{ "id": 253214, "title": "Question about external data", "url": "/competitions/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/253214", "author_name": "", "post_date": "2021-07-15T12:42:14.714677100Z", "votes": 2, "comment_count": 1, "views": 0, "content": "

Hi, @juliaelliott.
\nThank you very much for organizing such an interesting competition.

\n

As the title states, I have one question about external data.
\nIn the competition rules tab, it says that external data can be used as long as it is public and can be accessed by anyone, as shown in the quote below.

\n
\n

C. External Data.
\n You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).

\n
\n

However, on the other hand, Baid et al. (“The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classification”, arXiv:2107.02314, 2021) states that it is not allowed to use data sets other than the BraTS data, as shown below.

\n
\n

2.1 Data
\n Participants are not allowed to use additional public and/or private data (from their own institutions) for extending the provided BraTS data, for the training of the algorithm chosen to be ranked.

\n
\n

Which one should I consider correct?
\nAlso, if the content of the above paper published by the competition organizers is correct, is it allowed to use pre-trained Neural Network weights?

\n

I would be grateful for your answer.

\n
\n

Thank you, @lechuck0 !
\nWell clarified 😊

", "messages": [ { "id": "1389073", "postDate": "07/15/2021 12:42:14", "content": "

Hi, @juliaelliott.
\nThank you very much for organizing such an interesting competition.

\n

As the title states, I have one question about external data.
\nIn the competition rules tab, it says that external data can be used as long as it is public and can be accessed by anyone, as shown in the quote below.

\n
\n

C. External Data.
\n You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).

\n
\n

However, on the other hand, Baid et al. (“The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classification”, arXiv:2107.02314, 2021) states that it is not allowed to use data sets other than the BraTS data, as shown below.

\n
\n

2.1 Data
\n Participants are not allowed to use additional public and/or private data (from their own institutions) for extending the provided BraTS data, for the training of the algorithm chosen to be ranked.

\n
\n

Which one should I consider correct?
\nAlso, if the content of the above paper published by the competition organizers is correct, is it allowed to use pre-trained Neural Network weights?

\n

I would be grateful for your answer.

\n
\n

Thank you, @lechuck0 !
\nWell clarified 😊

", "rawMarkdown": "Hi, @juliaelliott. \nThank you very much for organizing such an interesting competition.\n\nAs the title states, I have one question about external data.\nIn the competition rules tab, it says that external data can be used as long as it is public and can be accessed by anyone, as shown in the quote below.\n\n> C. External Data. \nYou may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).\n\nHowever, on the other hand, [Baid et al.](https://arxiv.org/abs/2107.02314) (“The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classification”, arXiv:2107.02314, 2021) states that it is not allowed to use data sets other than the BraTS data, as shown below.\n\n> 2.1 Data\nParticipants are not allowed to use additional public and/or private data (from their own institutions) for extending the provided BraTS data, for the training of the algorithm chosen to be ranked. \n\n**Which one should I consider correct?** \n**Also, if the content of the above paper published by the competition organizers is correct, is it allowed to use pre-trained Neural Network weights?**\n\nI would be grateful for your answer.\n\n\n---\n\nThank you, @lechuck0 !\nWell clarified 😊", "votes": null }, { "id": "1389587", "postDate": "07/15/2021 20:57:35", "content": "

Thank you for pointing this out @maxwell110!

\n

Please use Kaggle's competition rules: \"You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).\"

\n

We plan to correct the arXiv publication to be in alignment with Kaggle's competition rules.

", "rawMarkdown": "Thank you for pointing this out @maxwell110!\n\nPlease use Kaggle's competition rules: \"You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).\"\n\nWe plan to correct the arXiv publication to be in alignment with Kaggle's competition rules.", "votes": null } ], "comments": [ { "id": 1389587, "author_name": "lechuck0", "author_url": "", "post_date": "07/15/2021 20:57:35", "content": "

Thank you for pointing this out @maxwell110!

\n

Please use Kaggle's competition rules: \"You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).\"

\n

We plan to correct the arXiv publication to be in alignment with Kaggle's competition rules.

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1389073": "Hi, @juliaelliott. \nThank you very much for organizing such an interesting competition.\n\nAs the title states, I have one question about external data.\nIn the competition rules tab, it says that external data can be used as long as it is public and can be accessed by anyone, as shown in the quote below.\n\n> C. External Data. \nYou may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).\n\nHowever, on the other hand, [Baid et al.](https://arxiv.org/abs/2107.02314) (“The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classification”, arXiv:2107.02314, 2021) states that it is not allowed to use data sets other than the BraTS data, as shown below.\n\n> 2.1 Data\nParticipants are not allowed to use additional public and/or private data (from their own institutions) for extending the provided BraTS data, for the training of the algorithm chosen to be ranked. \n\n**Which one should I consider correct?** \n**Also, if the content of the above paper published by the competition organizers is correct, is it allowed to use pre-trained Neural Network weights?**\n\nI would be grateful for your answer.\n\n\n---\n\nThank you, @lechuck0 !\nWell clarified 😊", "1389587": "Thank you for pointing this out @maxwell110!\n\nPlease use Kaggle's competition rules: \"You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).\"\n\nWe plan to correct the arXiv publication to be in alignment with Kaggle's competition rules." }, "source": "meta" }