{
  "id": 50791,
  "title": "Coco vs Imagenet vs from-scratch",
  "url": "/competitions/data-science-bowl-2018/discussion/50791",
  "author_name": "",
  "post_date": "2018-03-01T09:12:21.731669600Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hi guys,</p>\n\n<p>are you using e.g. Mask-RCNN with trained weights from Imagenet or Coco or do you even train from scratch?</p>",
  "messages": [
    {
      "id": "289270",
      "postDate": "03/01/2018 09:12:21",
      "content": "<p>Hi guys,</p>\n\n<p>are you using e.g. Mask-RCNN with trained weights from Imagenet or Coco or do you even train from scratch?</p>",
      "rawMarkdown": "Hi guys,\n\nare you using e.g. Mask-RCNN with trained weights from Imagenet or Coco or do you even train from scratch?",
      "votes": null
    },
    {
      "id": "290623",
      "postDate": "03/04/2018 15:17:45",
      "content": "<p>I suggest to use imagenet pretrain if you have it. Otherwise, you have to train from scratch</p>",
      "rawMarkdown": "I suggest to use imagenet pretrain if you have it. Otherwise, you have to train from scratch",
      "votes": null
    },
    {
      "id": "293672",
      "postDate": "03/10/2018 11:00:39",
      "content": "<p>The weights are  available in the public domain right? I\n'm just confused by your statement \"if you have it\". Could you please clarify?</p>",
      "rawMarkdown": "The weights are  available in the public domain right? I\n'm just confused by your statement \"if you have it\". Could you please clarify?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 290623,
      "author_name": "john1231983",
      "author_url": "",
      "post_date": "03/04/2018 15:17:45",
      "content": "<p>I suggest to use imagenet pretrain if you have it. Otherwise, you have to train from scratch</p>",
      "votes": null,
      "replies": [
        {
          "id": 293672,
          "author_name": "burhanusm",
          "author_url": "",
          "post_date": "03/10/2018 11:00:39",
          "content": "<p>The weights are  available in the public domain right? I\n'm just confused by your statement \"if you have it\". Could you please clarify?</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "289270": "Hi guys,\n\nare you using e.g. Mask-RCNN with trained weights from Imagenet or Coco or do you even train from scratch?",
    "290623": "I suggest to use imagenet pretrain if you have it. Otherwise, you have to train from scratch",
    "293672": "The weights are  available in the public domain right? I\n'm just confused by your statement \"if you have it\". Could you please clarify?"
  },
  "source": "meta"
}