{
  "id": 115807,
  "title": "I want to know augmentation for this task",
  "url": "/competitions/pku-autonomous-driving/discussion/115807",
  "author_name": "",
  "post_date": "2019-11-05T11:52:32.972797400Z",
  "votes": 5,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I'm challenging the object detection competition for the first time. Is data augmentation performed for object detection?\nFor example, mixup works well for normal classification tasks, but is object detection possible?</p>",
  "messages": [
    {
      "id": "665782",
      "postDate": "11/05/2019 11:52:32",
      "content": "<p>I'm challenging the object detection competition for the first time. Is data augmentation performed for object detection?\nFor example, mixup works well for normal classification tasks, but is object detection possible?</p>",
      "rawMarkdown": "I'm challenging the object detection competition for the first time. Is data augmentation performed for object detection?\nFor example, mixup works well for normal classification tasks, but is object detection possible?",
      "votes": null
    },
    {
      "id": "671641",
      "postDate": "11/13/2019 01:23:37",
      "content": "<p>As you can see, only flipping is the augumentation of public best kernel.\nNow LB scoring is discussed to be wrong, then we cannot conclude that augumentation high scoring at lb works well,\nbut about Local cv I'm just testing  :)\n<a href=\"https://www.kaggle.com/hocop1/centernet-baseline\">https://www.kaggle.com/hocop1/centernet-baseline</a></p>",
      "rawMarkdown": "As you can see, only flipping is the augumentation of public best kernel.\nNow LB scoring is discussed to be wrong, then we cannot conclude that augumentation high scoring at lb works well,\nbut about Local cv I'm just testing  :)\nhttps://www.kaggle.com/hocop1/centernet-baseline",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 671641,
      "author_name": "ryunosukeishizaki",
      "author_url": "",
      "post_date": "11/13/2019 01:23:37",
      "content": "<p>As you can see, only flipping is the augumentation of public best kernel.\nNow LB scoring is discussed to be wrong, then we cannot conclude that augumentation high scoring at lb works well,\nbut about Local cv I'm just testing  :)\n<a href=\"https://www.kaggle.com/hocop1/centernet-baseline\">https://www.kaggle.com/hocop1/centernet-baseline</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "665782": "I'm challenging the object detection competition for the first time. Is data augmentation performed for object detection?\nFor example, mixup works well for normal classification tasks, but is object detection possible?",
    "671641": "As you can see, only flipping is the augumentation of public best kernel.\nNow LB scoring is discussed to be wrong, then we cannot conclude that augumentation high scoring at lb works well,\nbut about Local cv I'm just testing  :)\nhttps://www.kaggle.com/hocop1/centernet-baseline"
  },
  "source": "meta"
}