![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F1197382%2Ffde67bdc74aa7482f65b3d8f5315af52%2Flb1.png?generation=1688000726548843&alt=media\")
An interesting fact.
", "messages": [ { "id": 2321881, "postDate": "2023-06-29T01:05:44.550Z", "content": "
An interesting fact.
", "rawMarkdown": "\n\nAn interesting fact.", "votes": 58 }, { "id": 2321911, "postDate": "2023-06-29T01:45:38.797Z", "content": "This competition is full of coincidences…
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2099265%2Fd8769286d927ce24a37172fa05e96358%2Fc2.png?generation=1688003115230046&alt=media\")
![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2099265%2Fb751d2040b0caefb70cb52a55d34ae82%2Fc1.png?generation=1688003105953391&alt=media\")
Life is unfair.
", "rawMarkdown": "Life is unfair.", "votes": 1, "replies": [ { "id": 2341268, "postDate": "2023-07-12T02:09:11.793Z", "content": "Yes, but not this time. All of these teams have been removed from LB.
", "rawMarkdown": "Yes, but not this time. All of these teams have been removed from LB.", "votes": 1 } ] } ] }, { "id": 2321896, "postDate": "2023-06-29T01:19:53.780Z", "content": "The same shift from public to private leaderboard for 4 teams plus a lot of newbie with few submissions (9 submissions for 5 users, for example) achieving high leaderboard positions seems too suspicious to be a coincidence )
", "rawMarkdown": "The same shift from public to private leaderboard for 4 teams plus a lot of newbie with few submissions (9 submissions for 5 users, for example) achieving high leaderboard positions seems too suspicious to be a coincidence )", "votes": 9, "replies": [ { "id": 2324041, "postDate": "2023-06-30T10:44:07.727Z", "content": "Submitting public notebook could achieve this.
", "rawMarkdown": "Submitting public notebook could achieve this. " } ] }, { "id": 2341253, "postDate": "2023-07-12T01:45:49.697Z", "content": "Good news: All teams with silver medal mentioned above have been deleted.
", "rawMarkdown": "Good news: All teams with silver medal mentioned above have been deleted.", "votes": 5 }, { "id": 2322472, "postDate": "2023-06-29T09:55:03.080Z", "content": "![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6674022%2Fe1dd99adffe0335eeead2fddebd781ac%2FScreenshot%20from%202023-06-29%2011-45-45.png?generation=1688032167996173&alt=media\")
This notebook trained locally gives you 0.698 (probably a different version of catboost, but same hyperparams, other hyperparams made it worse for me). It was better than anything else I tried LOL.
\nMaybe others also got their best score by combining public notebooks in the same way. My best score didn't ensemble.
", "rawMarkdown": "\n\nThis [notebook](https://www.kaggle.com/code/vadimkamaev/catboost-new) trained locally gives you 0.698 (probably a different version of catboost, but same hyperparams, other hyperparams made it worse for me). It was better than anything else I tried LOL.\n\nMaybe others also got their best score by combining public notebooks in the same way. My best score didn't ensemble.", "votes": 4 }, { "id": 2321925, "postDate": "2023-06-29T02:07:55.810Z", "content": "Please upvote this thread to highlight it with the host. Not sure yet if they are gonna do anything but let's try our best to make it visible.
", "rawMarkdown": "Please upvote this thread to highlight it with the host. Not sure yet if they are gonna do anything but let's try our best to make it visible.", "votes": 3 }, { "id": 2341232, "postDate": "2023-07-12T00:56:54.953Z", "content": "Now they disappeared from the leaderboard 🚔
", "rawMarkdown": "Now they disappeared from the leaderboard 🚔", "votes": 4, "replies": [ { "id": 2341264, "postDate": "2023-07-12T02:07:51.210Z", "content": "", "rawMarkdown": "These too: https://www.kaggle.com/competitions/predict-student-performance-from-game-play/discussion/420026#2321911", "votes": 2 }, { "id": 2341601, "postDate": "2023-07-12T07:31:44.747Z", "content": "Apparently the top-1 public LB team as well. One of their team members has profile page that is now inaccessible.
