Hey Kagglers,
\nWe’re happy to announce the conclusion of the TensorFlow - Help Protect the Great Barrier Reef competition! We’ve worked with the TensorFlow team in the past on other competitions, and when they told us about the upcoming $1 Billion (AUD) commitment to investing in Australia through the Digital Future Initiative which included a partnership with CSIRO, we were happy to help! We’re proud to be a home where we can make a positive impact on the world, especially when those opportunities help further the state-of-the-art in machine learning and help Kagglers grow.
\nThis competition ended with 14,549 registrations, 2,613 participants on 2,026 teams! We had 61,174 submissions from 93 countries! For 518 users (including 19 users on top 100 teams!), this was their first competition. An additional congratulations goes out to our 12 new Kaggle Competition Masters and 68 new Kaggle Competition Experts at the conclusion of this competition! Thank you all for your hard work in this competition and congratulations to our winners and to those who gained a new ranking!
\nA few more fun facts:
\n1) The winning submission was made on the last day of the competition, with only 11 hours left before the deadline!
\n2) The top TensorFlow Performance submission finished in the Top 10% and ran in only 71 minutes!
I also want to give a massive thanks to our many hosts of this competition. From Google/TensorFlow: Glenn, and Megs, and at CSIRO: Ryan and Brano for bringing the idea to our platform and working to make this such a successful event!
\nWe're also continually pleased to see how Kaggle Competitions serve as a great medium and vehicle through which the greater data science community can learn and develop their machine learning abilities and skills. Beginners and experts can come together to start, grow, and succeed. Further, in this competition, many can use their machine learning skills to help a great cause like this one.
\nWe've cleaned the leaderboard and disqualified some teams that have violated the rules. If you think you were removed by mistake, or believe you have evidence that suggests another team cheated, please contact compliance. Please fill in all the fields honestly.
\nThe top potential winning teams (both for the accuracy prize and Tensorflow Performance Prize) will be contacted shortly via email to provide their winning solutions for host review, which is underway. Should we need to move down the leaderboard for any reason, we will do so and continue to make contact with subsequent teams.
\nYou should have already seen your points and medals awarded, however it may take a few hours to propagate through if yours are still missing.
\nIn the meantime, I've created this thread to be a curated list of your work. As you post your solutions, top notebooks, and general discussion posts, I'll update this thread to keep track. If you write a paper, thesis, or present at a conference, and would like to share your work, please let us know so we can share with the Kaggle community! Additionally, if you performed any data manipulation, or used a new technique in this competition that you'd like to share to further the industry, please let us know and we'll post it here! That way, should you be looking for takeaways in the future, or if you've just now stumbled across the competition, you have a place to see what surfaced from the work performed.
\nHappy Modeling!
\nKaggle Team
Hey Kagglers,
\nWe’re happy to announce the conclusion of the TensorFlow - Help Protect the Great Barrier Reef competition! We’ve worked with the TensorFlow team in the past on other competitions, and when they told us about the upcoming $1 Billion (AUD) commitment to investing in Australia through the Digital Future Initiative which included a partnership with CSIRO, we were happy to help! We’re proud to be a home where we can make a positive impact on the world, especially when those opportunities help further the state-of-the-art in machine learning and help Kagglers grow.
\nThis competition ended with 14,549 registrations, 2,613 participants on 2,026 teams! We had 61,174 submissions from 93 countries! For 518 users (including 19 users on top 100 teams!), this was their first competition. An additional congratulations goes out to our 12 new Kaggle Competition Masters and 68 new Kaggle Competition Experts at the conclusion of this competition! Thank you all for your hard work in this competition and congratulations to our winners and to those who gained a new ranking!
\nA few more fun facts:
\n1) The winning submission was made on the last day of the competition, with only 11 hours left before the deadline!
\n2) The top TensorFlow Performance submission finished in the Top 10% and ran in only 71 minutes!
I also want to give a massive thanks to our many hosts of this competition. From Google/TensorFlow: Glenn, and Megs, and at CSIRO: Ryan and Brano for bringing the idea to our platform and working to make this such a successful event!
