{ "id": 203342, "title": "[SEE EDIT] Preprocessed datasets/notebook outputs are prohibited in this competition!", "url": "/competitions/ranzcr-clip-catheter-line-classification/discussion/203342", "author_name": "", "post_date": "2020-12-14T21:20:21.866051100Z", "votes": 6, "comment_count": 7, "views": 0, "content": "

EDIT: It is confirmed that it's possible to create pre-processing dataset with certain conditions. See this comment by @maggiemd.

\n
\n

From the competition rules:

\n
\n

A. Data Access and Use. You may access and use the Competition Data for non-commercial purposes only, including for participating in the Competition and on Kaggle.com forums, and for academic research and education. The Competition Sponsor reserves the right to disqualify any participant who uses the Competition Data other than as permitted by the Competition Website and these Rules.

\n

Imaging studies used in this Challenge are de-identified patient data, meaning that reasonable care has been taken to remove from them all personally identifiable information. As a participant in the Competition, you agree 1) not to make copies of, or in any way redistribute, any of the data made available to you, 2) not to attempt to re-identify any personal information based on the data and 3) to notify the Competition organizers of any personally identifiable information you encounter in the data by posting a message to the Kaggle community discussion forums.

\n

B. Data Security. You agree to use reasonable and suitable measures to prevent persons who have not formally agreed to these Rules from gaining access to the Competition Data. You agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Competition Data to any party not participating in the Competition. You agree to notify Kaggle immediately upon learning of any possible unauthorized transmission of or unauthorized access to the Competition Data and agree to work with Kaggle to rectify any unauthorized transmission or access.

\n
\n

Please see edit

", "messages": [ { "id": "1112744", "postDate": "12/14/2020 21:20:21", "content": "

EDIT: It is confirmed that it's possible to create pre-processing dataset with certain conditions. See this comment by @maggiemd.

\n
\n

From the competition rules:

\n
\n

A. Data Access and Use. You may access and use the Competition Data for non-commercial purposes only, including for participating in the Competition and on Kaggle.com forums, and for academic research and education. The Competition Sponsor reserves the right to disqualify any participant who uses the Competition Data other than as permitted by the Competition Website and these Rules.

\n

Imaging studies used in this Challenge are de-identified patient data, meaning that reasonable care has been taken to remove from them all personally identifiable information. As a participant in the Competition, you agree 1) not to make copies of, or in any way redistribute, any of the data made available to you, 2) not to attempt to re-identify any personal information based on the data and 3) to notify the Competition organizers of any personally identifiable information you encounter in the data by posting a message to the Kaggle community discussion forums.

\n

B. Data Security. You agree to use reasonable and suitable measures to prevent persons who have not formally agreed to these Rules from gaining access to the Competition Data. You agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Competition Data to any party not participating in the Competition. You agree to notify Kaggle immediately upon learning of any possible unauthorized transmission of or unauthorized access to the Competition Data and agree to work with Kaggle to rectify any unauthorized transmission or access.

\n
\n

Please see edit

", "rawMarkdown": "**EDIT**: It is confirmed that it's possible to create pre-processing dataset with certain conditions. See [this comment](https://www.kaggle.com/c/ranzcr-clip-catheter-line-classification/discussion/203342#1113111) by @maggiemd.\n\n---\n\nFrom the competition rules:\n\n> A. Data Access and Use. You may access and use the Competition Data for non-commercial purposes only, including for participating in the Competition and on Kaggle.com forums, and for academic research and education. The Competition Sponsor reserves the right to disqualify any participant who uses the Competition Data other than as permitted by the Competition Website and these Rules.\n\n> Imaging studies used in this Challenge are de-identified patient data, meaning that reasonable care has been taken to remove from them all personally identifiable information. As a participant in the Competition, you agree 1) not to make copies of, or in any way redistribute, any of the data made available to you, 2) not to attempt to re-identify any personal information based on the data and 3) to notify the Competition organizers of any personally identifiable information you encounter in the data by posting a message to the Kaggle community discussion forums.\n\n> B. Data Security. You agree to use reasonable and suitable measures to prevent persons who have not formally agreed to these Rules from gaining access to the Competition Data. **You agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Competition Data to any party not participating in the Competition**. You agree to notify Kaggle immediately upon learning of any possible unauthorized transmission of or unauthorized access to the Competition Data and agree to work with Kaggle to rectify any unauthorized transmission or access.\n\n~~If I understand correctly, this means you can't create public dataset or notebooks for pre-processing purposes.~~ *Please see edit*", "votes": null }, { "id": "1112745", "postDate": "12/14/2020 21:20:40", "content": "

@kaggle and organizers please correct if I'm misunderstanding the terms of the competition.

