{
  "id": 175906,
  "title": "3D ConvModel",
  "url": "/competitions/osic-pulmonary-fibrosis-progression/discussion/175906",
  "author_name": "",
  "post_date": "2020-08-19T19:41:58.287092900Z",
  "votes": 2,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hi,</p>\n<p>I am trying a 3D ConvModel with several data augmentations, but it is not working to me, i only want to predict first FCV from Scan images, has anyone the same trouble?</p>\n<p>Maybe there are few patients or maybe im doing something wrong.</p>\n<p>Thanks in advance</p>",
  "messages": [
    {
      "id": "977913",
      "postDate": "08/19/2020 19:41:58",
      "content": "<p>Hi,</p>\n<p>I am trying a 3D ConvModel with several data augmentations, but it is not working to me, i only want to predict first FCV from Scan images, has anyone the same trouble?</p>\n<p>Maybe there are few patients or maybe im doing something wrong.</p>\n<p>Thanks in advance</p>",
      "rawMarkdown": "Hi,\n\nI am trying a 3D ConvModel with several data augmentations, but it is not working to me, i only want to predict first FCV from Scan images, has anyone the same trouble?\n\nMaybe there are few patients or maybe im doing something wrong.\n\nThanks in advance",
      "votes": null
    },
    {
      "id": "978127",
      "postDate": "08/20/2020 01:08:51",
      "content": "<p>I don't think traditional deep learning of the lung volumes will produce good results. There is simply not enough training data. Only a hundred-something samples cannot train a CNN which likely has hundreds of thousands of parameters.</p>",
      "rawMarkdown": "I don't think traditional deep learning of the lung volumes will produce good results. There is simply not enough training data. Only a hundred-something samples cannot train a CNN which likely has hundreds of thousands of parameters.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 978127,
      "author_name": "matthewmasters",
      "author_url": "",
      "post_date": "08/20/2020 01:08:51",
      "content": "<p>I don't think traditional deep learning of the lung volumes will produce good results. There is simply not enough training data. Only a hundred-something samples cannot train a CNN which likely has hundreds of thousands of parameters.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "977913": "Hi,\n\nI am trying a 3D ConvModel with several data augmentations, but it is not working to me, i only want to predict first FCV from Scan images, has anyone the same trouble?\n\nMaybe there are few patients or maybe im doing something wrong.\n\nThanks in advance",
    "978127": "I don't think traditional deep learning of the lung volumes will produce good results. There is simply not enough training data. Only a hundred-something samples cannot train a CNN which likely has hundreds of thousands of parameters."
  },
  "source": "meta"
}