\nWe're also continually pleased to see how Kaggle Competitions serve as a great medium and vehicle through which the greater data science community can learn and develop their machine learning abilities and skills. Beginners and experts can come together to start, grow, and succeed. Further, in this competition, many can use their machine learning skills to help a great cause like this one.
\nWe've cleaned the leaderboard and disqualified some teams that have violated the rules. If you think you were removed by mistake, or believe you have evidence that suggests another team cheated, please contact compliance. Please fill in all the fields honestly.
\nThe top potential winning teams (both for the accuracy prize and Tensorflow Performance Prize) will be contacted shortly via email to provide their winning solutions for host review, which is underway. Should we need to move down the leaderboard for any reason, we will do so and continue to make contact with subsequent teams.
\nYou should have already seen your points and medals awarded, however it may take a few hours to propagate through if yours are still missing.
\nIn the meantime, I've created this thread to be a curated list of your work. As you post your solutions, top notebooks, and general discussion posts, I'll update this thread to keep track. If you write a paper, thesis, or present at a conference, and would like to share your work, please let us know so we can share with the Kaggle community! Additionally, if you performed any data manipulation, or used a new technique in this competition that you'd like to share to further the industry, please let us know and we'll post it here! That way, should you be looking for takeaways in the future, or if you've just now stumbled across the competition, you have a place to see what surfaced from the work performed.
\nHappy Modeling!
\nKaggle Team
@addisonhoward it is really strange to see my disqualification from 12th place (0.718, gold medal). At least it would be useful to see some notifcation with a potential reason.
\nI did not spend much time on this challenge, just made a strong RCNN baseline in one week, selected submissions based on CV and waited a few weeks for a shakeup.
\nThe shakeup was as expected but removing from LB was definitely not the thing that I expected.
Seems like Kaggle uses some predictive model that identifies violations and kicks out teams based solely on (mis)predictions…
\nWithout a manual review, especially for high scoring submissions that does make Kaggle more like a gambling platform.
Hi there,
\nAll removals are made with individual manual of each violation. Should you believe you were removed by accident, please contact Kaggle Compliance.
", "rawMarkdown": "Hi there,\n\nAll removals are made with individual manual of each violation. Should you believe you were removed by accident, please contact Kaggle Compliance." }, { "id": 1697570, "postDate": "2022-02-19T17:35:30.780Z", "content": "Thanks for the quick response!
\nYes, I filled a compliance form.
Congrats to all the winners! We have learned so much from all the competitors. It has been a great pleasure to participate in the competition. Thank you, our sponsors, all the participants and Kaggle staff.
", "rawMarkdown": "Congrats to all the winners! We have learned so much from all the competitors. It has been a great pleasure to participate in the competition. Thank you, our sponsors, all the participants and Kaggle staff.", "votes": 1 }, { "id": 1694998, "postDate": "2022-02-17T22:02:33.603Z", "content": "\n\n\nWe've cleaned the leaderboard and disqualified some teams that have violated the rules.
\n
Does it mean that you didn't disqualify all teams that violated the rules?
\nDoes it mean that even publicly admitted private sharing long before forming a team is now allowed?
\nYou personally in that discussion admitted that it had been a rule violation and therefore you must have enough case evidence but the team is still on the leaderboard!
\nDoes it mean that Kaggle completely gave up to keep the competitions fair because rule violations are so wide spread that you don't manage to follow all the cheating?
Hi Allie,
\nPrivate sharing before forming a team is not allowed. The team in question has been removed.
\nSweeping generalizations and leading questions are unnecessary. For any team that you believe has violated competition rules, and you have evidence to support such, you may contact compliance.
", "rawMarkdown": "Hi Allie,\n\nPrivate sharing before forming a team is not allowed. The team in question has been removed. \n\nSweeping generalizations and leading questions are unnecessary. For any team that you believe has violated competition rules, and you have evidence to support such, you may contact [compliance](https://www.kaggle.com/compliance).", "votes": 1 }, { "id": 1695060, "postDate": "2022-02-17T23:27:15.953Z", "content": "@addisonhoward
\nall my reaction was only inspired by your words and the fact how many teams were this time disqualified (now 83).
