UPDATED: June 5th at 11:59 PM UTC is the Final Submission Deadline. After this date, we will not be taking any more submissions. Remember to select your two best submissions to be rescored during the re-run period.
\n\nWe will update the end date of the competition tomorrow to reflect the re-run period end date of June 11, 2019
", "messages": [ { "id": 550712, "postDate": "2019-06-12T00:42:23.890Z", "content": "Hi @maggiemd you’ve released the medals, and with out any explanations for the public Top 5. I’m very disappointed and cannot understand why. Previously when I got a medal I was happy, no matter a silver or a bronze. But this time, what I expected is the explanations and got nothing : )
", "rawMarkdown": "Hi @maggiemd you’ve released the medals, and with out any explanations for the public Top 5. I’m very disappointed and cannot understand why. Previously when I got a medal I was happy, no matter a silver or a bronze. But this time, what I expected is the explanations and got nothing : )", "votes": 9, "replies": [ { "id": 550715, "postDate": "2019-06-12T00:49:29.813Z", "content": "I think at least Kaggle should have a conclusion for this incident, whether they are cheating. If there is no violation, that would be great since everyone follows the rule
", "rawMarkdown": "I think at least Kaggle should have a conclusion for this incident, whether they are cheating. If there is no violation, that would be great since everyone follows the rule", "votes": 1 }, { "id": 550719, "postDate": "2019-06-12T00:52:14.597Z", "content": "", "rawMarkdown": "@wcukierski", "votes": 1 }, { "id": 550723, "postDate": "2019-06-12T00:59:41.520Z", "content": "obviously we have no clue to say someone has cheated 100%. but as this is the largest and the most famous data science competition platform, we all know that it is popularized by its high quality competitions and strict rules. But really?
\n\nthere are sufficient amount of discussions and even someone who jumped out to reveal the possibly inappropriate behaviors done by a small group of people. Similar situation has occured several times and what we got from the kaggle admin team during the past two weeks is \"nothing\".
\n\nChecking their code and kernel and even uploaded model are not that hard for Kaggle. Can we understand this is a unofficial announcement for the tolerance to possibly inappropriate behaviors?
", "rawMarkdown": "obviously we have no clue to say someone has cheated 100%. but as this is the largest and the most famous data science competition platform, we all know that it is popularized by its high quality competitions and strict rules. But really?\n\nthere are sufficient amount of discussions and even someone who jumped out to reveal the possibly inappropriate behaviors done by a small group of people. Similar situation has occured several times and what we got from the kaggle admin team during the past two weeks is \"nothing\".\n\nChecking their code and kernel and even uploaded model are not that hard for Kaggle. Can we understand this is a unofficial announcement for the tolerance to possibly inappropriate behaviors?", "votes": 1 }, { "id": 550732, "postDate": "2019-06-12T01:22:53.933Z", "content": "I don't know why kaggle choose to ignore this :(
", "rawMarkdown": "I don't know why kaggle choose to ignore this :( " }, { "id": 550768, "postDate": "2019-06-12T02:27:19.883Z", "content": "I disappointed. I don't know whether cheating is used in final results. But we hope explains from kaggle and top teams.
", "rawMarkdown": "I disappointed. I don't know whether cheating is used in final results. But we hope explains from kaggle and top teams.", "votes": 2 }, { "id": 550794, "postDate": "2019-06-12T03:21:16.313Z", "content": "\n\nSome of top teams have explained: \"we overfitted leaderboard\", and \"oh, kaggle cannot check my code since I am out of medal zone\".
\n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/95330#latest-550702
", "rawMarkdown": "@takuok \n\nSome of top teams have explained: \"we overfitted leaderboard\", and \"oh, kaggle cannot check my code since I am out of medal zone\". \n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/95330#latest-550702", "votes": 2 }, { "id": 550802, "postDate": "2019-06-12T03:30:06.360Z", "content": "Now, cheating looks best solution;) hahaha.