", "rawMarkdown": "Apparently the top-1 **public** LB team as well. One of their team members has profile page that is now inaccessible.", "votes": 2 } ] }, { "id": 2322643, "postDate": "2023-06-29T11:59:32.393Z", "content": "All it says is that they probably used the same submission code. It does no say that they had private sharing outside teams. As others pointed out in their comments, they may just use the same public notebook.
\nPublic finger pointing is not great IMHO. If you have doubt then you can share privately with Kaggle and provide the evidence you may have about potential cheaters. I think you can send emails to compliance@kaggle.com for that.
", "rawMarkdown": "All it says is that they probably used the same submission code. It does no say that they had private sharing outside teams. As others pointed out in their comments, they may just use the same public notebook. \n\nPublic finger pointing is not great IMHO. If you have doubt then you can share privately with Kaggle and provide the evidence you may have about potential cheaters. I think you can send emails to compliance@kaggle.com for that.\n", "votes": 3 }, { "id": 2322331, "postDate": "2023-06-29T08:01:57.920Z", "content": "One possible answer would be that they used the same public notebook with the same type of slight adjustment, but I don't have enough experience to say how likely that is, just want to add a voice of caution before going to conclusions.
", "rawMarkdown": "One possible answer would be that they used the same public notebook with the same type of slight adjustment, but I don't have enough experience to say how likely that is, just want to add a voice of caution before going to conclusions.", "votes": 2, "replies": [ { "id": 2323333, "postDate": "2023-06-29T21:29:40.620Z", "content": "Could be. But how would you explain (almost) identical number of submissions and the same time of the last submission on these screenshots?
", "rawMarkdown": "Could be. But how would you explain (almost) identical number of submissions and the same time of the last submission on [these screenshots](https://www.kaggle.com/competitions/predict-student-performance-from-game-play/discussion/420026#2321911)?", "votes": 1 } ] }, { "id": 2321959, "postDate": "2023-06-29T02:41:25.310Z", "content": "Kaggle will remove the cheating results.
", "rawMarkdown": "Kaggle will remove the cheating results.", "votes": 2 }, { "id": 2322596, "postDate": "2023-06-29T11:16:03.863Z", "rawMarkdown": "", "isDeleted": true } ], "comments": [ { "id": 2321911, "author_name": "Oleksiy Kononenko", "author_url": "", "post_date": "2023-06-29T01:45:38.797000", "content": "This competition is full of coincidences…
\n
Life is unfair.
", "votes": 1, "replies": [ { "id": 2341268, "author_name": "Oleksiy Kononenko", "author_url": "", "post_date": "2023-07-12T02:09:11.793000", "content": "Yes, but not this time. All of these teams have been removed from LB.
", "votes": 1, "replies": [] } ] } ] }, { "id": 2321896, "author_name": "Vitaly Kudelya", "author_url": "", "post_date": "2023-06-29T01:19:53.780000", "content": "The same shift from public to private leaderboard for 4 teams plus a lot of newbie with few submissions (9 submissions for 5 users, for example) achieving high leaderboard positions seems too suspicious to be a coincidence )
", "votes": 9, "replies": [ { "id": 2324041, "author_name": "Lucas Morin", "author_url": "", "post_date": "2023-06-30T10:44:07.727000", "content": "Submitting public notebook could achieve this.
", "votes": 0, "replies": [] } ] }, { "id": 2341253, "author_name": "嘴爷", "author_url": "", "post_date": "2023-07-12T01:45:49.697000", "content": "Good news: All teams with silver medal mentioned above have been deleted.
", "votes": 5, "replies": [] }, { "id": 2322472, "author_name": "CroDoc", "author_url": "", "post_date": "2023-06-29T09:55:03.080000", "content": "
This notebook trained locally gives you 0.698 (probably a different version of catboost, but same hyperparams, other hyperparams made it worse for me). It was better than anything else I tried LOL.
\nMaybe others also got their best score by combining public notebooks in the same way. My best score didn't ensemble.