\nI hope that your expression \"… we disqualified some teams that have violated the rules\" was only misunderstood by me.
Correct - the \"some\" in \"we disqualified some teams\" should be understood as \"of all the teams that participated, some of them violated the rules, and we removed those teams\" and not \"of all those who violated the rules, we removed some of the violators\"
", "rawMarkdown": "Correct - the \"some\" in \"we disqualified some teams\" should be understood as \"of all the teams that participated, some of them violated the rules, and we removed those teams\" and not \"of all those who violated the rules, we removed some of the violators\"", "votes": 1 }, { "id": 1695202, "postDate": "2022-02-18T02:35:02.280Z", "content": "@addisonhoward @blankaf there's a big misunderstanding here. My teammate just said in the commnet that \"he wanted to do this\", but in practice, this situation did not occur. In fact, this is my first time at Kaggle, and I am not familiar with many operations on Kaggle, So I didn't take immediate action. The discussion lasted about a day. It was not until the end of the discussion in the comment area, after we merged the team, I saw and added my teammate's 0.669 model in my code.
", "rawMarkdown": "@addisonhoward @blankaf there's a big misunderstanding here. My teammate just said in the commnet that \"he wanted to do this\", but in practice, this situation did not occur. In fact, this is my first time at Kaggle, and I am not familiar with many operations on Kaggle, So I didn't take immediate action. The discussion lasted about a day. It was not until the end of the discussion in the comment area, after we merged the team, I saw and added my teammate's 0.669 model in my code." }, { "id": 1695399, "postDate": "2022-02-18T05:50:32.917Z", "content": "I think it is unfair for us to directly negate the achievements of others only based on one person's one-sided remarks and conclusions.
", "rawMarkdown": "I think it is unfair for us to directly negate the achievements of others only based on one person's one-sided remarks and conclusions." }, { "id": 1695677, "postDate": "2022-02-18T10:04:10.967Z", "content": "It wasn't only one comment made by Calvchen but all the discussion thread which he then deleted. And it wasn't only him but also Charlie who in another discussion wrote: \"I just want to share this code to my teammate\" although at that time he didn't have any teammate.
\nSo your disqualification is completely fair because your team repeatedly and evidently violated rules.
“So your disqualification is completely fair because your team repeatedly and evidently violated rules.”
\nBased on two comments from one of my teammate?
“He just want to share this code”, but in fact, he made it public.
\nAlthough we are all Chinese, we come from different countries and regions, and we only find each other through this Kaggle competition's discussion. At that time, we haven't trust each other, and we didn't reach a consensus on the team merger communication.
In fact, Charlie reluctantly agreed to the team merger because, during the discussion, he realized he had to merge the team with someone else or it would be against the rules. He wasn't familiar with the rest of the team at the time, and he was scoring more points than everyone else, and other teams were scoring far less than him.
\nIn fact, the rest of us were able to join Charlie's high-scoring team at the time because of that discussion.
", "rawMarkdown": "“So your disqualification is completely fair because your team repeatedly and evidently violated rules.”\nBased on two comments from one of my teammate? \n\n“He just want to share this code”, but in fact, he made it public. \nAlthough we are all Chinese, we come from different countries and regions, and we only find each other through this Kaggle competition's discussion. At that time, we haven't trust each other, and we didn't reach a consensus on the team merger communication. \n\nIn fact, Charlie reluctantly agreed to the team merger because, during the discussion, he realized he had to merge the team with someone else or it would be against the rules. He wasn't familiar with the rest of the team at the time, and he was scoring more points than everyone else, and other teams were scoring far less than him.\n\nIn fact, the rest of us were able to join Charlie's high-scoring team at the time because of that discussion. ", "votes": -1 } ] }, { "id": 1710709, "postDate": "2022-03-03T08:59:11.937Z", "content": "the 7th Place link is a 2th Place link.it is a error
", "rawMarkdown": "the 7th Place link is a 2th Place link.it is a error" }, { "id": 1701127, "postDate": "2022-02-22T14:36:58.780Z", "content": "congrats to all winners
", "rawMarkdown": "congrats to all winners" }, { "id": 1699868, "postDate": "2022-02-21T13:47:44.707Z", "content": "\nThank you for holding the competition.
