So, it seems the notebook and the discussion for the 25.6 LB leak have been deleted. If someone does not make it public again, this will constitute private sharing, and all the people that used it should be disqualified.
\nEDIT: the submission csv is now available at https://www.kaggle.com/code/extraxim/ensemble-csv-for-more-files-lb-25-6
", "messages": [ { "id": 3226449, "postDate": "2025-06-17T17:00:49.457Z", "content": "Why do you call it a leak?
\nUsually, on Kaggle, a leak is some signal from test labels found in training data.
", "rawMarkdown": "Why do you call it a leak?\n\nUsually, on Kaggle, a leak is some signal from test labels found in training data. \n", "votes": 2, "replies": [ { "id": 3226587, "postDate": "2025-06-17T21:00:45.733Z", "content": "I agree. Still, we can discuss semantics, but any information that goes out and is only known by a select number of people is basically a leak in some way. The situation would be different if the thing that improves their score is something they discovered independently from each other. But it doesn't matter now, the csv is now public (although the methodology is still hidden, but it would be naive to think someone is going to share it).
", "rawMarkdown": "I agree. Still, we can discuss semantics, but any information that goes out and is only known by a select number of people is basically a leak in some way. The situation would be different if the thing that improves their score is something they discovered independently from each other. But it doesn't matter now, the csv is now public (although the methodology is still hidden, but it would be naive to think someone is going to share it).", "votes": -1, "replies": [ { "id": 3226602, "postDate": "2025-06-17T21:31:05.227Z", "content": "I agree that sharing a good score notebook for a while then removing it is unfair to all those who didn't see it.
\nThe notebook author did the right thing: make it public again.
", "rawMarkdown": "I agree that sharing a good score notebook for a while then removing it is unfair to all those who didn't see it.\n\nThe notebook author did the right thing: make it public again.", "votes": 2 }, { "id": 3226826, "postDate": "2025-06-18T06:21:39.573Z", "content": "I just want to clarify that the notebook has always been public and was never hidden. Once a notebook is made public, it cannot be hidden again.
", "rawMarkdown": "I just want to clarify that the notebook has always been public and was never hidden. Once a notebook is made public, it cannot be hidden again.", "votes": 1 }, { "id": 3226879, "postDate": "2025-06-18T07:33:56.107Z", "content": "I just checked that I could create a notebook, make it public, then make it private again.
", "rawMarkdown": "I just checked that I could create a notebook, make it public, then make it private again." }, { "id": 3226894, "postDate": "2025-06-18T08:01:07.737Z", "content": "you are correct ….
", "rawMarkdown": "you are correct ....", "votes": 1 }, { "id": 3227001, "postDate": "2025-06-18T10:56:08.160Z", "content": "only can't datasets and models
", "rawMarkdown": "only can't datasets and models", "votes": 1 } ] } ] }, { "id": 3226018, "postDate": "2025-06-17T06:39:58.297Z", "content": "That is a public kernel was shared, the original intent was just to share the code, not the CSV file.
\nHowever, it wasn't taken into account that the CSV file would eventually be displayed in the output.
\nIt's been there ever since, and everyone can copy and use it for themselves.
\nhttps://www.kaggle.com/code/atom1231/ensemble-csv-for-more-files
\nAnd of course, some people have already done so.
", "rawMarkdown": "That is a public kernel was shared, the original intent was just to share the code, not the CSV file.\n\nHowever, it wasn't taken into account that the CSV file would eventually be displayed in the output.\n\nIt's been there ever since, and everyone can copy and use it for themselves.\n\nhttps://www.kaggle.com/code/atom1231/ensemble-csv-for-more-files\n\nAnd of course, some people have already done so.", "votes": 1 }, { "id": 3225966, "postDate": "2025-06-17T05:14:16.967Z", "content": "So, it seems the notebook and the discussion for the 25.6 LB leak have been deleted. If someone does not make it public again, this will constitute private sharing, and all the people that used it should be disqualified.
\nEDIT: the submission csv is now available at https://www.kaggle.com/code/extraxim/ensemble-csv-for-more-files-lb-25-6
", "rawMarkdown": "So, it seems the notebook and the discussion for the 25.6 LB leak have been deleted. If someone does not make it public again, this will constitute private sharing, and all the people that used it should be disqualified. \n\nEDIT: the submission csv is now available at https://www.kaggle.com/code/extraxim/ensemble-csv-for-more-files-lb-25-6", "votes": -7 }, { "id": 3226008, "postDate": "2025-06-17T06:20:43.167Z", "content": "I have no idea what this is all about, but I doubt Kaggle would disqualify somebody because somebody else deleted a notebook or data. Is there precedent for this?
