{
  "id": 182313,
  "title": "Private datasets need to be made public for submission?",
  "url": "/competitions/landmark-recognition-2020/discussion/182313",
  "author_name": "",
  "post_date": "2020-09-12T07:02:42.471224400Z",
  "votes": 1,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I have a doubt regarding final submissions for scorings:-</p>\n<p>As of now we are storing our pre-trained models or some embeddings/vectors(generated offline) in a private dataset on Kaggle.</p>\n<p>Do we need to make them public for final score calculations?</p>",
  "messages": [
    {
      "id": "1007447",
      "postDate": "09/12/2020 07:02:42",
      "content": "<p>I have a doubt regarding final submissions for scorings:-</p>\n<p>As of now we are storing our pre-trained models or some embeddings/vectors(generated offline) in a private dataset on Kaggle.</p>\n<p>Do we need to make them public for final score calculations?</p>",
      "rawMarkdown": "I have a doubt regarding final submissions for scorings:-\n\nAs of now we are storing our pre-trained models or some embeddings/vectors(generated offline) in a private dataset on Kaggle.\n\nDo we need to make them public for final score calculations?",
      "votes": null
    },
    {
      "id": "1007658",
      "postDate": "09/12/2020 11:04:52",
      "content": "<p>Hello <a href=\"https://www.kaggle.com/rohitdeepu17\" target=\"_blank\">@rohitdeepu17</a></p>\n<p>Yes, you need to make the datasets public. Please refer to the <a href=\"https://www.kaggle.com/c/landmark-recognition-2020/overview/code-requirements\" target=\"_blank\">Code requirements section</a></p>\n<blockquote>\n  <p>Freely &amp; publicly available external data is allowed, including pre-trained models</p>\n</blockquote>",
      "rawMarkdown": "Hello @rohitdeepu17\n\nYes, you need to make the datasets public. Please refer to the [Code requirements section](https://www.kaggle.com/c/landmark-recognition-2020/overview/code-requirements)\n\n> Freely & publicly available external data is allowed, including pre-trained models",
      "votes": null
    },
    {
      "id": "1010173",
      "postDate": "09/14/2020 14:58:49",
      "content": "<p>If you created the pre-trained model, you do <strong>not</strong> need to share it publicly. Anything you create (code, embeddings, models) based on the provided competition data is considered part of your approach and not external data.</p>",
      "rawMarkdown": "If you created the pre-trained model, you do **not** need to share it publicly. Anything you create (code, embeddings, models) based on the provided competition data is considered part of your approach and not external data.",
      "votes": null
    },
    {
      "id": "1012466",
      "postDate": "09/16/2020 05:37:03",
      "content": "<p><a href=\"https://www.kaggle.com/wcukierski\" target=\"_blank\">@wcukierski</a> thanks for your reply</p>",
      "rawMarkdown": "wcukierski thanks for your reply",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1007658,
      "author_name": "trikialaaa",
      "author_url": "",
      "post_date": "09/12/2020 11:04:52",
      "content": "<p>Hello <a href=\"https://www.kaggle.com/rohitdeepu17\" target=\"_blank\">@rohitdeepu17</a></p>\n<p>Yes, you need to make the datasets public. Please refer to the <a href=\"https://www.kaggle.com/c/landmark-recognition-2020/overview/code-requirements\" target=\"_blank\">Code requirements section</a></p>\n<blockquote>\n  <p>Freely &amp; publicly available external data is allowed, including pre-trained models</p>\n</blockquote>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1010173,
      "author_name": "wcukierski",
      "author_url": "",
      "post_date": "09/14/2020 14:58:49",
      "content": "<p>If you created the pre-trained model, you do <strong>not</strong> need to share it publicly. Anything you create (code, embeddings, models) based on the provided competition data is considered part of your approach and not external data.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1012466,
          "author_name": "nishaaggarwal",
          "author_url": "",
          "post_date": "09/16/2020 05:37:03",
          "content": "<p><a href=\"https://www.kaggle.com/wcukierski\" target=\"_blank\">@wcukierski</a> thanks for your reply</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1007447": "I have a doubt regarding final submissions for scorings:-\n\nAs of now we are storing our pre-trained models or some embeddings/vectors(generated offline) in a private dataset on Kaggle.\n\nDo we need to make them public for final score calculations?",
    "1007658": "Hello @rohitdeepu17\n\nYes, you need to make the datasets public. Please refer to the [Code requirements section](https://www.kaggle.com/c/landmark-recognition-2020/overview/code-requirements)\n\n> Freely & publicly available external data is allowed, including pre-trained models",
    "1010173": "If you created the pre-trained model, you do **not** need to share it publicly. Anything you create (code, embeddings, models) based on the provided competition data is considered part of your approach and not external data.",
    "1012466": "wcukierski thanks for your reply"
  },
  "source": "meta"
}