", "rawMarkdown": "Now, cheating looks best solution;) hahaha.", "votes": -4 }, { "id": 550846, "postDate": "2019-06-12T04:44:36.733Z", "content": "So private sharing and scraping are officially allowed? @inversion
", "rawMarkdown": "So private sharing and scraping are officially allowed? @inversion ", "votes": 1 }, { "id": 550865, "postDate": "2019-06-12T05:16:58.710Z", "content": "Sure, kaggle allow this! I will cheat in the next competition. ;)
", "rawMarkdown": "Sure, kaggle allow this! I will cheat in the next competition. ;)", "votes": -1 }, { "id": 550870, "postDate": "2019-06-12T05:27:25.200Z", "content": "@qiaojian Join X5! Join the GM/Ms! Oh, they just stole the solo gold from you. Don‘t worry, you will have solo gooooolds！
", "rawMarkdown": "@qiaojian Join X5! Join the GM/Ms! Oh, they just stole the solo gold from you. Don‘t worry, you will have solo gooooolds！", "votes": 2 }, { "id": 550871, "postDate": "2019-06-12T05:27:27.277Z", "content": "X5 is allowed to cheat in kaggle, and I have joined ods.ai. I think I can cheat in the next competition. :) By the way, X5 will set up an interview to me!
", "rawMarkdown": "X5 is allowed to cheat in kaggle, and I have joined ods.ai. I think I can cheat in the next competition. :) By the way, X5 will set up an interview to me!" }, { "id": 550897, "postDate": "2019-06-12T06:03:11.290Z", "content": "@qiaojian contact us https://www.x5.ru/en/Pages/About/Contacts.aspx
", "rawMarkdown": "@qiaojian contact us https://www.x5.ru/en/Pages/About/Contacts.aspx", "votes": 3 }, { "id": 550905, "postDate": "2019-06-12T06:08:00.250Z", "content": "@onodera， it‘s time for me to learn crawler!
", "rawMarkdown": "@onodera， it‘s time for me to learn crawler!", "votes": 3 }, { "id": 551305, "postDate": "2019-06-12T15:15:13.457Z", "content": "Hi all, a couple of reminders and remarks on this thread:
\n\nIn general, Kaggle does not make public comments about users, compliance investigations, or other private matters. We do not engage in public speculation. The absence of a removal does not mean a rule wasn't broken (we might simply lack the \"evidence beyond a reasonable doubt\"). A removal does not mean a person had malicious intent (accidents happen, English is a 2nd language for many, and most people simply don't read the rules).
Competition closes take time. We are not \"choosing to ignore\" a situation because two days have passed.
In all competitions, we finalize the leaderboard (the action that releases medals) before the full winner obligations are fulfilled. This is done to prevent everyone from having to wait for weeks while we close the competition. We have and can remove teams after medals have gone out. We have removed teams from leaderboards years after the competition closed due to discovery of new evidence.
\"Can we understand this is a[n] unofficial announcement for the tolerance to possibly inappropriate behaviors?\" No. The rules, site terms, your morals, and our collective community guidelines should govern your behavior here. You and only you can choose to conduct yourself with professionalism and integrity, or you can choose cut corners, pass blame, and justify inappropriate behaviors because others might be cheating the system.
Thanks for your reply, which is constructive and reasonable.\nThere are that many discussions because we believe it is very easy to test those abnormally high score kernels, whether or not there are external data involved or just an overfitted model as the people claimed.
\n\nThis is also a shaded area: what if there are some external data used in the kernels that are not selected for final submission? To some extent, the \"violation\" may not directly affect the final, selected kernel. The problem is, by submitting the weights for final inference purpose, it is relative hard to detect whether the external data are used for training that (those) model(s). But consider the GPU resources Kaggle have, everything should become easy. Just use their code and retrain 1 group. See how many downvotes for the \"public private 2nd place\" post.
\n\nBUT, this time, everything seems to be too coincidence (there are similar situation before but not that kind of \"coincidence\"). As a part of the community has drawn the concerns and also (I believe) it is not hard to validate what we talked about, we hope Kaggle can carefully review this case.
