{ "id": 251810, "title": "[introduction] New YOLO family \"YOLOR\"", "url": "/competitions/siim-covid19-detection/discussion/251810", "author_name": "", "post_date": "2021-07-08T23:18:09.387997100Z", "votes": 7, "comment_count": 7, "views": 0, "content": "

(This is just for introduction.)
\nNew YOLO family YOLOR, You Only Learn One Representation, has come.

\n

In this article, How to Train YOLOR on a Custom Dataset, you can learn step by step(also offering Google Colaboratory notebook);

\n

I haven't checked the details yet, but I wanted to post this information to share.
\nI often use YOLOv5 for object detection, but if YOLOR is easier to use and has better performance, I would like to replace.

\n

Thanks😊

\n

The license of yolor is GNU General Public License v3.0.
\nIf you are worried, please check the competition regulations.

\n

\"img\"
\nYOLOR leverages multi-task implicit knowledge along with task-specific explicit knowledge(from this article)

\n

\"img\"
\nYOLOR models sets the state of the art in object detection in terms of speed and accuracy(from this article)

", "messages": [ { "id": "1381374", "postDate": "07/08/2021 23:18:09", "content": "

(This is just for introduction.)
\nNew YOLO family YOLOR, You Only Learn One Representation, has come.

\n

In this article, How to Train YOLOR on a Custom Dataset, you can learn step by step(also offering Google Colaboratory notebook);

\n

I haven't checked the details yet, but I wanted to post this information to share.
\nI often use YOLOv5 for object detection, but if YOLOR is easier to use and has better performance, I would like to replace.

\n

Thanks😊

\n

The license of yolor is GNU General Public License v3.0.
\nIf you are worried, please check the competition regulations.

\n

\"img\"
\nYOLOR leverages multi-task implicit knowledge along with task-specific explicit knowledge(from this article)

\n

\"img\"
\nYOLOR models sets the state of the art in object detection in terms of speed and accuracy(from this article)

", "rawMarkdown": "(*This is just for introduction.*)\nNew YOLO family **YOLOR**, *You Only Learn One Representation*, has come.\n\nIn this [article](https://blog.roboflow.com/train-yolor-on-a-custom-dataset/), *How to Train YOLOR on a Custom Dataset*, you can learn step by step(also offering Google Colaboratory notebook);\n\n\nI haven't checked the details yet, but I wanted to post this information to share.\nI often use YOLOv5 for object detection, but if YOLOR is easier to use and has better performance, I would like to replace.\n\nThanks😊\n\n*The license of yolor is GNU General Public License v3.0.*\n*If you are worried, please check the competition regulations.*\n\n![img](https://blog.roboflow.com/content/images/size/w1600/2021/07/unifued_network-1-.png)\n*YOLOR leverages multi-task implicit knowledge along with task-specific explicit knowledge(from this [article](https://blog.roboflow.com/train-yolor-on-a-custom-dataset/))*\n\n![img](https://blog.roboflow.com/content/images/2021/07/performance-1-.png)\n*YOLOR models sets the state of the art in object detection in terms of speed and accuracy(from this [article](https://blog.roboflow.com/train-yolor-on-a-custom-dataset/))*", "votes": null }, { "id": "1381545", "postDate": "07/09/2021 05:07:42", "content": "

Thanks for sharing!

", "rawMarkdown": "Thanks for sharing!", "votes": null }, { "id": "1382738", "postDate": "07/10/2021 07:38:42", "content": "

Thanks for sharing!

\n

Have you applied YOLOR in this competition? Is it really derive better score than YOLOv5?

", "rawMarkdown": "Thanks for sharing!\n\nHave you applied YOLOR in this competition? Is it really derive better score than YOLOv5?", "votes": null }, { "id": "1383056", "postDate": "07/10/2021 13:19:41", "content": "

I am about to ask question, I am currently fine tuning YOLO model. Keen to know if YOLOR is really going to be better.

", "rawMarkdown": "I am about to ask question, I am currently fine tuning YOLO model. Keen to know if YOLOR is really going to be better.", "votes": null }, { "id": "1386441", "postDate": "07/13/2021 13:41:05", "content": "

Thank you☺️

\n

Sorry but I've just recently found YOLOR😅
\nAt first, I would like to implement it and then compare it with the result of YOLOv5!

", "rawMarkdown": "Thank you☺️\n\nSorry but I've just recently found YOLOR😅\nAt first, I would like to implement it and then compare it with the result of YOLOv5!", "votes": null }, { "id": "1386444", "postDate": "07/13/2021 13:42:44", "content": "

Thank you☺️

\n

No, not yet😅
\nI've just recently found YOLOR.
\nBut I want to compare the result with the one of YOLOv5!

