{ "id": 152446, "title": "Oh Great", "url": "/competitions/open-images-object-detection-rvc-2020/discussion/152446", "author_name": "Bojan Tunguz", "post_date": "2020-05-19T22:07:26.628000", "votes": 38, "comment_count": 18, "views": 0, "content": "

Two new image competitions, with huge datasets, requiring a boatload of computational power, and NO points or medals. Yeah, these are going to be enormously successful. 🙄

", "messages": [ { "id": 854238, "postDate": "2020-05-19T22:07:26.630Z", "content": "

Two new image competitions, with huge datasets, requiring a boatload of computational power, and NO points or medals. Yeah, these are going to be enormously successful. 🙄

", "rawMarkdown": "Two new image competitions, with huge datasets, requiring a boatload of computational power, and **NO** points or medals. Yeah, these are going to be enormously successful. 🙄 ", "votes": 38 }, { "id": 854328, "postDate": "2020-05-20T00:52:49.700Z", "content": "

\"no

", "rawMarkdown": "![no points](https://pbs.twimg.com/media/EYbBAQCWAAE6I7u?format=png&name=900x900)", "votes": 17 }, { "id": 854309, "postDate": "2020-05-20T00:25:29.270Z", "content": "

\"\"

", "rawMarkdown": "![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2405813%2Fbfc64fb4c0de243ab58627a3d9284d08%2Fkaggle_4.jpg?generation=1589934326039195&alt=media)\n", "votes": 11 }, { "id": 854322, "postDate": "2020-05-20T00:42:51.310Z", "content": "

lol I wish there were more research contests like the one Francois Chollet made so that students or people who cannot afford the computational power can take part in interesting problems which doesn't just require throwing computational power at it

", "rawMarkdown": "lol I wish there were more research contests like the one Francois Chollet made so that students or people who cannot afford the computational power can take part in interesting problems which doesn't just require throwing computational power at it", "votes": 8, "replies": [ { "id": 854327, "postDate": "2020-05-20T00:49:12.180Z", "content": "

Kaggle used to be all about tabular data problems. Most of those you could do on a modestly powered laptop. But then things changed. And kept changing.

", "rawMarkdown": "Kaggle used to be all about tabular data problems. Most of those you could do on a modestly powered laptop. But then things changed. And kept changing. ", "votes": 6 }, { "id": 854414, "postDate": "2020-05-20T02:34:47.093Z", "rawMarkdown": "", "votes": -3, "isDeleted": true }, { "id": 855782, "postDate": "2020-05-21T07:49:52.077Z", "content": "

@jesudasdsouza Please win this competition

", "rawMarkdown": "@jesudasdsouza Please win this competition", "votes": 1 }, { "id": 855806, "postDate": "2020-05-21T08:33:23.447Z", "content": "

No. Too much effort for too less reward.

", "rawMarkdown": "No. Too much effort for too less reward. ", "votes": 1 }, { "id": 855817, "postDate": "2020-05-21T08:45:20.907Z", "content": "

separate GPU quota could have been a sensible move

", "rawMarkdown": "separate GPU quota could have been a sensible move", "votes": 1 } ] }, { "id": 854266, "postDate": "2020-05-19T23:12:41.893Z", "content": "

For real though because competitions that have extravagant requirements should at least have some sort of reward (the bare minimum points or medals) in order to incentivize data scientists taking on such a large task.

", "rawMarkdown": "For real though because competitions that have extravagant requirements should at least have some sort of reward (the bare minimum points or medals) in order to incentivize data scientists taking on such a large task.", "votes": 6, "replies": [ { "id": 854300, "postDate": "2020-05-20T00:15:04.837Z", "content": "

I think they got burned badly last year with some highly notorious cheaters, so they decided not to give any points or medals this year. Bad move though - the competitions will likely be ghost towns.

", "rawMarkdown": "I think they got burned badly last year with some highly notorious cheaters, so they decided not to give any points or medals this year. Bad move though - the competitions will likely be ghost towns. ", "votes": 3 }, { "id": 855363, "postDate": "2020-05-20T21:17:01.760Z", "content": "

so that's the reason ha??

", "rawMarkdown": "so that's the reason ha?? ", "votes": 2 }, { "id": 855372, "postDate": "2020-05-20T21:37:24.947Z", "content": "

Yes, that's my strong suspicion.

