{
  "id": 422411,
  "title": "TTA for mask-rcnn",
  "url": "/competitions/hubmap-hacking-the-human-vasculature/discussion/422411",
  "author_name": "",
  "post_date": "2023-07-09T18:08:33.945038700Z",
  "votes": 5,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hi, I've seen some people use TTA, but I was wondering how can we use it for instance segmentation models like mask rcnn now, when there are multiple masks being predicted? Should I just combine them all into a single mask and then average? Thanks.</p>",
  "messages": [
    {
      "id": "2336923",
      "postDate": "07/09/2023 18:08:33",
      "content": "<p>Hi, I've seen some people use TTA, but I was wondering how can we use it for instance segmentation models like mask rcnn now, when there are multiple masks being predicted? Should I just combine them all into a single mask and then average? Thanks.</p>",
      "rawMarkdown": "Hi, I've seen some people use TTA, but I was wondering how can we use it for instance segmentation models like mask rcnn now, when there are multiple masks being predicted? Should I just combine them all into a single mask and then average? Thanks.",
      "votes": null
    },
    {
      "id": "2346490",
      "postDate": "07/16/2023 09:41:07",
      "content": "<p>That's one way to go.<br>\nYou can also apply TTA to all the masks first and then average.<br>\nOf course, the only way to decide is by checking the CV score.</p>",
      "rawMarkdown": "That's one way to go.\nYou can also apply TTA to all the masks first and then average.\nOf course, the only way to decide is by checking the CV score.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2346490,
      "author_name": "yassinealouini",
      "author_url": "",
      "post_date": "07/16/2023 09:41:07",
      "content": "<p>That's one way to go.<br>\nYou can also apply TTA to all the masks first and then average.<br>\nOf course, the only way to decide is by checking the CV score.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2336923": "Hi, I've seen some people use TTA, but I was wondering how can we use it for instance segmentation models like mask rcnn now, when there are multiple masks being predicted? Should I just combine them all into a single mask and then average? Thanks.",
    "2346490": "That's one way to go.\nYou can also apply TTA to all the masks first and then average.\nOf course, the only way to decide is by checking the CV score."
  },
  "source": "meta"
}