{ "id": 245523, "title": "Any Update here?", "url": "/competitions/siim-covid19-detection/discussion/245523", "author_name": "", "post_date": "2021-06-11T07:48:13.619442500Z", "votes": 31, "comment_count": 12, "views": 0, "content": "

Weekends are coming and there is still no updating info.\"u_4277415444,176027722_fm_26_gp_0.jpg\"

", "messages": [ { "id": "1344915", "postDate": "06/11/2021 07:48:13", "content": "

Weekends are coming and there is still no updating info.\"u_4277415444,176027722_fm_26_gp_0.jpg\"

", "rawMarkdown": "Weekends are coming and there is still no updating info.![u_4277415444,176027722_fm_26_gp_0.jpg](https://i.loli.net/2021/06/11/AToCMiSu51jZs2K.jpg)", "votes": null }, { "id": "1344933", "postDate": "06/11/2021 08:00:06", "content": "

They maybe think it’s useless because scientists aren’t scientists : https://blogs.sciencemag.org/pipeline/archives/2021/06/02/machine-learning-deserves-better-than-this

", "rawMarkdown": "They maybe think it’s useless because scientists aren’t scientists : https://blogs.sciencemag.org/pipeline/archives/2021/06/02/machine-learning-deserves-better-than-this", "votes": null }, { "id": "1345116", "postDate": "06/11/2021 10:22:50", "content": "

Really hope this is not one of the case of \"garbage in\". It seems to be the normality in ML and covid chest X-Ray apparently : https://lukeoakdenrayner.wordpress.com/2020/03/23/ct-scanning-is-just-awful-for-diagnosing-covid-19/ ….

", "rawMarkdown": "Really hope this is not one of the case of \"garbage in\". It seems to be the normality in ML and covid chest X-Ray apparently : https://lukeoakdenrayner.wordpress.com/2020/03/23/ct-scanning-is-just-awful-for-diagnosing-covid-19/ ....", "votes": null }, { "id": "1345170", "postDate": "06/11/2021 11:19:06", "content": "

Lottery game again.

", "rawMarkdown": "Lottery game again.", "votes": null }, { "id": "1345358", "postDate": "06/11/2021 14:10:09", "content": "

Don't panic.
\n\"\"

", "rawMarkdown": "Don't panic.\n![](https://uploda1.ysklog.net/uploda/024bfd4d7b.jpg)", "votes": null }, { "id": "1348072", "postDate": "06/13/2021 17:52:30", "content": "

If there is a significant change in data, an extension of the date should be given imo

", "rawMarkdown": "If there is a significant change in data, an extension of the date should be given imo", "votes": null }, { "id": "1348107", "postDate": "06/13/2021 18:42:50", "content": "

Thanks for sharing that blog post. Its very unfortunate that people these days care mostly about publishing every other day to be at the top of the \"AI race\", especially big companies. I wished more people focused on data and actual real world impact rather than the miniscule improvements over ImageNet. IMO Kaggle competitions in that sense are closer to have a real world impact than most of the papers thanks to people competing and validating the problem/data. This reminds me of https://arxiv.org/abs/1206.4656 which is a paper from 2012 but I guess not much changed over almost a decade.

", "rawMarkdown": "Thanks for sharing that blog post. Its very unfortunate that people these days care mostly about publishing every other day to be at the top of the \"AI race\", especially big companies. I wished more people focused on data and actual real world impact rather than the miniscule improvements over ImageNet. IMO Kaggle competitions in that sense are closer to have a real world impact than most of the papers thanks to people competing and validating the problem/data. This reminds me of https://arxiv.org/abs/1206.4656 which is a paper from 2012 but I guess not much changed over almost a decade.", "votes": null }, { "id": "1348535", "postDate": "06/14/2021 05:48:33", "content": "

\"\"

", "rawMarkdown": "![](https://i.imgflip.com/5d6lqh.jpg)", "votes": null }, { "id": "1349695", "postDate": "06/15/2021 03:10:15", "content": "

Metric seems updated. But duplicate images still exist.

", "rawMarkdown": "Metric seems updated. But duplicate images still exist.", "votes": null }, { "id": "1349698", "postDate": "06/15/2021 03:15:24", "content": "

I think they are not exact dups, consider them as free augmentations :)

", "rawMarkdown": "I think they are not exact dups, consider them as free augmentations :)", "votes": null }, { "id": "1349700", "postDate": "06/15/2021 03:23:57", "content": "

Good idea. I will consider it.