", "rawMarkdown": "Thanks for your reply, which is constructive and reasonable.\nThere are that many discussions because we believe it is very easy to test those abnormally high score kernels, whether or not there are external data involved or just an overfitted model as the people claimed.\n\nThis is also a shaded area: what if there are some external data used in the kernels that are not selected for final submission? To some extent, the \"violation\" may not directly affect the final, selected kernel. The problem is, by submitting the weights for final inference purpose, it is relative hard to detect whether the external data are used for training that (those) model(s). But consider the GPU resources Kaggle have, everything should become easy. Just use their code and retrain 1 group. See how many downvotes for the \"public private 2nd place\" post.\n\nBUT, this time, everything seems to be too coincidence (there are similar situation before but not that kind of \"coincidence\"). As a part of the community has drawn the concerns and also (I believe) it is not hard to validate what we talked about, we hope Kaggle can carefully review this case. \n\n\n", "votes": 6 }, { "id": 551388, "postDate": "2019-06-12T17:01:13.693Z", "content": "@wcukierski Thanks for your statement! I can understand that Kaggle usually does not make public comment on such an incident. But since this time the incident draws a lot of attention and also @dempton posted publicly that he has evidence that some teams cheating in the competition, I hope that Kaggle can share the information about the investigation when the timing is appropriate.
\n\nHere are the questions I am concerned with:\n1. The top teams on public LB claimed their solution overfitting the LB, is that true? Do they choose the same kernel that achieved a high score in 1st stage for the 2nd stage? It it not, what is the 2nd stage socre of the high score kernel in the 1st stage?
\n\n@dempton said he has chat record with some of the top teams in public LB. So has kaggle teams got a chance to review his evidence?
Is this incident a single event or actually may suggest other violations in previous competitions? This time the abnormal high score draw people's attention, but similar behavior may not be easily noticed to give the characteristics of kaggle competition.
Is there any actions or plan that kaggle to prevent similar behavior we suspect here happened? Especially for the competition like imet, most people may use a similar model, so the dataset will play a vital role in the final grade. However, due to the blackbox of the deep learning model, it may not easy to disentangle how different the data used in the training process.
I really hope that kaggle team can answer my questions here! Thanks!
", "rawMarkdown": "@wcukierski Thanks for your statement! I can understand that Kaggle usually does not make public comment on such an incident. But since this time the incident draws a lot of attention and also @dempton posted publicly that he has evidence that some teams cheating in the competition, I hope that Kaggle can share the information about the investigation when the timing is appropriate.\n\nHere are the questions I am concerned with:\n1. The top teams on public LB claimed their solution overfitting the LB, is that true? Do they choose the same kernel that achieved a high score in 1st stage for the 2nd stage? It it not, what is the 2nd stage socre of the high score kernel in the 1st stage?\n\n2. @dempton said he has chat record with some of the top teams in public LB. So has kaggle teams got a chance to review his evidence? \n\n3. Is this incident a single event or actually may suggest other violations in previous competitions? This time the abnormal high score draw people's attention, but similar behavior may not be easily noticed to give the characteristics of kaggle competition.\n\n4. Is there any actions or plan that kaggle to prevent similar behavior we suspect here happened? Especially for the competition like imet, most people may use a similar model, so the dataset will play a vital role in the final grade. However, due to the blackbox of the deep learning model, it may not easy to disentangle how different the data used in the training process.\n\nI really hope that kaggle team can answer my questions here! Thanks!\n", "votes": 3 } ] }, { "id": 542433, "postDate": "2019-06-03T23:17:40.987Z", "content": "UPDATED: June 5th at 11:59 PM UTC is the Final Submission Deadline. After this date, we will not be taking any more submissions. Remember to select your two best submissions to be rescored during the re-run period.
\n\nWe will update the end date of the competition tomorrow to reflect the re-run period end date of June 11, 2019
", "rawMarkdown": "UPDATED: June 5th at 11:59 PM UTC is the Final Submission Deadline. After this date, we will not be taking any more submissions. Remember to select your two best submissions to be rescored during the re-run period. \n\nWe will update the end date of the competition tomorrow to reflect the re-run period end date of June 11, 2019", "votes": 6 }, { "id": 551230, "postDate": "2019-06-12T13:39:22.393Z", "content": "Daily check-in: what is the conclusion for the top 5 teams abnormal high score on public LB @inversion @maggiemd @wcukierski @codingcliff
", "rawMarkdown": "Daily check-in: what is the conclusion for the top 5 teams abnormal high score on public LB @inversion @maggiemd @wcukierski @codingcliff ", "votes": 4, "replies": [ { "id": 551249, "postDate": "2019-06-12T13:57:43.840Z", "content": "X5 is allowed to cheat :), let's join X5 and cheat!
