{
  "id": 669784,
  "title": "Final Check: Rules on Local Training and Kaggle Notebook Submissions",
  "url": "/competitions/csiro-biomass/discussion/669784",
  "author_name": "",
  "post_date": "2026-01-24T10:20:49.798196Z",
  "votes": -1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>I would like to ask for clarification regarding the interpretation of the competition rules,\nespecially as we approach the final day of the competition.</p>\n<p>Currently, I have downloaded the entire official dataset of this competition and am training my models locally.\nAfter training, I upload the trained model weights to Kaggle and perform inference and submission\nusing a Kaggle Notebook.</p>\n<p>I would like to confirm the following point:</p>\n<ul>\n<li>As I understand it, the competition rules require that only publicly available and generally free-to-use\ndata and model architectures are used. Is there any requirement for participants to formally prove this?\nIn other words, since the organizers cannot directly verify the full training procedure performed locally,\nis there any risk that this could be considered a rule violation?</li>\n</ul>\n<p>I would appreciate any guidance from the organizers or the community. Thank you!</p>",
  "messages": [
    {
      "id": "3396101",
      "postDate": "01/24/2026 10:20:49",
      "content": "<p>I would like to ask for clarification regarding the interpretation of the competition rules,\nespecially as we approach the final day of the competition.</p>\n<p>Currently, I have downloaded the entire official dataset of this competition and am training my models locally.\nAfter training, I upload the trained model weights to Kaggle and perform inference and submission\nusing a Kaggle Notebook.</p>\n<p>I would like to confirm the following point:</p>\n<ul>\n<li>As I understand it, the competition rules require that only publicly available and generally free-to-use\ndata and model architectures are used. Is there any requirement for participants to formally prove this?\nIn other words, since the organizers cannot directly verify the full training procedure performed locally,\nis there any risk that this could be considered a rule violation?</li>\n</ul>\n<p>I would appreciate any guidance from the organizers or the community. Thank you!</p>",
      "rawMarkdown": "I would like to ask for clarification regarding the interpretation of the competition rules,\nespecially as we approach the final day of the competition.\n\nCurrently, I have downloaded the entire official dataset of this competition and am training my models locally.\nAfter training, I upload the trained model weights to Kaggle and perform inference and submission\nusing a Kaggle Notebook.\n\nI would like to confirm the following point:\n\n-  As I understand it, the competition rules require that only publicly available and generally free-to-use\n   data and model architectures are used. Is there any requirement for participants to formally prove this?\n   In other words, since the organizers cannot directly verify the full training procedure performed locally,\n   is there any risk that this could be considered a rule violation?\n\nI would appreciate any guidance from the organizers or the community. Thank you!",
      "votes": null
    },
    {
      "id": "3396234",
      "postDate": "01/24/2026 15:12:40",
      "content": "<p>Yes - there is a requirement.  Here's a piece of it.  If you win the money you will need to verify.  Don't win the money - no worries.  The text comes right out of the Rules tab.</p>\n<p>*You may be required by the Sponsor to provide a detailed description of how the winning Submission was generated, to the Competition Sponsor’s specifications, as outlined in Section 2.8, Winner’s Obligations. This may include a detailed description of methodology, where one must be able to reproduce the approach by reading the description, and includes a detailed explanation of the architecture, preprocessing, loss function, training details, hyper-parameters, etc. The description should also include a link to a code repository with complete and detailed instructions so that the results obtained can be reproduced.\n*</p>",
      "rawMarkdown": "Yes - there is a requirement.  Here's a piece of it.  If you win the money you will need to verify.  Don't win the money - no worries.  The text comes right out of the Rules tab.\n\n*You may be required by the Sponsor to provide a detailed description of how the winning Submission was generated, to the Competition Sponsor’s specifications, as outlined in Section 2.8, Winner’s Obligations. This may include a detailed description of methodology, where one must be able to reproduce the approach by reading the description, and includes a detailed explanation of the architecture, preprocessing, loss function, training details, hyper-parameters, etc. The description should also include a link to a code repository with complete and detailed instructions so that the results obtained can be reproduced.\n*",
      "votes": null
    },
    {
      "id": "3396336",
      "postDate": "01/24/2026 18:36:00",
      "content": "<p>I made a jest the other day that someone did not need to worry about getting cash payment in his country since he was new and at 1000th place.  But I got a warning on my account and the post was deleted, so not going to repeat that jest.</p>",
      "rawMarkdown": "I made a jest the other day that someone did not need to worry about getting cash payment in his country since he was new and at 1000th place.  But I got a warning on my account and the post was deleted, so not going to repeat that jest.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3396234,
      "author_name": "pcjimmmy",
      "author_url": "",
      "post_date": "01/24/2026 15:12:40",
      "content": "<p>Yes - there is a requirement.  Here's a piece of it.  If you win the money you will need to verify.  Don't win the money - no worries.  The text comes right out of the Rules tab.</p>\n<p>*You may be required by the Sponsor to provide a detailed description of how the winning Submission was generated, to the Competition Sponsor’s specifications, as outlined in Section 2.8, Winner’s Obligations. This may include a detailed description of methodology, where one must be able to reproduce the approach by reading the description, and includes a detailed explanation of the architecture, preprocessing, loss function, training details, hyper-parameters, etc. The description should also include a link to a code repository with complete and detailed instructions so that the results obtained can be reproduced.\n*</p>",
      "votes": null,
      "replies": [
        {
          "id": 3396336,
          "author_name": "pcjimmmy",
          "author_url": "",
          "post_date": "01/24/2026 18:36:00",
          "content": "<p>I made a jest the other day that someone did not need to worry about getting cash payment in his country since he was new and at 1000th place.  But I got a warning on my account and the post was deleted, so not going to repeat that jest.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3396101": "I would like to ask for clarification regarding the interpretation of the competition rules,\nespecially as we approach the final day of the competition.\n\nCurrently, I have downloaded the entire official dataset of this competition and am training my models locally.\nAfter training, I upload the trained model weights to Kaggle and perform inference and submission\nusing a Kaggle Notebook.\n\nI would like to confirm the following point:\n\n-  As I understand it, the competition rules require that only publicly available and generally free-to-use\n   data and model architectures are used. Is there any requirement for participants to formally prove this?\n   In other words, since the organizers cannot directly verify the full training procedure performed locally,\n   is there any risk that this could be considered a rule violation?\n\nI would appreciate any guidance from the organizers or the community. Thank you!",
    "3396234": "Yes - there is a requirement.  Here's a piece of it.  If you win the money you will need to verify.  Don't win the money - no worries.  The text comes right out of the Rules tab.\n\n*You may be required by the Sponsor to provide a detailed description of how the winning Submission was generated, to the Competition Sponsor’s specifications, as outlined in Section 2.8, Winner’s Obligations. This may include a detailed description of methodology, where one must be able to reproduce the approach by reading the description, and includes a detailed explanation of the architecture, preprocessing, loss function, training details, hyper-parameters, etc. The description should also include a link to a code repository with complete and detailed instructions so that the results obtained can be reproduced.\n*",
    "3396336": "I made a jest the other day that someone did not need to worry about getting cash payment in his country since he was new and at 1000th place.  But I got a warning on my account and the post was deleted, so not going to repeat that jest."
  },
  "source": "meta"
}