", "rawMarkdown": "Good idea. I will consider it.", "votes": null }, { "id": "1350670", "postDate": "06/15/2021 16:28:16", "content": "

Updates are underway. Refer to this post: https://www.kaggle.com/c/siim-covid19-detection/discussion/246296

", "rawMarkdown": "Updates are underway. Refer to this post: https://www.kaggle.com/c/siim-covid19-detection/discussion/246296", "votes": null }, { "id": "1354129", "postDate": "06/17/2021 11:54:34", "content": "

Lol. One of the reason I love reading discussions are data science memes

", "rawMarkdown": "Lol. One of the reason I love reading discussions are **data science memes**", "votes": null } ], "comments": [ { "id": 1344933, "author_name": "josephamigo", "author_url": "", "post_date": "06/11/2021 08:00:06", "content": "

They maybe think it’s useless because scientists aren’t scientists : https://blogs.sciencemag.org/pipeline/archives/2021/06/02/machine-learning-deserves-better-than-this

", "votes": null, "replies": [ { "id": 1345116, "author_name": "josephamigo", "author_url": "", "post_date": "06/11/2021 10:22:50", "content": "

Really hope this is not one of the case of \"garbage in\". It seems to be the normality in ML and covid chest X-Ray apparently : https://lukeoakdenrayner.wordpress.com/2020/03/23/ct-scanning-is-just-awful-for-diagnosing-covid-19/ ….

", "votes": null, "replies": [] }, { "id": 1345170, "author_name": "southsakura", "author_url": "", "post_date": "06/11/2021 11:19:06", "content": "

Lottery game again.

", "votes": null, "replies": [] }, { "id": 1348107, "author_name": "keremt", "author_url": "", "post_date": "06/13/2021 18:42:50", "content": "

Thanks for sharing that blog post. Its very unfortunate that people these days care mostly about publishing every other day to be at the top of the \"AI race\", especially big companies. I wished more people focused on data and actual real world impact rather than the miniscule improvements over ImageNet. IMO Kaggle competitions in that sense are closer to have a real world impact than most of the papers thanks to people competing and validating the problem/data. This reminds me of https://arxiv.org/abs/1206.4656 which is a paper from 2012 but I guess not much changed over almost a decade.

", "votes": null, "replies": [] } ] }, { "id": 1345358, "author_name": "yujiariyasu", "author_url": "", "post_date": "06/11/2021 14:10:09", "content": "

Don't panic.
\n\"\"

", "votes": null, "replies": [ { "id": 1348535, "author_name": "kyoshioka47", "author_url": "", "post_date": "06/14/2021 05:48:33", "content": "

\"\"

", "votes": null, "replies": [] }, { "id": 1354129, "author_name": "shaz13", "author_url": "", "post_date": "06/17/2021 11:54:34", "content": "

Lol. One of the reason I love reading discussions are data science memes

", "votes": null, "replies": [] } ] }, { "id": 1348072, "author_name": "mrinath", "author_url": "", "post_date": "06/13/2021 17:52:30", "content": "

If there is a significant change in data, an extension of the date should be given imo

", "votes": null, "replies": [] }, { "id": 1349695, "author_name": "southsakura", "author_url": "", "post_date": "06/15/2021 03:10:15", "content": "

Metric seems updated. But duplicate images still exist.

", "votes": null, "replies": [ { "id": 1349698, "author_name": "keremt", "author_url": "", "post_date": "06/15/2021 03:15:24", "content": "

I think they are not exact dups, consider them as free augmentations :)

", "votes": null, "replies": [] }, { "id": 1349700, "author_name": "southsakura", "author_url": "", "post_date": "06/15/2021 03:23:57", "content": "

Good idea. I will consider it.

", "votes": null, "replies": [] } ] }, { "id": 1350670, "author_name": "juliaelliott", "author_url": "", "post_date": "06/15/2021 16:28:16", "content": "

Updates are underway. Refer to this post: https://www.kaggle.com/c/siim-covid19-detection/discussion/246296

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1344915": "Weekends are coming and there is still no updating info.![u_4277415444,176027722_fm_26_gp_0.jpg](https://i.loli.net/2021/06/11/AToCMiSu51jZs2K.jpg)", "1344933": "They maybe think it’s useless because scientists aren’t scientists : https://blogs.sciencemag.org/pipeline/archives/2021/06/02/machine-learning-deserves-better-than-this", "1345116": "Really hope this is not one of the case of \"garbage in\". It seems to be the normality in ML and covid chest X-Ray apparently : https://lukeoakdenrayner.wordpress.com/2020/03/23/ct-scanning-is-just-awful-for-diagnosing-covid-19/ ....", "1345170": "Lottery game again.", "1345358": "Don't panic.\n![](https://uploda1.ysklog.net/uploda/024bfd4d7b.jpg)", "1348072": "If there is a significant change in data, an extension of the date should be given imo", "1348107": "Thanks for sharing that blog post. Its very unfortunate that people these days care mostly about publishing every other day to be at the top of the \"AI race\", especially big companies. I wished more people focused on data and actual real world impact rather than the miniscule improvements over ImageNet. IMO Kaggle competitions in that sense are closer to have a real world impact than most of the papers thanks to people competing and validating the problem/data. This reminds me of https://arxiv.org/abs/1206.4656 which is a paper from 2012 but I guess not much changed over almost a decade.", "1348535": "![](https://i.imgflip.com/5d6lqh.jpg)", "1349695": "Metric seems updated. But duplicate images still exist.", "1349698": "I think they are not exact dups, consider them as free augmentations :)", "1349700": "Good idea. I will consider it.", "1350670": "Updates are underway. Refer to this post: https://www.kaggle.com/c/siim-covid19-detection/discussion/246296", "1354129": "Lol. One of the reason I love reading discussions are **data science memes**" }, "source": "meta" }