Are YOLO Models allowed for this competition?
", "messages": [ { "id": 3241833, "postDate": "2025-07-05T08:53:17.433Z", "content": "https://github.com/ultralytics/ultralytics
", "rawMarkdown": "https://github.com/ultralytics/ultralytics", "votes": 1 }, { "id": 1132946, "postDate": "2020-12-30T20:04:24.327Z", "content": "Are YOLO Models allowed for this competition?
", "rawMarkdown": "Are YOLO Models allowed for this competition?", "votes": 12 }, { "id": 1150581, "postDate": "2021-01-12T17:25:32.247Z", "content": "Allowances for to use of external data and enforcement of winning models compliance is ultimately at the discretion of the host. As @andy2709 has clarified below, this host permits the use of YOLOv5.
", "rawMarkdown": "Allowances for to use of external data and enforcement of winning models compliance is ultimately at the discretion of the host. As @andy2709 has clarified below, this host permits the use of YOLOv5.", "votes": 5, "replies": [ { "id": 1150601, "postDate": "2021-01-12T17:40:23.450Z", "content": "Thank you for the clarification.
", "rawMarkdown": "Thank you for the clarification.", "votes": 1 }, { "id": 1150966, "postDate": "2021-01-13T02:35:57.617Z", "content": "@matthiasanderer Just to confirm what Julia said above, our team allows people to use YOLOv5 in their solutions
", "rawMarkdown": "@matthiasanderer Just to confirm what Julia said above, our team **allows** people to use YOLOv5 in their solutions", "votes": 2 }, { "id": 1150970, "postDate": "2021-01-13T02:54:34.157Z", "content": "Great, thank you for the information.
", "rawMarkdown": "Great, thank you for the information.", "votes": 1 }, { "id": 1151205, "postDate": "2021-01-13T07:28:09Z", "content": "Yeah, thank you for the clarification.
", "rawMarkdown": "Yeah, thank you for the clarification." } ] }, { "id": 1133151, "postDate": "2020-12-31T01:42:33.600Z", "content": "Yes, you can use YOLOv5 (https://github.com/ultralytics/yolov5)
", "rawMarkdown": "Yes, you can use YOLOv5 (https://github.com/ultralytics/yolov5)", "votes": 4, "replies": [ { "id": 1133189, "postDate": "2020-12-31T02:35:06.473Z", "content": "Currently, I'm using YOLOv4 (root version \"https://github.com/AlexeyAB/darknet\") for many tasks, so i have a question:
\nPersonally I prefer Yolov5 because it is implemented in PyTorch which is easier to handle and work with. You might look at this blog from roboflow which presents comparison of these two models in different aspects.
", "rawMarkdown": "Personally I prefer Yolov5 because it is implemented in PyTorch which is easier to handle and work with. You might look at this [blog from roboflow](https://blog.roboflow.com/yolov4-versus-yolov5/#:~:text=YOLOv5%20trains%20faster%20on%20the,optimized%20to%20achieve%20higher%20FPS.) which presents comparison of these two models in different aspects.", "votes": 9 }, { "id": 1142159, "postDate": "2021-01-07T07:48:09.320Z", "content": "@andy2709 Earlier Yolo was not allowed because of GPLv3 license.
\nHere is Kaggle's official reply in Global Wheat Detection competition where all Yolo submissions were disqualified.
\nI see the Rules sections [8.C] are similar for both competitions - so want to avoid Global Wheat Detection situation where participants had to re-write their code using a different model.
