{
  "id": 297800,
  "title": "Has anyone tried creating 'fake' train images?",
  "url": "/competitions/sartorius-cell-instance-segmentation/discussion/297800",
  "author_name": "",
  "post_date": "2021-12-29T17:14:50.725859900Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hey,</p>\n<p>Did anyone tried creating 'fake' images from the given train images to increase the amount of train data?</p>\n<p>Thanks!</p>",
  "messages": [
    {
      "id": "1632466",
      "postDate": "12/29/2021 17:14:50",
      "content": "<p>Hey,</p>\n<p>Did anyone tried creating 'fake' images from the given train images to increase the amount of train data?</p>\n<p>Thanks!</p>",
      "rawMarkdown": "Hey,\n\nDid anyone tried creating 'fake' images from the given train images to increase the amount of train data?\n\nThanks!",
      "votes": null
    },
    {
      "id": "1632558",
      "postDate": "12/29/2021 18:56:07",
      "content": "<p>I thought about generating/noising relevant geometrical shapes but given the noise level of the original dataset decided that's not worth it.</p>",
      "rawMarkdown": "I thought about generating/noising relevant geometrical shapes but given the noise level of the original dataset decided that's not worth it.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1632558,
      "author_name": "rednikotin",
      "author_url": "",
      "post_date": "12/29/2021 18:56:07",
      "content": "<p>I thought about generating/noising relevant geometrical shapes but given the noise level of the original dataset decided that's not worth it.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1632466": "Hey,\n\nDid anyone tried creating 'fake' images from the given train images to increase the amount of train data?\n\nThanks!",
    "1632558": "I thought about generating/noising relevant geometrical shapes but given the noise level of the original dataset decided that's not worth it."
  },
  "source": "meta"
}