\nWe created a feedback post for this competition to make the future competition better.
\nI know you are busy, but could you please check it out?
\nA summary of the feedback has been created below.
Post: https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/308329
\nSummary: https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/308329#1699343
Thank you @addisonhoward for such great competition. Once again thank you everybody and see you on next competition.
", "rawMarkdown": "Thank you @addisonhoward for such great competition. Once again thank you everybody and see you on next competition. " }, { "id": 1696297, "postDate": "2022-02-18T18:19:58.853Z", "content": "Thank you @addisonhoward for featuring our write-up but unfortunately we got disqualified. I have clarified reasons here: https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/308455
", "rawMarkdown": "Thank you @addisonhoward for featuring our write-up but unfortunately we got disqualified. I have clarified reasons here: https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/308455" }, { "id": 1695531, "postDate": "2022-02-18T07:47:05.840Z", "content": "Why I was excluded from the final protocol? I was on 58 position with 0.689. Now I'm not in public and private leaderboard.
", "rawMarkdown": "Why I was excluded from the final protocol? I was on 58 position with 0.689. Now I'm not in public and private leaderboard." }, { "id": 1701679, "postDate": "2022-02-23T01:20:16.617Z", "rawMarkdown": "", "isDeleted": true } ], "comments": [ { "id": 1697298, "author_name": "Selim Seferbekov", "author_url": "", "post_date": "2022-02-19T14:16:12.833000", "content": "@addisonhoward it is really strange to see my disqualification from 12th place (0.718, gold medal). At least it would be useful to see some notifcation with a potential reason.
\nI did not spend much time on this challenge, just made a strong RCNN baseline in one week, selected submissions based on CV and waited a few weeks for a shakeup.
\nThe shakeup was as expected but removing from LB was definitely not the thing that I expected.
Seems like Kaggle uses some predictive model that identifies violations and kicks out teams based solely on (mis)predictions…
\nWithout a manual review, especially for high scoring submissions that does make Kaggle more like a gambling platform.
Hi there,
\nAll removals are made with individual manual of each violation. Should you believe you were removed by accident, please contact Kaggle Compliance.
", "votes": 0, "replies": [] }, { "id": 1697570, "author_name": "Selim Seferbekov", "author_url": "", "post_date": "2022-02-19T17:35:30.780000", "content": "Thanks for the quick response!
\nYes, I filled a compliance form.
Congrats to all the winners! We have learned so much from all the competitors. It has been a great pleasure to participate in the competition. Thank you, our sponsors, all the participants and Kaggle staff.
", "votes": 1, "replies": [] }, { "id": 1694998, "author_name": "Allie K.", "author_url": "", "post_date": "2022-02-17T22:02:33.603000", "content": "\n\n\nWe've cleaned the leaderboard and disqualified some teams that have violated the rules.
\n
Does it mean that you didn't disqualify all teams that violated the rules?
\nDoes it mean that even publicly admitted private sharing long before forming a team is now allowed?
\nYou personally in that discussion admitted that it had been a rule violation and therefore you must have enough case evidence but the team is still on the leaderboard!
\nDoes it mean that Kaggle completely gave up to keep the competitions fair because rule violations are so wide spread that you don't manage to follow all the cheating?
Hi Allie,
\nPrivate sharing before forming a team is not allowed. The team in question has been removed.
\nSweeping generalizations and leading questions are unnecessary. For any team that you believe has violated competition rules, and you have evidence to support such, you may contact compliance.
", "votes": 1, "replies": [] }, { "id": 1695060, "author_name": "Allie K.", "author_url": "", "post_date": "2022-02-17T23:27:15.953000", "content": "@addisonhoward
\nall my reaction was only inspired by your words and the fact how many teams were this time disqualified (now 83).
\nI hope that your expression \"… we disqualified some teams that have violated the rules\" was only misunderstood by me.