", "rawMarkdown": "X5 is allowed to cheat :), let's join X5 and cheat!", "votes": -1 } ] }, { "id": 544782, "postDate": "2019-06-05T23:06:17.090Z", "content": "The submissions have been disabled and now we will be moving to the next phase of the competition.
", "rawMarkdown": "The submissions have been disabled and now we will be moving to the next phase of the competition. ", "votes": -3, "replies": [ { "id": 544795, "postDate": "2019-06-05T23:18:13.187Z", "content": "\n\nI think we have still 1 hour to that deadline.\nDeadline is 11:59pm on 5th of June, right? Why disabled now?
", "rawMarkdown": "@maggiemd \n\nI think we have still 1 hour to that deadline.\nDeadline is 11:59pm on 5th of June, right? Why disabled now?", "votes": 2 }, { "id": 544798, "postDate": "2019-06-05T23:22:01.203Z", "content": "Can you give us the chance to submit our results until 11:59pm as @wcukierski promised ?
\n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94501#latest-544805
\n\n> we will be extending the deadline of this competition by one day to compensate for any disruption.
\n\nThis is very disappointing. \nI had 4 of my best models to be submitted 1 hour before the deadline but couldn't !
", "rawMarkdown": "Can you give us the chance to submit our results until 11:59pm as @wcukierski promised ?\n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94501#latest-544805\n\n> we will be extending the deadline of this competition by one day to compensate for any disruption.\n\nThis is very disappointing. \nI had 4 of my best models to be submitted 1 hour before the deadline but couldn't !", "votes": 1 }, { "id": 544814, "postDate": "2019-06-05T23:43:32.327Z", "content": "@maggiemd \nIt was announced that 'we are extending the deadline by one day'. Shouldn't the deadline refer to 11:59PM UTC June 4th?
", "rawMarkdown": "@maggiemd \nIt was announced that 'we are extending the **deadline** by one day'. Shouldn't the deadline refer to 11:59PM UTC June 4th? ", "votes": 2 }, { "id": 544835, "postDate": "2019-06-06T00:18:31.713Z", "content": "I've opened up submissions and will allow them until 12:59 a.m. to account for the error in time translation.
", "rawMarkdown": "I've opened up submissions and will allow them until 12:59 a.m. to account for the error in time translation. ", "votes": 1 }, { "id": 545846, "postDate": "2019-06-06T00:34:28.180Z", "content": "Thanks @maggiemd , really appreciate it!
", "rawMarkdown": "Thanks @maggiemd , really appreciate it!" }, { "id": 545852, "postDate": "2019-06-06T00:42:24.637Z", "content": "Thanks indeed!
", "rawMarkdown": "Thanks indeed!" }, { "id": 545898, "postDate": "2019-06-06T02:44:44.273Z", "rawMarkdown": "" }, { "id": 545972, "postDate": "2019-06-06T05:12:53.367Z", "content": "Only tens of minutes reopened? I could not be staring at the discussion at that time!
", "rawMarkdown": "Only tens of minutes reopened? I could not be staring at the discussion at that time!" } ] }, { "id": 543235, "postDate": "2019-06-04T12:02:37.013Z", "content": "thank you , can we have more submissions in last day ?
", "rawMarkdown": "thank you , can we have more submissions in last day ?", "votes": -1 }, { "id": 549333, "postDate": "2019-06-10T15:01:42.163Z", "rawMarkdown": "", "votes": 1, "isDeleted": true }, { "id": 543861, "postDate": "2019-06-04T22:28:22.760Z", "rawMarkdown": "", "isDeleted": true }, { "id": 543852, "postDate": "2019-06-04T22:23:16.667Z", "rawMarkdown": "", "isDeleted": true }, { "id": 542629, "postDate": "2019-06-04T02:27:50.770Z", "content": "Thanks for info.
