{
  "id": 134428,
  "title": "Do I have to disclose my own model that I use?",
  "url": "/competitions/deepfake-detection-challenge/discussion/134428",
  "author_name": "",
  "post_date": "2020-03-08T03:43:26.896321900Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>So, I realize I need to disclose if I use some kind of open source available model. But as we are encouraged to train our models outside of Kaggle and only then upload them to Kaggle as external data, do we need to openly share our original work and trained model? Or is it only for additional datasets we used, or prebuilt open source models that others created and we are \"burrowing\"?</p>\n\n<p>Thanks. </p>\n\n<p>p.s. links to sources of the rules will be appreciated i.e. \"Rules\" \"Getting started\" etc...</p>",
  "messages": [
    {
      "id": "766359",
      "postDate": "03/08/2020 03:43:26",
      "content": "<p>So, I realize I need to disclose if I use some kind of open source available model. But as we are encouraged to train our models outside of Kaggle and only then upload them to Kaggle as external data, do we need to openly share our original work and trained model? Or is it only for additional datasets we used, or prebuilt open source models that others created and we are \"burrowing\"?</p>\n\n<p>Thanks. </p>\n\n<p>p.s. links to sources of the rules will be appreciated i.e. \"Rules\" \"Getting started\" etc...</p>",
      "rawMarkdown": "So, I realize I need to disclose if I use some kind of open source available model. But as we are encouraged to train our models outside of Kaggle and only then upload them to Kaggle as external data, do we need to openly share our original work and trained model? Or is it only for additional datasets we used, or prebuilt open source models that others created and we are \"burrowing\"?\n\nThanks. \n\np.s. links to sources of the rules will be appreciated i.e. \"Rules\" \"Getting started\" etc...",
      "votes": null
    },
    {
      "id": "766994",
      "postDate": "03/09/2020 02:27:42",
      "content": "<p>Please refer to <a href=\"/juliaelliott\">@juliaelliott</a> answer here <a href=\"https://www.kaggle.com/c/deepfake-detection-challenge/discussion/121203#693874\">https://www.kaggle.com/c/deepfake-detection-challenge/discussion/121203#693874</a></p>",
      "rawMarkdown": "Please refer to @juliaelliott answer here [https://www.kaggle.com/c/deepfake-detection-challenge/discussion/121203#693874](https://www.kaggle.com/c/deepfake-detection-challenge/discussion/121203#693874)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 766994,
      "author_name": "reachkishore",
      "author_url": "",
      "post_date": "03/09/2020 02:27:42",
      "content": "<p>Please refer to <a href=\"/juliaelliott\">@juliaelliott</a> answer here <a href=\"https://www.kaggle.com/c/deepfake-detection-challenge/discussion/121203#693874\">https://www.kaggle.com/c/deepfake-detection-challenge/discussion/121203#693874</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "766359": "So, I realize I need to disclose if I use some kind of open source available model. But as we are encouraged to train our models outside of Kaggle and only then upload them to Kaggle as external data, do we need to openly share our original work and trained model? Or is it only for additional datasets we used, or prebuilt open source models that others created and we are \"burrowing\"?\n\nThanks. \n\np.s. links to sources of the rules will be appreciated i.e. \"Rules\" \"Getting started\" etc...",
    "766994": "Please refer to @juliaelliott answer here [https://www.kaggle.com/c/deepfake-detection-challenge/discussion/121203#693874](https://www.kaggle.com/c/deepfake-detection-challenge/discussion/121203#693874)"
  },
  "source": "meta"
}