Correct - the \"some\" in \"we disqualified some teams\" should be understood as \"of all the teams that participated, some of them violated the rules, and we removed those teams\" and not \"of all those who violated the rules, we removed some of the violators\"
", "votes": 1, "replies": [] }, { "id": 1695202, "author_name": "Phoenix Ash", "author_url": "", "post_date": "2022-02-18T02:35:02.280000", "content": "@addisonhoward @blankaf there's a big misunderstanding here. My teammate just said in the commnet that \"he wanted to do this\", but in practice, this situation did not occur. In fact, this is my first time at Kaggle, and I am not familiar with many operations on Kaggle, So I didn't take immediate action. The discussion lasted about a day. It was not until the end of the discussion in the comment area, after we merged the team, I saw and added my teammate's 0.669 model in my code.
", "votes": 0, "replies": [] }, { "id": 1695399, "author_name": "Phoenix Ash", "author_url": "", "post_date": "2022-02-18T05:50:32.917000", "content": "I think it is unfair for us to directly negate the achievements of others only based on one person's one-sided remarks and conclusions.
", "votes": 0, "replies": [] }, { "id": 1695677, "author_name": "Allie K.", "author_url": "", "post_date": "2022-02-18T10:04:10.967000", "content": "It wasn't only one comment made by Calvchen but all the discussion thread which he then deleted. And it wasn't only him but also Charlie who in another discussion wrote: \"I just want to share this code to my teammate\" although at that time he didn't have any teammate.
\nSo your disqualification is completely fair because your team repeatedly and evidently violated rules.
“So your disqualification is completely fair because your team repeatedly and evidently violated rules.”
\nBased on two comments from one of my teammate?
“He just want to share this code”, but in fact, he made it public.
\nAlthough we are all Chinese, we come from different countries and regions, and we only find each other through this Kaggle competition's discussion. At that time, we haven't trust each other, and we didn't reach a consensus on the team merger communication.
In fact, Charlie reluctantly agreed to the team merger because, during the discussion, he realized he had to merge the team with someone else or it would be against the rules. He wasn't familiar with the rest of the team at the time, and he was scoring more points than everyone else, and other teams were scoring far less than him.
\nIn fact, the rest of us were able to join Charlie's high-scoring team at the time because of that discussion.
", "votes": -1, "replies": [] } ] }, { "id": 1710709, "author_name": "Kira yang", "author_url": "", "post_date": "2022-03-03T08:59:11.937000", "content": "the 7th Place link is a 2th Place link.it is a error
", "votes": 0, "replies": [] }, { "id": 1701127, "author_name": "AngadY", "author_url": "", "post_date": "2022-02-22T14:36:58.780000", "content": "congrats to all winners
", "votes": 0, "replies": [] }, { "id": 1699868, "author_name": "Bilzard", "author_url": "", "post_date": "2022-02-21T13:47:44.707000", "content": "\nThank you for holding the competition.
\nWe created a feedback post for this competition to make the future competition better.
\nI know you are busy, but could you please check it out?
\nA summary of the feedback has been created below.
Post: https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/308329
\nSummary: https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/308329#1699343
Thank you @addisonhoward for such great competition. Once again thank you everybody and see you on next competition.
", "votes": 0, "replies": [] }, { "id": 1696297, "author_name": "Sanchit Vijay", "author_url": "", "post_date": "2022-02-18T18:19:58.853000", "content": "Thank you @addisonhoward for featuring our write-up but unfortunately we got disqualified. I have clarified reasons here: https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/308455
", "votes": 0, "replies": [] }, { "id": 1695531, "author_name": "Aleksandr Kruchinin", "author_url": "", "post_date": "2022-02-18T07:47:05.840000", "content": "Why I was excluded from the final protocol? I was on 58 position with 0.689. Now I'm not in public and private leaderboard.