", "rawMarkdown": "kaggle and organizers please correct if I'm misunderstanding the terms of the competition.", "votes": null }, { "id": "1112789", "postDate": "12/14/2020 22:23:42", "content": "

There are multiple completed and continuing competitions where users created resized/preprocessed datasets and loaded them on Kaggle. I haven't ever seen Kaggle ask anybody not to do that. In fact, within Kaggle, they seem to encourage it.

\n

I would be more cautious (and get some approval) if loading outside of Kaggle.

\n

Not an official position, just my two cents.

\n

-Rich

", "rawMarkdown": "There are multiple completed and continuing competitions where users created resized/preprocessed datasets and loaded them on Kaggle. I haven't ever seen Kaggle ask anybody not to do that. In fact, within Kaggle, they seem to encourage it.\n\nI would be more cautious (and get some approval) if loading outside of Kaggle.\n\nNot an official position, just my two cents.\n\n-Rich", "votes": null }, { "id": "1112798", "postDate": "12/14/2020 22:37:40", "content": "

I'm raising this because anyone can download a notebook output without even logging in Kaggle. Also public datasets without a properly specified license can be accessed by anyone that has a Kaggle account without accepting the terms and conditions of the competition.

\n

I think Kaggle should have a semi-public (or competition-bound) dataset option for cases where pre-processing is crucial but the conditions limit the data usage to specific points.

", "rawMarkdown": "I'm raising this because anyone can download a notebook output without even logging in Kaggle. Also public datasets without a properly specified license can be accessed by anyone that has a Kaggle account without accepting the terms and conditions of the competition.\n\nI think Kaggle should have a semi-public (or competition-bound) dataset option for cases where pre-processing is crucial but the conditions limit the data usage to specific points.", "votes": null }, { "id": "1113105", "postDate": "12/15/2020 07:20:50", "content": "

@richardepstein I agree with your interpretation. As long as you are not wholesale sharing the data external to the competition, you can tie your work on the competition with the ties to the original competition linked to your shared dataset.

", "rawMarkdown": "richardepstein I agree with your interpretation. As long as you are not wholesale sharing the data external to the competition, you can tie your work on the competition with the ties to the original competition linked to your shared dataset.", "votes": null }, { "id": "1113111", "postDate": "12/15/2020 07:23:27", "content": "

The spirit is that you will not directly take the data and share it externally. If you are using the data and updating it and tying it to your work in this competition then this is okay. This does not translate to all competitions.

", "rawMarkdown": "The spirit is that you will not directly take the data and share it externally. If you are using the data and updating it and tying it to your work in this competition then this is okay. This does not translate to all competitions.", "votes": null }, { "id": "1113577", "postDate": "12/15/2020 15:01:47", "content": "

Thank you for the clarification, this is reassuring.

", "rawMarkdown": "Thank you for the clarification, this is reassuring.", "votes": null }, { "id": "1113738", "postDate": "12/15/2020 16:46:23", "content": "

I updated the post to reflect this update!

", "rawMarkdown": "I updated the post to reflect this update!", "votes": null } ], "comments": [ { "id": 1112745, "author_name": "xhlulu", "author_url": "", "post_date": "12/14/2020 21:20:40", "content": "

@kaggle and organizers please correct if I'm misunderstanding the terms of the competition.

", "votes": null, "replies": [ { "id": 1113111, "author_name": "maggiemd", "author_url": "", "post_date": "12/15/2020 07:23:27", "content": "

The spirit is that you will not directly take the data and share it externally. If you are using the data and updating it and tying it to your work in this competition then this is okay. This does not translate to all competitions.

", "votes": null, "replies": [] }, { "id": 1113577, "author_name": "xhlulu", "author_url": "", "post_date": "12/15/2020 15:01:47", "content": "

Thank you for the clarification, this is reassuring.

