make sure to multiple height and width by factor."
  },
  "source": "meta"
}