{
  "id": 296505,
  "title": "can I train my model on my own server?",
  "url": "/competitions/tensorflow-great-barrier-reef/discussion/296505",
  "author_name": "",
  "post_date": "2021-12-22T01:39:54.593363Z",
  "votes": 2,
  "comment_count": 11,
  "views": 0,
  "content": "<p>hi , I just came into contact with the kernel only competition recently. I have a few questions to ask: can I train the model on my own server, and then upload the model and submit the results in the submitted kernel interface?</p>",
  "messages": [
    {
      "id": "1625604",
      "postDate": "12/22/2021 01:39:54",
      "content": "<p>hi , I just came into contact with the kernel only competition recently. I have a few questions to ask: can I train the model on my own server, and then upload the model and submit the results in the submitted kernel interface?</p>",
      "rawMarkdown": "hi , I just came into contact with the kernel only competition recently. I have a few questions to ask: can I train the model on my own server, and then upload the model and submit the results in the submitted kernel interface?",
      "votes": null
    },
    {
      "id": "1625831",
      "postDate": "12/22/2021 07:13:00",
      "content": "<p>yes, yes, yes :)</p>",
      "rawMarkdown": "yes, yes, yes :)",
      "votes": null
    },
    {
      "id": "1625900",
      "postDate": "12/22/2021 08:52:07",
      "content": "<p>Sure! Upload your model checkpoint as a dataset, attach it to the kernel. Only inference notebook should be committed</p>",
      "rawMarkdown": "Sure! Upload your model checkpoint as a dataset, attach it to the kernel. Only inference notebook should be committed",
      "votes": null
    },
    {
      "id": "1626570",
      "postDate": "12/23/2021 02:21:50",
      "content": "<p>thanks a lot!</p>",
      "rawMarkdown": "thanks a lot!",
      "votes": null
    },
    {
      "id": "1626571",
      "postDate": "12/23/2021 02:22:07",
      "content": "<p>thanks you!</p>",
      "rawMarkdown": "thanks you!",
      "votes": null
    },
    {
      "id": "1627009",
      "postDate": "12/23/2021 12:31:22",
      "content": "<p>Yes, you can.</p>",
      "rawMarkdown": "Yes, you can.",
      "votes": null
    },
    {
      "id": "1652694",
      "postDate": "01/16/2022 22:31:41",
      "content": "<p>Reviving this old topic, given the rules state \"Freely &amp; publicly available external data is allowed, including pre-trained models\", does that mean pre-trained models that I trained locally and uploaded as a \"private dataset\" is not allowed?</p>\n<p>The wording of this rule is rather ambiguous, or have I missed something <a href=\"https://www.kaggle.com/wcukierski\" target=\"_blank\">@wcukierski</a>?</p>",
      "rawMarkdown": "Reviving this old topic, given the rules state \"Freely & publicly available external data is allowed, including pre-trained models\", does that mean pre-trained models that I trained locally and uploaded as a \"private dataset\" is not allowed?\n\nThe wording of this rule is rather ambiguous, or have I missed something @wcukierski?",
      "votes": null
    },
    {
      "id": "1652697",
      "postDate": "01/16/2022 22:32:52",
      "content": "<p>It is allowed to train model locally, upload as a dataset and then use it in competition. </p>",
      "rawMarkdown": "It is allowed to train model locally, upload as a dataset and then use it in competition.",
      "votes": null
    },
    {
      "id": "1652716",
      "postDate": "01/16/2022 23:24:20",
      "content": "<p><a href=\"https://www.kaggle.com/remekkinas\" target=\"_blank\">@remekkinas</a> Thanks. Do you know whether it is allowed to upload it as a dataset that is \"private\" ? Given the wording of the competition rules?</p>",
      "rawMarkdown": "remekkinas Thanks. Do you know whether it is allowed to upload it as a dataset that is \"private\" ? Given the wording of the competition rules?",
      "votes": null
    },
    {
      "id": "1652992",
      "postDate": "01/17/2022 06:46:00",
      "content": "<p>You can use models trained by yourself (otherwise, there would be no point in holding competitions).</p>",
      "rawMarkdown": "You can use models trained by yourself (otherwise, there would be no point in holding competitions).",
      "votes": null
    },
    {
      "id": "1653078",
      "postDate": "01/17/2022 08:30:39",
      "content": "<p><a href=\"https://www.kaggle.com/alexchwong\" target=\"_blank\">@alexchwong</a> I think it could be private.</p>",
      "rawMarkdown": "alexchwong I think it could be private.",
      "votes": null
    },
    {
      "id": "1653153",
      "postDate": "01/17/2022 09:32:22",
      "content": "<p>Thanks guys.</p>",