", "rawMarkdown": "Yes, that's my strong suspicion. ", "votes": 1 }, { "id": 855742, "postDate": "2020-05-21T06:58:34.103Z", "content": "

makes sense Bojan

", "rawMarkdown": "makes sense Bojan", "votes": 1 }, { "id": 855813, "postDate": "2020-05-21T08:42:22.790Z", "content": "

https://www.theregister.co.uk/2020/01/21/ai_kaggle_contest_cheat/ The rank 1 team did web-scraping, leading to a big difference between rank 1 score and rank 2 score

", "rawMarkdown": "https://www.theregister.co.uk/2020/01/21/ai_kaggle_contest_cheat/ The rank 1 team did web-scraping, leading to a big difference between rank 1 score and rank 2 score" } ] }, { "id": 854639, "postDate": "2020-05-20T07:03:38.550Z", "content": "

I don't understand the point of this... they are killing the research competitions. \nExample: Imet2019 = 521 teams ... Imet2020 (no prizes/medals) = 80 teams.

\n\n

People is free to join or not depending on how well organized is the comp, the timing, the topic and imo one of the most important factors are the prizes. An easy way to encourage/motivate top kagglers who can provide such solutions (and will do in many cases) is to offer great prizes/points. I think the main point, the reason why kaggle works is because a lot of people compete that's the point imo. Motivate people to join, get interested in the subject and compete… and obviously have fun!

\n\n

Maybe this new policy is due to last year cheating scandals...

", "rawMarkdown": "I don't understand the point of this... they are killing the research competitions. \nExample: Imet2019 = **521** teams ... Imet2020 (no prizes/medals) = **80** teams.\n\nPeople is free to join or not depending on how well organized is the comp, the timing, the topic and imo one of the most important factors are the prizes. An easy way to encourage/motivate top kagglers who can provide such solutions (and will do in many cases) is to offer great prizes/points. I think the main point, the reason why kaggle works is because a lot of people compete that's the point imo. Motivate people to join, get interested in the subject and compete… and obviously have fun!\n\nMaybe this new policy is due to last year cheating scandals...", "votes": 3 }, { "id": 854632, "postDate": "2020-05-20T06:58:15.137Z", "content": "

Forgive me if I am wrong but there is a prize for these challenges:\nlink

\n\n
\n

1st Place: $1200\n 2nd Place: $600

\n
\n\n

That being said, I don't see why they should not award Kaggle points for these competitions.

", "rawMarkdown": "Forgive me if I am wrong but there is a prize for these challenges:\n[link](http://www.robustvision.net/)\n> 1st Place: $1200\n> 2nd Place: $600\n\nThat being said, I don't see why they should not award Kaggle points for these competitions. ", "votes": 2 }, { "id": 854379, "postDate": "2020-05-20T02:04:10.440Z", "content": "

I was so looking forward to these... shame.

", "rawMarkdown": "I was so looking forward to these... shame.", "votes": 1 }, { "id": 854601, "postDate": "2020-05-20T06:21:32.060Z", "rawMarkdown": "", "isDeleted": true } ], "comments": [ { "id": 854328, "author_name": "Bojan Tunguz", "author_url": "", "post_date": "2020-05-20T00:52:49.700000", "content": "

\"no

", "votes": 17, "replies": [] }, { "id": 854309, "author_name": "Konstantin Yakovlev", "author_url": "", "post_date": "2020-05-20T00:25:29.270000", "content": "

\"\"

", "votes": 11, "replies": [] }, { "id": 854322, "author_name": "Sri.", "author_url": "", "post_date": "2020-05-20T00:42:51.310000", "content": "

lol I wish there were more research contests like the one Francois Chollet made so that students or people who cannot afford the computational power can take part in interesting problems which doesn't just require throwing computational power at it

", "votes": 8, "replies": [ { "id": 854327, "author_name": "Bojan Tunguz", "author_url": "", "post_date": "2020-05-20T00:49:12.180000", "content": "

Kaggle used to be all about tabular data problems. Most of those you could do on a modestly powered laptop. But then things changed. And kept changing.

", "votes": 6, "replies": [] }, { "id": 854414, "author_name": "", "author_url": "", "post_date": "2020-05-20T02:34:47.093000", "content": "", "votes": -3, "replies": [] }, { "id": 855782, "author_name": "Manoj", "author_url": "", "post_date": "2020-05-21T07:49:52.077000", "content": "

@jesudasdsouza Please win this competition

", "votes": 1, "replies": [] }, { "id": 855806, "author_name": "Jesudas DSouza", "author_url": "", "post_date": "2020-05-21T08:33:23.447000", "content": "

No. Too much effort for too less reward.

", "votes": 1, "replies": [] }, { "id": 855817, "author_name": "Manoj", "author_url": "", "post_date": "2020-05-21T08:45:20.907000", "content": "

separate GPU quota could have been a sensible move

", "votes": 1, "replies": [] } ] }, { "id": 854266, "author_name": "", "author_url": "", "post_date": "2020-05-19T23:12:41.893000", "content": "

For real though because competitions that have extravagant requirements should at least have some sort of reward (the bare minimum points or medals) in order to incentivize data scientists taking on such a large task.