", "rawMarkdown": "Thank you☺️\n\nNo, not yet😅\nI've just recently found YOLOR.\nBut I want to compare the result with the one of YOLOv5!", "votes": null }, { "id": "1386618", "postDate": "07/13/2021 15:25:39", "content": "

We tried it but ti gives not better results

", "rawMarkdown": "We tried it but ti gives not better results", "votes": null }, { "id": "1387291", "postDate": "07/14/2021 04:33:44", "content": "

Thanks for sharing @yhirakawa ! Worth a try definitely!

", "rawMarkdown": "Thanks for sharing @yhirakawa ! Worth a try definitely!", "votes": null } ], "comments": [ { "id": 1381545, "author_name": "minhtien1405", "author_url": "", "post_date": "07/09/2021 05:07:42", "content": "

Thanks for sharing!

", "votes": null, "replies": [] }, { "id": 1382738, "author_name": "baekseungyun", "author_url": "", "post_date": "07/10/2021 07:38:42", "content": "

Thanks for sharing!

\n

Have you applied YOLOR in this competition? Is it really derive better score than YOLOv5?

", "votes": null, "replies": [ { "id": 1386444, "author_name": "yhirakawa", "author_url": "", "post_date": "07/13/2021 13:42:44", "content": "

Thank you☺️

\n

No, not yet😅
\nI've just recently found YOLOR.
\nBut I want to compare the result with the one of YOLOv5!

", "votes": null, "replies": [] } ] }, { "id": 1383056, "author_name": "hawkeat", "author_url": "", "post_date": "07/10/2021 13:19:41", "content": "

I am about to ask question, I am currently fine tuning YOLO model. Keen to know if YOLOR is really going to be better.

", "votes": null, "replies": [ { "id": 1386441, "author_name": "yhirakawa", "author_url": "", "post_date": "07/13/2021 13:41:05", "content": "

Thank you☺️

\n

Sorry but I've just recently found YOLOR😅
\nAt first, I would like to implement it and then compare it with the result of YOLOv5!

", "votes": null, "replies": [] } ] }, { "id": 1386618, "author_name": "jaideepvalani", "author_url": "", "post_date": "07/13/2021 15:25:39", "content": "

We tried it but ti gives not better results

", "votes": null, "replies": [] }, { "id": 1387291, "author_name": "saurabhbagchi", "author_url": "", "post_date": "07/14/2021 04:33:44", "content": "

Thanks for sharing @yhirakawa ! Worth a try definitely!

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1381374": "(*This is just for introduction.*)\nNew YOLO family **YOLOR**, *You Only Learn One Representation*, has come.\n\nIn this [article](https://blog.roboflow.com/train-yolor-on-a-custom-dataset/), *How to Train YOLOR on a Custom Dataset*, you can learn step by step(also offering Google Colaboratory notebook);\n\n\nI haven't checked the details yet, but I wanted to post this information to share.\nI often use YOLOv5 for object detection, but if YOLOR is easier to use and has better performance, I would like to replace.\n\nThanks😊\n\n*The license of yolor is GNU General Public License v3.0.*\n*If you are worried, please check the competition regulations.*\n\n![img](https://blog.roboflow.com/content/images/size/w1600/2021/07/unifued_network-1-.png)\n*YOLOR leverages multi-task implicit knowledge along with task-specific explicit knowledge(from this [article](https://blog.roboflow.com/train-yolor-on-a-custom-dataset/))*\n\n![img](https://blog.roboflow.com/content/images/2021/07/performance-1-.png)\n*YOLOR models sets the state of the art in object detection in terms of speed and accuracy(from this [article](https://blog.roboflow.com/train-yolor-on-a-custom-dataset/))*", "1381545": "Thanks for sharing!", "1382738": "Thanks for sharing!\n\nHave you applied YOLOR in this competition? Is it really derive better score than YOLOv5?", "1383056": "I am about to ask question, I am currently fine tuning YOLO model. Keen to know if YOLOR is really going to be better.", "1386441": "Thank you☺️\n\nSorry but I've just recently found YOLOR😅\nAt first, I would like to implement it and then compare it with the result of YOLOv5!", "1386444": "Thank you☺️\n\nNo, not yet😅\nI've just recently found YOLOR.\nBut I want to compare the result with the one of YOLOv5!", "1386618": "We tried it but ti gives not better results", "1387291": "Thanks for sharing @yhirakawa ! Worth a try definitely!" }, "source": "meta" }