", "rawMarkdown": "Thanks for info." } ], "comments": [ { "id": 550712, "author_name": "Yiheng Wang", "author_url": "", "post_date": "2019-06-12T00:42:23.890000", "content": "Hi @maggiemd you’ve released the medals, and with out any explanations for the public Top 5. I’m very disappointed and cannot understand why. Previously when I got a medal I was happy, no matter a silver or a bronze. But this time, what I expected is the explanations and got nothing : )
", "votes": 9, "replies": [ { "id": 550715, "author_name": "Strideradu", "author_url": "", "post_date": "2019-06-12T00:49:29.813000", "content": "I think at least Kaggle should have a conclusion for this incident, whether they are cheating. If there is no violation, that would be great since everyone follows the rule
", "votes": 1, "replies": [] }, { "id": 550719, "author_name": "Strideradu", "author_url": "", "post_date": "2019-06-12T00:52:14.597000", "content": "", "votes": 1, "replies": [] }, { "id": 550723, "author_name": "Kulbear", "author_url": "", "post_date": "2019-06-12T00:59:41.520000", "content": "obviously we have no clue to say someone has cheated 100%. but as this is the largest and the most famous data science competition platform, we all know that it is popularized by its high quality competitions and strict rules. But really?
\n\nthere are sufficient amount of discussions and even someone who jumped out to reveal the possibly inappropriate behaviors done by a small group of people. Similar situation has occured several times and what we got from the kaggle admin team during the past two weeks is \"nothing\".
\n\nChecking their code and kernel and even uploaded model are not that hard for Kaggle. Can we understand this is a unofficial announcement for the tolerance to possibly inappropriate behaviors?
", "votes": 1, "replies": [] }, { "id": 550732, "author_name": "SeuTao", "author_url": "", "post_date": "2019-06-12T01:22:53.933000", "content": "I don't know why kaggle choose to ignore this :(
", "votes": 0, "replies": [] }, { "id": 550768, "author_name": "takuoko", "author_url": "", "post_date": "2019-06-12T02:27:19.883000", "content": "I disappointed. I don't know whether cheating is used in final results. But we hope explains from kaggle and top teams.
", "votes": 2, "replies": [] }, { "id": 550794, "author_name": "Xuan Cao", "author_url": "", "post_date": "2019-06-12T03:21:16.313000", "content": "\n\nSome of top teams have explained: \"we overfitted leaderboard\", and \"oh, kaggle cannot check my code since I am out of medal zone\".
\n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/95330#latest-550702
", "votes": 2, "replies": [] }, { "id": 550802, "author_name": "takuoko", "author_url": "", "post_date": "2019-06-12T03:30:06.360000", "content": "Now, cheating looks best solution;) hahaha.
", "votes": -4, "replies": [] }, { "id": 550846, "author_name": "Ahmet Erdem", "author_url": "", "post_date": "2019-06-12T04:44:36.733000", "content": "So private sharing and scraping are officially allowed? @inversion
", "votes": 1, "replies": [] }, { "id": 550865, "author_name": "earhian", "author_url": "", "post_date": "2019-06-12T05:16:58.710000", "content": "Sure, kaggle allow this! I will cheat in the next competition. ;)
", "votes": -1, "replies": [] }, { "id": 550870, "author_name": "SeuTao", "author_url": "", "post_date": "2019-06-12T05:27:25.200000", "content": "@qiaojian Join X5! Join the GM/Ms! Oh, they just stole the solo gold from you. Don‘t worry, you will have solo gooooolds！
", "votes": 2, "replies": [] }, { "id": 550871, "author_name": "earhian", "author_url": "", "post_date": "2019-06-12T05:27:27.277000", "content": "X5 is allowed to cheat in kaggle, and I have joined ods.ai. I think I can cheat in the next competition. :) By the way, X5 will set up an interview to me!
", "votes": 0, "replies": [] }, { "id": 550897, "author_name": "ONODERA", "author_url": "", "post_date": "2019-06-12T06:03:11.290000", "content": "@qiaojian contact us https://www.x5.ru/en/Pages/About/Contacts.aspx
", "votes": 3, "replies": [] }, { "id": 550905, "author_name": "earhian", "author_url": "", "post_date": "2019-06-12T06:08:00.250000", "content": "@onodera， it‘s time for me to learn crawler!