", "votes": 6, "replies": [ { "id": 854300, "author_name": "Bojan Tunguz", "author_url": "", "post_date": "2020-05-20T00:15:04.837000", "content": "

I think they got burned badly last year with some highly notorious cheaters, so they decided not to give any points or medals this year. Bad move though - the competitions will likely be ghost towns.

", "votes": 3, "replies": [] }, { "id": 855363, "author_name": "Firat Gonen", "author_url": "", "post_date": "2020-05-20T21:17:01.760000", "content": "

so that's the reason ha??

", "votes": 2, "replies": [] }, { "id": 855372, "author_name": "Bojan Tunguz", "author_url": "", "post_date": "2020-05-20T21:37:24.947000", "content": "

Yes, that's my strong suspicion.

", "votes": 1, "replies": [] }, { "id": 855742, "author_name": "Firat Gonen", "author_url": "", "post_date": "2020-05-21T06:58:34.103000", "content": "

makes sense Bojan

", "votes": 1, "replies": [] }, { "id": 855813, "author_name": "Jesudas DSouza", "author_url": "", "post_date": "2020-05-21T08:42:22.790000", "content": "

https://www.theregister.co.uk/2020/01/21/ai_kaggle_contest_cheat/ The rank 1 team did web-scraping, leading to a big difference between rank 1 score and rank 2 score

", "votes": 0, "replies": [] } ] }, { "id": 854639, "author_name": "Nanashi", "author_url": "", "post_date": "2020-05-20T07:03:38.550000", "content": "

I don't understand the point of this... they are killing the research competitions. \nExample: Imet2019 = 521 teams ... Imet2020 (no prizes/medals) = 80 teams.

\n\n

People is free to join or not depending on how well organized is the comp, the timing, the topic and imo one of the most important factors are the prizes. An easy way to encourage/motivate top kagglers who can provide such solutions (and will do in many cases) is to offer great prizes/points. I think the main point, the reason why kaggle works is because a lot of people compete that's the point imo. Motivate people to join, get interested in the subject and compete… and obviously have fun!

\n\n

Maybe this new policy is due to last year cheating scandals...

", "votes": 3, "replies": [] }, { "id": 854632, "author_name": "Utsav Nandi", "author_url": "", "post_date": "2020-05-20T06:58:15.137000", "content": "

Forgive me if I am wrong but there is a prize for these challenges:\nlink

\n\n
\n

1st Place: $1200\n 2nd Place: $600

\n
\n\n

That being said, I don't see why they should not award Kaggle points for these competitions.

", "votes": 2, "replies": [] }, { "id": 854379, "author_name": "Trigram", "author_url": "", "post_date": "2020-05-20T02:04:10.440000", "content": "

I was so looking forward to these... shame.

", "votes": 1, "replies": [] }, { "id": 854601, "author_name": "", "author_url": "", "post_date": "2020-05-20T06:21:32.060000", "content": "", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "854238": "Two new image competitions, with huge datasets, requiring a boatload of computational power, and **NO** points or medals. Yeah, these are going to be enormously successful. 🙄 ", "854328": "![no points](https://pbs.twimg.com/media/EYbBAQCWAAE6I7u?format=png&name=900x900)", "854309": "![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2405813%2Fbfc64fb4c0de243ab58627a3d9284d08%2Fkaggle_4.jpg?generation=1589934326039195&alt=media)\n", "854322": "lol I wish there were more research contests like the one Francois Chollet made so that students or people who cannot afford the computational power can take part in interesting problems which doesn't just require throwing computational power at it", "854266": "For real though because competitions that have extravagant requirements should at least have some sort of reward (the bare minimum points or medals) in order to incentivize data scientists taking on such a large task.", "854639": "I don't understand the point of this... they are killing the research competitions. \nExample: Imet2019 = **521** teams ... Imet2020 (no prizes/medals) = **80** teams.\n\nPeople is free to join or not depending on how well organized is the comp, the timing, the topic and imo one of the most important factors are the prizes. An easy way to encourage/motivate top kagglers who can provide such solutions (and will do in many cases) is to offer great prizes/points. I think the main point, the reason why kaggle works is because a lot of people compete that's the point imo. Motivate people to join, get interested in the subject and compete… and obviously have fun!\n\nMaybe this new policy is due to last year cheating scandals...", "854632": "Forgive me if I am wrong but there is a prize for these challenges:\n[link](http://www.robustvision.net/)\n> 1st Place: $1200\n> 2nd Place: $600\n\nThat being said, I don't see why they should not award Kaggle points for these competitions. ", "854379": "I was so looking forward to these... shame.", "854601": "" } }