      "rawMarkdown": "Thanks guys.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1625831,
      "author_name": "remekkinas",
      "author_url": "",
      "post_date": "12/22/2021 07:13:00",
      "content": "<p>yes, yes, yes :)</p>",
      "votes": null,
      "replies": [
        {
          "id": 1626571,
          "author_name": "junyun1002",
          "author_url": "",
          "post_date": "12/23/2021 02:22:07",
          "content": "<p>thanks you!</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1625900,
      "author_name": "meowmeowmeowmeowmeow",
      "author_url": "",
      "post_date": "12/22/2021 08:52:07",
      "content": "<p>Sure! Upload your model checkpoint as a dataset, attach it to the kernel. Only inference notebook should be committed</p>",
      "votes": null,
      "replies": [
        {
          "id": 1626570,
          "author_name": "junyun1002",
          "author_url": "",
          "post_date": "12/23/2021 02:21:50",
          "content": "<p>thanks a lot!</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1627009,
      "author_name": "susnato",
      "author_url": "",
      "post_date": "12/23/2021 12:31:22",
      "content": "<p>Yes, you can.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1652694,
      "author_name": "alexchwong",
      "author_url": "",
      "post_date": "01/16/2022 22:31:41",
      "content": "<p>Reviving this old topic, given the rules state \"Freely &amp; publicly available external data is allowed, including pre-trained models\", does that mean pre-trained models that I trained locally and uploaded as a \"private dataset\" is not allowed?</p>\n<p>The wording of this rule is rather ambiguous, or have I missed something <a href=\"https://www.kaggle.com/wcukierski\" target=\"_blank\">@wcukierski</a>?</p>",
      "votes": null,
      "replies": [
        {
          "id": 1652697,
          "author_name": "remekkinas",
          "author_url": "",
          "post_date": "01/16/2022 22:32:52",
          "content": "<p>It is allowed to train model locally, upload as a dataset and then use it in competition. </p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1652716,
          "author_name": "alexchwong",
          "author_url": "",
          "post_date": "01/16/2022 23:24:20",
          "content": "<p><a href=\"https://www.kaggle.com/remekkinas\" target=\"_blank\">@remekkinas</a> Thanks. Do you know whether it is allowed to upload it as a dataset that is \"private\" ? Given the wording of the competition rules?</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1652992,
          "author_name": "atamazian",
          "author_url": "",
          "post_date": "01/17/2022 06:46:00",
          "content": "<p>You can use models trained by yourself (otherwise, there would be no point in holding competitions).</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1653078,
          "author_name": "remekkinas",
          "author_url": "",
          "post_date": "01/17/2022 08:30:39",
          "content": "<p><a href=\"https://www.kaggle.com/alexchwong\" target=\"_blank\">@alexchwong</a> I think it could be private.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1653153,
          "author_name": "alexchwong",
          "author_url": "",
          "post_date": "01/17/2022 09:32:22",
          "content": "<p>Thanks guys.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1625604": "hi , I just came into contact with the kernel only competition recently. I have a few questions to ask: can I train the model on my own server, and then upload the model and submit the results in the submitted kernel interface?",
    "1625831": "yes, yes, yes :)",
    "1625900": "Sure! Upload your model checkpoint as a dataset, attach it to the kernel. Only inference notebook should be committed",
    "1626570": "thanks a lot!",
    "1626571": "thanks you!",
    "1627009": "Yes, you can.",
    "1652694": "Reviving this old topic, given the rules state \"Freely & publicly available external data is allowed, including pre-trained models\", does that mean pre-trained models that I trained locally and uploaded as a \"private dataset\" is not allowed?\n\nThe wording of this rule is rather ambiguous, or have I missed something @wcukierski?",
    "1652697": "It is allowed to train model locally, upload as a dataset and then use it in competition.",
    "1652716": "remekkinas Thanks. Do you know whether it is allowed to upload it as a dataset that is \"private\" ? Given the wording of the competition rules?",
    "1652992": "You can use models trained by yourself (otherwise, there would be no point in holding competitions).",
    "1653078": "alexchwong I think it could be private.",
    "1653153": "Thanks guys."
  },
  "source": "meta"
}