", "votes": 3, "replies": [] }, { "id": 551305, "author_name": "Will Cukierski", "author_url": "", "post_date": "2019-06-12T15:15:13.457000", "content": "Hi all, a couple of reminders and remarks on this thread:
\n\nIn general, Kaggle does not make public comments about users, compliance investigations, or other private matters. We do not engage in public speculation. The absence of a removal does not mean a rule wasn't broken (we might simply lack the \"evidence beyond a reasonable doubt\"). A removal does not mean a person had malicious intent (accidents happen, English is a 2nd language for many, and most people simply don't read the rules).
Competition closes take time. We are not \"choosing to ignore\" a situation because two days have passed.
In all competitions, we finalize the leaderboard (the action that releases medals) before the full winner obligations are fulfilled. This is done to prevent everyone from having to wait for weeks while we close the competition. We have and can remove teams after medals have gone out. We have removed teams from leaderboards years after the competition closed due to discovery of new evidence.
\"Can we understand this is a[n] unofficial announcement for the tolerance to possibly inappropriate behaviors?\" No. The rules, site terms, your morals, and our collective community guidelines should govern your behavior here. You and only you can choose to conduct yourself with professionalism and integrity, or you can choose cut corners, pass blame, and justify inappropriate behaviors because others might be cheating the system.
Thanks for your reply, which is constructive and reasonable.\nThere are that many discussions because we believe it is very easy to test those abnormally high score kernels, whether or not there are external data involved or just an overfitted model as the people claimed.
\n\nThis is also a shaded area: what if there are some external data used in the kernels that are not selected for final submission? To some extent, the \"violation\" may not directly affect the final, selected kernel. The problem is, by submitting the weights for final inference purpose, it is relative hard to detect whether the external data are used for training that (those) model(s). But consider the GPU resources Kaggle have, everything should become easy. Just use their code and retrain 1 group. See how many downvotes for the \"public private 2nd place\" post.
\n\nBUT, this time, everything seems to be too coincidence (there are similar situation before but not that kind of \"coincidence\"). As a part of the community has drawn the concerns and also (I believe) it is not hard to validate what we talked about, we hope Kaggle can carefully review this case.
", "votes": 6, "replies": [] }, { "id": 551388, "author_name": "Strideradu", "author_url": "", "post_date": "2019-06-12T17:01:13.693000", "content": "@wcukierski Thanks for your statement! I can understand that Kaggle usually does not make public comment on such an incident. But since this time the incident draws a lot of attention and also @dempton posted publicly that he has evidence that some teams cheating in the competition, I hope that Kaggle can share the information about the investigation when the timing is appropriate.
\n\nHere are the questions I am concerned with:\n1. The top teams on public LB claimed their solution overfitting the LB, is that true? Do they choose the same kernel that achieved a high score in 1st stage for the 2nd stage? It it not, what is the 2nd stage socre of the high score kernel in the 1st stage?
\n\n@dempton said he has chat record with some of the top teams in public LB. So has kaggle teams got a chance to review his evidence?
Is this incident a single event or actually may suggest other violations in previous competitions? This time the abnormal high score draw people's attention, but similar behavior may not be easily noticed to give the characteristics of kaggle competition.
Is there any actions or plan that kaggle to prevent similar behavior we suspect here happened? Especially for the competition like imet, most people may use a similar model, so the dataset will play a vital role in the final grade. However, due to the blackbox of the deep learning model, it may not easy to disentangle how different the data used in the training process.
I really hope that kaggle team can answer my questions here! Thanks!
", "votes": 3, "replies": [] } ] }, { "id": 551230, "author_name": "Strideradu", "author_url": "", "post_date": "2019-06-12T13:39:22.393000", "content": "Daily check-in: what is the conclusion for the top 5 teams abnormal high score on public LB @inversion @maggiemd @wcukierski @codingcliff
", "votes": 4, "replies": [ { "id": 551249, "author_name": "earhian", "author_url": "", "post_date": "2019-06-12T13:57:43.840000", "content": "X5 is allowed to cheat :), let's join X5 and cheat!
", "votes": -1, "replies": [] } ] }, { "id": 544782, "author_name": "Maggie", "author_url": "", "post_date": "2019-06-05T23:06:17.090000", "content": "The submissions have been disabled and now we will be moving to the next phase of the competition.