", "votes": 4, "replies": [] }, { "id": 2321925, "author_name": "Hoang Nguyen", "author_url": "", "post_date": "2023-06-29T02:07:55.810000", "content": "Please upvote this thread to highlight it with the host. Not sure yet if they are gonna do anything but let's try our best to make it visible.
", "votes": 3, "replies": [] }, { "id": 2341232, "author_name": "heng", "author_url": "", "post_date": "2023-07-12T00:56:54.953000", "content": "Now they disappeared from the leaderboard 🚔
", "votes": 4, "replies": [ { "id": 2341264, "author_name": "Oleksiy Kononenko", "author_url": "", "post_date": "2023-07-12T02:07:51.210000", "content": "", "votes": 2, "replies": [] }, { "id": 2341601, "author_name": "Hoang Nguyen", "author_url": "", "post_date": "2023-07-12T07:31:44.747000", "content": "Apparently the top-1 public LB team as well. One of their team members has profile page that is now inaccessible.
", "votes": 2, "replies": [] } ] }, { "id": 2322643, "author_name": "CPMP", "author_url": "", "post_date": "2023-06-29T11:59:32.393000", "content": "All it says is that they probably used the same submission code. It does no say that they had private sharing outside teams. As others pointed out in their comments, they may just use the same public notebook.
\nPublic finger pointing is not great IMHO. If you have doubt then you can share privately with Kaggle and provide the evidence you may have about potential cheaters. I think you can send emails to compliance@kaggle.com for that.
", "votes": 3, "replies": [] }, { "id": 2322331, "author_name": "Raki", "author_url": "", "post_date": "2023-06-29T08:01:57.920000", "content": "One possible answer would be that they used the same public notebook with the same type of slight adjustment, but I don't have enough experience to say how likely that is, just want to add a voice of caution before going to conclusions.
", "votes": 2, "replies": [ { "id": 2323333, "author_name": "Oleksiy Kononenko", "author_url": "", "post_date": "2023-06-29T21:29:40.620000", "content": "Could be. But how would you explain (almost) identical number of submissions and the same time of the last submission on these screenshots?
", "votes": 1, "replies": [] } ] }, { "id": 2321959, "author_name": "Alexander Suen", "author_url": "", "post_date": "2023-06-29T02:41:25.310000", "content": "Kaggle will remove the cheating results.
", "votes": 2, "replies": [] }, { "id": 2322596, "author_name": "", "author_url": "", "post_date": "2023-06-29T11:16:03.863000", "content": "", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "2321881": "\n\nAn interesting fact.", "2321911": "This competition is full of coincidences...\n\n\n\n\n\n\n\n\n\n", "2321896": "The same shift from public to private leaderboard for 4 teams plus a lot of newbie with few submissions (9 submissions for 5 users, for example) achieving high leaderboard positions seems too suspicious to be a coincidence )", "2341253": "Good news: All teams with silver medal mentioned above have been deleted.", "2322472": "\n\nThis [notebook](https://www.kaggle.com/code/vadimkamaev/catboost-new) trained locally gives you 0.698 (probably a different version of catboost, but same hyperparams, other hyperparams made it worse for me). It was better than anything else I tried LOL.\n\nMaybe others also got their best score by combining public notebooks in the same way. My best score didn't ensemble.", "2321925": "Please upvote this thread to highlight it with the host. Not sure yet if they are gonna do anything but let's try our best to make it visible.", "2341232": "Now they disappeared from the leaderboard 🚔", "2322643": "All it says is that they probably used the same submission code. It does no say that they had private sharing outside teams. As others pointed out in their comments, they may just use the same public notebook. \n\nPublic finger pointing is not great IMHO. If you have doubt then you can share privately with Kaggle and provide the evidence you may have about potential cheaters. I think you can send emails to compliance@kaggle.com for that.\n", "2322331": "One possible answer would be that they used the same public notebook with the same type of slight adjustment, but I don't have enough experience to say how likely that is, just want to add a voice of caution before going to conclusions.", "2321959": "Kaggle will remove the cheating results.", "2322596": "" } }