", "votes": null, "replies": [] }, { "id": 1113738, "author_name": "xhlulu", "author_url": "", "post_date": "12/15/2020 16:46:23", "content": "

I updated the post to reflect this update!

", "votes": null, "replies": [] } ] }, { "id": 1112789, "author_name": "richardepstein", "author_url": "", "post_date": "12/14/2020 22:23:42", "content": "

There are multiple completed and continuing competitions where users created resized/preprocessed datasets and loaded them on Kaggle. I haven't ever seen Kaggle ask anybody not to do that. In fact, within Kaggle, they seem to encourage it.

\n

I would be more cautious (and get some approval) if loading outside of Kaggle.

\n

Not an official position, just my two cents.

\n

-Rich

", "votes": null, "replies": [ { "id": 1112798, "author_name": "xhlulu", "author_url": "", "post_date": "12/14/2020 22:37:40", "content": "

I'm raising this because anyone can download a notebook output without even logging in Kaggle. Also public datasets without a properly specified license can be accessed by anyone that has a Kaggle account without accepting the terms and conditions of the competition.

\n

I think Kaggle should have a semi-public (or competition-bound) dataset option for cases where pre-processing is crucial but the conditions limit the data usage to specific points.

", "votes": null, "replies": [] }, { "id": 1113105, "author_name": "maggiemd", "author_url": "", "post_date": "12/15/2020 07:20:50", "content": "

@richardepstein I agree with your interpretation. As long as you are not wholesale sharing the data external to the competition, you can tie your work on the competition with the ties to the original competition linked to your shared dataset.

", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "1112744": "**EDIT**: It is confirmed that it's possible to create pre-processing dataset with certain conditions. See [this comment](https://www.kaggle.com/c/ranzcr-clip-catheter-line-classification/discussion/203342#1113111) by @maggiemd.\n\n---\n\nFrom the competition rules:\n\n> A. Data Access and Use. You may access and use the Competition Data for non-commercial purposes only, including for participating in the Competition and on Kaggle.com forums, and for academic research and education. The Competition Sponsor reserves the right to disqualify any participant who uses the Competition Data other than as permitted by the Competition Website and these Rules.\n\n> Imaging studies used in this Challenge are de-identified patient data, meaning that reasonable care has been taken to remove from them all personally identifiable information. As a participant in the Competition, you agree 1) not to make copies of, or in any way redistribute, any of the data made available to you, 2) not to attempt to re-identify any personal information based on the data and 3) to notify the Competition organizers of any personally identifiable information you encounter in the data by posting a message to the Kaggle community discussion forums.\n\n> B. Data Security. You agree to use reasonable and suitable measures to prevent persons who have not formally agreed to these Rules from gaining access to the Competition Data. **You agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Competition Data to any party not participating in the Competition**. You agree to notify Kaggle immediately upon learning of any possible unauthorized transmission of or unauthorized access to the Competition Data and agree to work with Kaggle to rectify any unauthorized transmission or access.\n\n~~If I understand correctly, this means you can't create public dataset or notebooks for pre-processing purposes.~~ *Please see edit*", "1112745": "kaggle and organizers please correct if I'm misunderstanding the terms of the competition.", "1112789": "There are multiple completed and continuing competitions where users created resized/preprocessed datasets and loaded them on Kaggle. I haven't ever seen Kaggle ask anybody not to do that. In fact, within Kaggle, they seem to encourage it.\n\nI would be more cautious (and get some approval) if loading outside of Kaggle.\n\nNot an official position, just my two cents.\n\n-Rich", "1112798": "I'm raising this because anyone can download a notebook output without even logging in Kaggle. Also public datasets without a properly specified license can be accessed by anyone that has a Kaggle account without accepting the terms and conditions of the competition.\n\nI think Kaggle should have a semi-public (or competition-bound) dataset option for cases where pre-processing is crucial but the conditions limit the data usage to specific points.", "1113105": "richardepstein I agree with your interpretation. As long as you are not wholesale sharing the data external to the competition, you can tie your work on the competition with the ties to the original competition linked to your shared dataset.", "1113111": "The spirit is that you will not directly take the data and share it externally. If you are using the data and updating it and tying it to your work in this competition then this is okay. This does not translate to all competitions.", "1113577": "Thank you for the clarification, this is reassuring.", "1113738": "I updated the post to reflect this update!" }, "source": "meta" }