", "rawMarkdown": "I have no idea what this is all about, but I doubt Kaggle would disqualify somebody because somebody else deleted a notebook or data. Is there precedent for this?", "replies": [ { "id": 3226044, "postDate": "2025-06-17T07:24:10.197Z", "content": "Nah I don't believe they would. As long as everything was done in good faith, difficult to call it 'private sharing'.
", "rawMarkdown": "Nah I don't believe they would. As long as everything was done in good faith, difficult to call it 'private sharing'." } ] }, { "id": 3226010, "postDate": "2025-06-17T06:21:03.843Z", "rawMarkdown": "", "votes": -1, "isDeleted": true } ], "comments": [ { "id": 3226449, "author_name": "CPMP", "author_url": "", "post_date": "2025-06-17T17:00:49.457000", "content": "Why do you call it a leak?
\nUsually, on Kaggle, a leak is some signal from test labels found in training data.
", "votes": 2, "replies": [ { "id": 3226587, "author_name": "Federico Peccia", "author_url": "", "post_date": "2025-06-17T21:00:45.733000", "content": "I agree. Still, we can discuss semantics, but any information that goes out and is only known by a select number of people is basically a leak in some way. The situation would be different if the thing that improves their score is something they discovered independently from each other. But it doesn't matter now, the csv is now public (although the methodology is still hidden, but it would be naive to think someone is going to share it).
", "votes": -1, "replies": [ { "id": 3226602, "author_name": "CPMP", "author_url": "", "post_date": "2025-06-17T21:31:05.227000", "content": "I agree that sharing a good score notebook for a while then removing it is unfair to all those who didn't see it.
\nThe notebook author did the right thing: make it public again.
", "votes": 2, "replies": [] }, { "id": 3226826, "author_name": "atom1231", "author_url": "", "post_date": "2025-06-18T06:21:39.573000", "content": "I just want to clarify that the notebook has always been public and was never hidden. Once a notebook is made public, it cannot be hidden again.
", "votes": 1, "replies": [] }, { "id": 3226879, "author_name": "CPMP", "author_url": "", "post_date": "2025-06-18T07:33:56.107000", "content": "I just checked that I could create a notebook, make it public, then make it private again.
", "votes": 0, "replies": [] }, { "id": 3226894, "author_name": "atom1231", "author_url": "", "post_date": "2025-06-18T08:01:07.737000", "content": "you are correct ….
", "votes": 1, "replies": [] }, { "id": 3227001, "author_name": "Kurise", "author_url": "", "post_date": "2025-06-18T10:56:08.160000", "content": "only can't datasets and models
", "votes": 1, "replies": [] } ] } ] }, { "id": 3226018, "author_name": "atom1231", "author_url": "", "post_date": "2025-06-17T06:39:58.297000", "content": "That is a public kernel was shared, the original intent was just to share the code, not the CSV file.
\nHowever, it wasn't taken into account that the CSV file would eventually be displayed in the output.
\nIt's been there ever since, and everyone can copy and use it for themselves.
\nhttps://www.kaggle.com/code/atom1231/ensemble-csv-for-more-files
\nAnd of course, some people have already done so.
", "votes": 1, "replies": [] }, { "id": 3226008, "author_name": "Jeroen Cottaar", "author_url": "", "post_date": "2025-06-17T06:20:43.167000", "content": "I have no idea what this is all about, but I doubt Kaggle would disqualify somebody because somebody else deleted a notebook or data. Is there precedent for this?
", "votes": 0, "replies": [ { "id": 3226044, "author_name": "greySnow", "author_url": "", "post_date": "2025-06-17T07:24:10.197000", "content": "Nah I don't believe they would. As long as everything was done in good faith, difficult to call it 'private sharing'.
", "votes": 0, "replies": [] } ] }, { "id": 3226010, "author_name": "", "author_url": "", "post_date": "2025-06-17T06:21:03.843000", "content": "", "votes": -1, "replies": [] } ], "raw_markdown_by_id": { "3226449": "Why do you call it a leak?\n\nUsually, on Kaggle, a leak is some signal from test labels found in training data. \n", "3226018": "That is a public kernel was shared, the original intent was just to share the code, not the CSV file.\n\nHowever, it wasn't taken into account that the CSV file would eventually be displayed in the output.\n\nIt's been there ever since, and everyone can copy and use it for themselves.\n\nhttps://www.kaggle.com/code/atom1231/ensemble-csv-for-more-files\n\nAnd of course, some people have already done so.", "3225966": "So, it seems the notebook and the discussion for the 25.6 LB leak have been deleted. If someone does not make it public again, this will constitute private sharing, and all the people that used it should be disqualified. \n\nEDIT: the submission csv is now available at https://www.kaggle.com/code/extraxim/ensemble-csv-for-more-files-lb-25-6", "3226008": "I have no idea what this is all about, but I doubt Kaggle would disqualify somebody because somebody else deleted a notebook or data. Is there precedent for this?", "3226010": "" } }