", "votes": 0, "replies": [] }, { "id": 1701679, "author_name": "", "author_url": "", "post_date": "2022-02-23T01:20:16.617000", "content": "", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "1694952": "Hey Kagglers,\n\nWe’re happy to announce the conclusion of the TensorFlow - Help Protect the Great Barrier Reef competition! We’ve worked with the TensorFlow team in the past on other competitions, and when they told us about the upcoming $1 Billion (AUD) commitment to investing in Australia through the [Digital Future Initiative](https://blog.google/intl/en-au/company-news/outreach-initiatives/digital-future-initiative/) which included a partnership with CSIRO, we were happy to help! We’re proud to be a home where we can make a positive impact on the world, especially when those opportunities help further the state-of-the-art in machine learning and help Kagglers grow.\n\nThis competition ended with 14,549 registrations, 2,613 participants on 2,026 teams! We had 61,174 submissions from 93 countries! For 518 users (including 19 users on top 100 teams!), this was their first competition. An additional congratulations goes out to our 12 new Kaggle Competition Masters and 68 new Kaggle Competition Experts at the conclusion of this competition! Thank you all for your hard work in this competition and congratulations to our winners and to those who gained a new ranking! \n\nA few more fun facts: \n1) The winning submission was made on the last day of the competition, with only 11 hours left before the deadline! \n2) The top TensorFlow Performance submission finished in the Top 10% and ran in only 71 minutes!\n\nI also want to give a massive thanks to our many hosts of this competition. From Google/TensorFlow: [Glenn](https://www.kaggle.com/glennglen), and [Megs](https://www.kaggle.com/megsmalpani), and at CSIRO: [Ryan](https://www.kaggle.com/ryanliu2021) and Brano for bringing the idea to our platform and working to make this such a successful event! \n\nWe're also continually pleased to see how Kaggle Competitions serve as a great medium and vehicle through which the greater data science community can learn and develop their machine learning abilities and skills. Beginners and experts can come together to start, grow, and succeed. Further, in this competition, many can use their machine learning skills to help a great cause like this one.\n\nWe've cleaned the leaderboard and disqualified some teams that have violated the rules. If you think you were removed by mistake, or believe you have evidence that suggests another team cheated, please contact [compliance](https://www.kaggle.com/compliance). Please fill in all the fields honestly.\n\nThe top potential winning teams (both for the accuracy prize and Tensorflow Performance Prize) will be contacted shortly via email to provide their winning solutions for host review, which is underway. Should we need to move down the leaderboard for any reason, we will do so and continue to make contact with subsequent teams. \n\nYou should have already seen your points and medals awarded, however it may take a few hours to propagate through if yours are still missing.\n\nIn the meantime, I've created this thread to be a curated list of your work. As you post your solutions, top notebooks, and general discussion posts, I'll update this thread to keep track. If you write a paper, thesis, or present at a conference, and would like to share your work, please let us know so we can share with the Kaggle community! Additionally, if you performed any data manipulation, or used a new technique in this competition that you'd like to share to further the industry, please let us know and we'll post it here! That way, should you be looking for takeaways in the future, or if you've just now stumbled across the competition, you have a place to see what surfaced from the work performed.\n\nHappy Modeling!\nKaggle Team\n\n###Top Solution Write-ups\n\n- [1st Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307878)\n- [2nd Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307760)\n- [3rd Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307707)\n- [4th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307626)\n- [5th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/308007)\n- [6th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307619)\n- [7th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307760)\n- [8th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307735)\n- [10th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307756)\n- [11th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307718)\n- [12th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307753)\n- [15th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307691)\n- [22nd Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307622)\n- [31st Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307504)\n- [37th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307669)\n- [38th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307807)\n- [45th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307825)\n- [63rd Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307800)\n- [69th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307609)\n- [77th