{
  "id": 123671,
  "title": "Approach for Regr Loss",
  "url": "/competitions/pku-autonomous-driving/discussion/123671",
  "author_name": "",
  "post_date": "2019-12-29T13:32:46.655681600Z",
  "votes": 5,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Is it a good approach to regress YPR and x,y,z together ,what i find that Regr loss fails to converge well,in attempt to further reduce the loss  model overfits.</p>\n\n<p>Would be good if some one can share a right approach to do same. In some of git hub models i find Rotational loss used but have got no idea about it.</p>",
  "messages": [
    {
      "id": "705793",
      "postDate": "12/29/2019 13:32:46",
      "content": "<p>Is it a good approach to regress YPR and x,y,z together ,what i find that Regr loss fails to converge well,in attempt to further reduce the loss  model overfits.</p>\n\n<p>Would be good if some one can share a right approach to do same. In some of git hub models i find Rotational loss used but have got no idea about it.</p>",
      "rawMarkdown": "Is it a good approach to regress YPR and x,y,z together ,what i find that Regr loss fails to converge well,in attempt to further reduce the loss  model overfits.\n\nWould be good if some one can share a right approach to do same. In some of git hub models i find Rotational loss used but have got no idea about it.",
      "votes": null
    },
    {
      "id": "706153",
      "postDate": "12/30/2019 02:09:54",
      "content": "<p>Maybe, need ourself to explore it</p>",
      "rawMarkdown": "Maybe, need ourself to explore it",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 706153,
      "author_name": "cswwp347724",
      "author_url": "",
      "post_date": "12/30/2019 02:09:54",
      "content": "<p>Maybe, need ourself to explore it</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "705793": "Is it a good approach to regress YPR and x,y,z together ,what i find that Regr loss fails to converge well,in attempt to further reduce the loss  model overfits.\n\nWould be good if some one can share a right approach to do same. In some of git hub models i find Rotational loss used but have got no idea about it.",
    "706153": "Maybe, need ourself to explore it"
  },
  "source": "meta"
}