{ "id": 271639, "title": "Final LB prediction", "url": "/competitions/g2net-gravitational-wave-detection/discussion/271639", "author_name": "", "post_date": "2021-09-11T17:13:55.907961300Z", "votes": 4, "comment_count": 6, "views": 0, "content": "

This is an exciting competition, and I'm working hard on it as others.
\nFor now the top team has reached 0.8840. I just wanna to know how much the final LB will be.
\n0.89xx or 0.90xx? How do you think about it?

", "messages": [ { "id": "1509772", "postDate": "09/11/2021 17:13:55", "content": "

This is an exciting competition, and I'm working hard on it as others.
\nFor now the top team has reached 0.8840. I just wanna to know how much the final LB will be.
\n0.89xx or 0.90xx? How do you think about it?

", "rawMarkdown": "This is an exciting competition, and I'm working hard on it as others.\nFor now the top team has reached 0.8840. I just wanna to know how much the final LB will be.\n0.89xx or 0.90xx? How do you think about it?", "votes": null }, { "id": "1509796", "postDate": "09/11/2021 17:50:55", "content": "

>= 0.8840 :D

", "rawMarkdown": "`>= 0.8840` :D", "votes": null }, { "id": "1509829", "postDate": "09/11/2021 18:16:04", "content": "

@wuliaokaola - The teams in and just below the lower medal ranges have increased their scores a lot in the last few days, much of this being a result of a flood of higher scoring notebooks being posted. That is, people below the silver zone now have easy access to ~0.8765 models rather than ~0.8665 ones a week or so ago. I think the top ranking teams may find it less easy to improve by as much unless someone comes up with some new \"magic\". A rationalisation is that the stronger GW signals are relatively easy to find, but the harder ones are very difficult indeed.

", "rawMarkdown": "wuliaokaola - The teams in and just below the lower medal ranges have increased their scores a lot in the last few days, much of this being a result of a flood of higher scoring notebooks being posted. That is, people below the silver zone now have easy access to ~0.8765 models rather than ~0.8665 ones a week or so ago. I think the top ranking teams may find it less easy to improve by as much unless someone comes up with some new \"magic\". A rationalisation is that the stronger GW signals are relatively easy to find, but the harder ones are very difficult indeed.", "votes": null }, { "id": "1509872", "postDate": "09/11/2021 18:36:00", "content": "

I am always surprised by how good top teams can get. I hope we won't be too far from them…

", "rawMarkdown": "I am always surprised by how good top teams can get. I hope we won't be too far from them...", "votes": null }, { "id": "1509873", "postDate": "09/11/2021 18:36:38", "content": "

the limit is your training AUC. (if you achieve zero generalisation gap)

", "rawMarkdown": "the limit is your training AUC. (if you achieve zero generalisation gap)", "votes": null }, { "id": "1509970", "postDate": "09/11/2021 20:33:40", "content": "

I would bet around 0.8890.

", "rawMarkdown": "I would bet around 0.8890.", "votes": null }, { "id": "1559700", "postDate": "10/27/2021 07:07:33", "content": "

Hey All,

\n

Thank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.

\n

Thank you all,

\n

Regards,
\nChris

", "rawMarkdown": "Hey All,\n\nThank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.\n\nThank you all,\n\nRegards,\nChris", "votes": null } ], "comments": [ { "id": 1509796, "author_name": "cpmpml", "author_url": "", "post_date": "09/11/2021 17:50:55", "content": "

>= 0.8840 :D

", "votes": null, "replies": [] }, { "id": 1509829, "author_name": "jbomitchell", "author_url": "", "post_date": "09/11/2021 18:16:04", "content": "

@wuliaokaola - The teams in and just below the lower medal ranges have increased their scores a lot in the last few days, much of this being a result of a flood of higher scoring notebooks being posted. That is, people below the silver zone now have easy access to ~0.8765 models rather than ~0.8665 ones a week or so ago. I think the top ranking teams may find it less easy to improve by as much unless someone comes up with some new \"magic\". A rationalisation is that the stronger GW signals are relatively easy to find, but the harder ones are very difficult indeed.

", "votes": null, "replies": [ { "id": 1509872, "author_name": "cpmpml", "author_url": "", "post_date": "09/11/2021 18:36:00", "content": "

I am always surprised by how good top teams can get. I hope we won't be too far from them…

", "votes": null, "replies": [] } ] }, { "id": 1509873, "author_name": "hengck23", "author_url": "", "post_date": "09/11/2021 18:36:38", "content": "

the limit is your training AUC. (if you achieve zero generalisation gap)

", "votes": null, "replies": [] }, { "id": 1509970, "author_name": "yassinealouini", "author_url": "", "post_date": "09/11/2021 20:33:40", "content": "

I would bet around 0.8890.

", "votes": null, "replies": [] }, { "id": 1559700, "author_name": "zerafachris", "author_url": "", "post_date": "10/27/2021 07:07:33", "content": "

Hey All,

\n

Thank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.

\n

Thank you all,

\n

Regards,
\nChris

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1509772": "This is an exciting competition, and I'm working hard on it as others.\nFor now the top team has reached 0.8840. I just wanna to know how much the final LB will be.\n0.89xx or 0.90xx? How do you think about it?", "1509796": "`>= 0.8840` :D", "1509829": "wuliaokaola - The teams in and just below the lower medal ranges have increased their scores a lot in the last few days, much of this being a result of a flood of higher scoring notebooks being posted. That is, people below the silver zone now have easy access to ~0.8765 models rather than ~0.8665 ones a week or so ago. I think the top ranking teams may find it less easy to improve by as much unless someone comes up with some new \"magic\". A rationalisation is that the stronger GW signals are relatively easy to find, but the harder ones are very difficult indeed.", "1509872": "I am always surprised by how good top teams can get. I hope we won't be too far from them...", "1509873": "the limit is your training AUC. (if you achieve zero generalisation gap)", "1509970": "I would bet around 0.8890.", "1559700": "Hey All,\n\nThank you all for taking part in our competition. The participation has been overwhelmingly positive. We are currently conducting a survey to gauge the demographic and outreach achieved. Kindly spare 2min and fill in this survey https://forms.gle/QP9L16niPexozyhu5.\n\nThank you all,\n\nRegards,\nChris" }, "source": "meta" }