Place](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/307634)\n\n###Other Top Discussions\n- [🔥🔥Winning solution of Previous Object Detection Kaggle Challenges](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/289999)\n- [My Kaggle Lessons Learned in 2021](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/297863)\n- [[LB 0.579] Yolov5(higher resolution) is all you need](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/300638)\n- [[placeholder] how to get lb 0.560 with single model](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/300405)\n- [🔥🔥Kaggle Starter Notebooks for Object Detection](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/290016)\n- [Competition logbook - ideas to IMPROVE LB score 💪💥](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/296721)\n- [Be CAREFUL with your TRAIN/VALID splits and avoid LB GAPS](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/293723)\n- [Best Single Model CV-LB](https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/290757)\n\n###Top Notebooks\n- [Great-Barrier-Reef: YOLOv5 [train] 🌊](https://www.kaggle.com/awsaf49/great-barrier-reef-yolov5-train)\n- [YoloX training pipeline COTS dataset [LB 0.507] !!](https://www.kaggle.com/remekkinas/yolox-training-pipeline-cots-dataset-lb-0-507)\n- [Great-Barrier-Reef: YOLOv5 [infer] 🌊](https://www.kaggle.com/awsaf49/great-barrier-reef-yolov5-infer)\n- [YoloX inference + Tracking on COTS [LB 0.539]](https://www.kaggle.com/parapapapam/yolox-inference-tracking-on-cots-lb-0-539)\n- [YoloX inference on Kaggle for COTS [LB 0.507] !!!](https://www.kaggle.com/remekkinas/yolox-inference-on-kaggle-for-cots-lb-0-507)\n- [Learning to Sea: Underwater img Enhancement + EDA](https://www.kaggle.com/soumya9977/learning-to-sea-underwater-img-enhancement-eda)\n- [🐠 Reef - A CV strategy: subsequences!](https://www.kaggle.com/julian3833/reef-a-cv-strategy-subsequences)\n- [Yolov5 is all you need](https://www.kaggle.com/steamedsheep/yolov5-is-all-you-need)\n- [competition metric implementation](https://www.kaggle.com/bamps53/competition-metric-implementation)\n- [🐡GreatBarrierReef: Full Guide to BBoxAugmentation](https://www.kaggle.com/andradaolteanu/greatbarrierreef-full-guide-to-bboxaugmentation)\n- [yolov5 high resolution training](https://www.kaggle.com/steamedsheep/yolov5-high-resolution-training)\n- [⭐️ Great Barrier Reefs 🐠: EDA with Animations](https://www.kaggle.com/diegoalejogm/great-barrier-reefs-eda-with-animations)\n\n\n\n", "1697298": "@addisonhoward it is really strange to see my disqualification from 12th place (0.718, gold medal). At least it would be useful to see some notifcation with a potential reason. \nI did not spend much time on this challenge, just made a strong RCNN baseline in one week, selected submissions based on CV and waited a few weeks for a shakeup. \nThe shakeup was as expected but removing from LB was definitely not the thing that I expected.\n\nSeems like Kaggle uses some predictive model that identifies violations and kicks out teams based solely on (mis)predictions… \nWithout a manual review, especially for high scoring submissions that does make Kaggle more like a gambling platform. ", "1694986": "Congrats to all the winners! We have learned so much from all the competitors. It has been a great pleasure to participate in the competition. Thank you, our sponsors, all the participants and Kaggle staff.", "1694998": "@addisonhoward \n> We've cleaned the leaderboard and disqualified some teams that have violated the rules.\n\nDoes it mean that you didn't disqualify **all teams** that violated the rules?\nDoes it mean that even publicly admitted private sharing long before forming a team is now allowed? \n**You personally** in that discussion admitted that it had been a rule violation and therefore you must have enough case evidence but the team is still on the leaderboard!\nDoes it mean that Kaggle completely gave up to keep the competitions fair because rule violations are so wide spread that you don't manage to follow all the cheating? \n", "1710709": "the 7th Place link is a 2th Place link.it is a error", "1701127": "congrats to all winners", "1699868": "@addisonhoward \n\nThank you for holding the competition.\nWe created a feedback post for this competition to make the future competition better.\nI know you are busy, but could you please check it out?\nA summary of the feedback has been created below.\n\nPost: https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/308329\nSummary: https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/308329#1699343", "1696471": "Thank you @addisonhoward for such great competition. Once again thank you everybody and see you on next competition. ", "1696297": "Thank you @addisonhoward for featuring our write-up but unfortunately we got disqualified. I have clarified reasons here: https://www.kaggle.com/c/tensorflow-great-barrier-reef/discussion/308455", "1695531": "Why I was excluded from the final protocol? I was on 58 position with 0.689. Now I'm not in public and private leaderboard.", "1701679": "" } }