", "votes": -3, "replies": [ { "id": 544795, "author_name": "Tawara", "author_url": "", "post_date": "2019-06-05T23:18:13.187000", "content": "\n\nI think we have still 1 hour to that deadline.\nDeadline is 11:59pm on 5th of June, right? Why disabled now?
", "votes": 2, "replies": [] }, { "id": 544798, "author_name": "Haider Alwasiti", "author_url": "", "post_date": "2019-06-05T23:22:01.203000", "content": "Can you give us the chance to submit our results until 11:59pm as @wcukierski promised ?
\n\nhttps://www.kaggle.com/c/imet-2019-fgvc6/discussion/94501#latest-544805
\n\n> we will be extending the deadline of this competition by one day to compensate for any disruption.
\n\nThis is very disappointing. \nI had 4 of my best models to be submitted 1 hour before the deadline but couldn't !
", "votes": 1, "replies": [] }, { "id": 544814, "author_name": "Xuan Cao", "author_url": "", "post_date": "2019-06-05T23:43:32.327000", "content": "@maggiemd \nIt was announced that 'we are extending the deadline by one day'. Shouldn't the deadline refer to 11:59PM UTC June 4th?
", "votes": 2, "replies": [] }, { "id": 544835, "author_name": "Maggie", "author_url": "", "post_date": "2019-06-06T00:18:31.713000", "content": "I've opened up submissions and will allow them until 12:59 a.m. to account for the error in time translation.
", "votes": 1, "replies": [] }, { "id": 545846, "author_name": "Xuan Cao", "author_url": "", "post_date": "2019-06-06T00:34:28.180000", "content": "Thanks @maggiemd , really appreciate it!
", "votes": 0, "replies": [] }, { "id": 545852, "author_name": "Haider Alwasiti", "author_url": "", "post_date": "2019-06-06T00:42:24.637000", "content": "Thanks indeed!
", "votes": 0, "replies": [] }, { "id": 545898, "author_name": "ghostintheshell", "author_url": "", "post_date": "2019-06-06T02:44:44.273000", "content": "", "votes": 0, "replies": [] }, { "id": 545972, "author_name": "", "author_url": "", "post_date": "2019-06-06T05:12:53.367000", "content": "Only tens of minutes reopened? I could not be staring at the discussion at that time!
", "votes": 0, "replies": [] } ] }, { "id": 543235, "author_name": "soywu", "author_url": "", "post_date": "2019-06-04T12:02:37.013000", "content": "thank you , can we have more submissions in last day ?
", "votes": -1, "replies": [] }, { "id": 549333, "author_name": "", "author_url": "", "post_date": "2019-06-10T15:01:42.163000", "content": "", "votes": 1, "replies": [] }, { "id": 543861, "author_name": "", "author_url": "", "post_date": "2019-06-04T22:28:22.760000", "content": "", "votes": 0, "replies": [] }, { "id": 543852, "author_name": "", "author_url": "", "post_date": "2019-06-04T22:23:16.667000", "content": "", "votes": 0, "replies": [] }, { "id": 542629, "author_name": "Jar", "author_url": "", "post_date": "2019-06-04T02:27:50.770000", "content": "Thanks for info.
", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "550712": "Hi @maggiemd you’ve released the medals, and with out any explanations for the public Top 5. I’m very disappointed and cannot understand why. Previously when I got a medal I was happy, no matter a silver or a bronze. But this time, what I expected is the explanations and got nothing : )", "542433": "UPDATED: June 5th at 11:59 PM UTC is the Final Submission Deadline. After this date, we will not be taking any more submissions. Remember to select your two best submissions to be rescored during the re-run period. \n\nWe will update the end date of the competition tomorrow to reflect the re-run period end date of June 11, 2019", "551230": "Daily check-in: what is the conclusion for the top 5 teams abnormal high score on public LB @inversion @maggiemd @wcukierski @codingcliff ", "544782": "The submissions have been disabled and now we will be moving to the next phase of the competition. ", "543235": "thank you , can we have more submissions in last day ?", "549333": "", "543861": "", "543852": "", "542629": "Thanks for info." } }