", "rawMarkdown": "@andy2709 Earlier Yolo was not allowed because of [GPLv3 license](https://github.com/ultralytics/yolov5/blob/master/LICENSE).\n\nHere is Kaggle's [official reply in Global Wheat Detection](https://www.kaggle.com/c/global-wheat-detection/discussion/167298) competition where all Yolo submissions were [**disqualified**](https://www.kaggle.com/c/global-wheat-detection/discussion/168353).\n\nI see the [Rules](https://www.kaggle.com/c/vinbigdata-chest-xray-abnormalities-detection/rules) sections [8.C] are similar for both competitions - so want to avoid Global Wheat Detection situation where participants **had to re-write their code** using a different model.", "votes": 3 }, { "id": 1220175, "postDate": "2021-02-27T17:13:22.450Z", "rawMarkdown": "", "isDeleted": true } ] }, { "id": 1149978, "postDate": "2021-01-12T09:19:36.180Z", "content": "@juliaelliott and @andy2709 : As stated by @sirishks there is a clear contradiction between the clearance of yolov5 in this thread and the competition rules section 8.c:
\nCould you please give a clear go for GPL3 license in this competition.
\nOtherwise a lot of people will now use yolov5 and will be very angry later on if you decide not to allow it :-)
", "rawMarkdown": "@juliaelliott and @andy2709 : As stated by @sirishks there is a clear contradiction between the clearance of yolov5 in this thread and the competition rules section 8.c:\n\nCould you please give a clear go for GPL3 license in this competition.\n\nOtherwise a lot of people will now use yolov5 and will be very angry later on if you decide not to allow it :-)", "votes": 1 }, { "id": 1133015, "postDate": "2020-12-30T21:22:05.147Z", "content": "I think the competition is among the classification problems. As indicated on the sample submission csv file, only the labels are required - no bounding box values.
", "rawMarkdown": "I think the competition is among the classification problems. As indicated on the sample submission csv file, only the labels are required - no bounding box values. ", "votes": 1, "replies": [ { "id": 1133027, "postDate": "2020-12-30T21:35:23.103Z", "content": "The training data is in bounding box format. detectors produces box probabilities. You can use max(box probability) in your sample_submission file
", "rawMarkdown": "The training data is in bounding box format. detectors produces box probabilities. You can use max(box probability) in your sample_submission file", "votes": 1 }, { "id": 1133045, "postDate": "2020-12-30T21:56:03.267Z", "content": "You are right - looked the ddescription and the files again. Yeah, the competition is all about object detection. YOLO series are the ideal candidates.
", "rawMarkdown": "You are right - looked the ddescription and the files again. Yeah, the competition is all about object detection. YOLO series are the ideal candidates. ", "votes": 2 } ] }, { "id": 1132962, "postDate": "2020-12-30T20:19:43.430Z", "content": "Of course! I will be using them for sure
", "rawMarkdown": "Of course! I will be using them for sure", "votes": 1 } ], "comments": [ { "id": 3241833, "author_name": "Banuprasad B", "author_url": "", "post_date": "2025-07-05T08:53:17.433000", "content": "https://github.com/ultralytics/ultralytics
", "votes": 1, "replies": [] }, { "id": 1150581, "author_name": "Julia Elliott", "author_url": "", "post_date": "2021-01-12T17:25:32.247000", "content": "Allowances for to use of external data and enforcement of winning models compliance is ultimately at the discretion of the host. As @andy2709 has clarified below, this host permits the use of YOLOv5.
", "votes": 5, "replies": [ { "id": 1150601, "author_name": "Matthias", "author_url": "", "post_date": "2021-01-12T17:40:23.450000", "content": "Thank you for the clarification.
", "votes": 1, "replies": [] }, { "id": 1150966, "author_name": "NguyenThanhNhan", "author_url": "", "post_date": "2021-01-13T02:35:57.617000", "content": "@matthiasanderer Just to confirm what Julia said above, our team allows people to use YOLOv5 in their solutions
", "votes": 2, "replies": [] }, { "id": 1150970, "author_name": "Mohammed Aliy", "author_url": "", "post_date": "2021-01-13T02:54:34.157000", "content": "Great, thank you for the information.
", "votes": 1, "replies": [] }, { "id": 1151205, "author_name": "Pythonian", "author_url": "", "post_date": "2021-01-13T07:28:09", "content": "Yeah, thank you for the clarification.
", "votes": 0, "replies": [] } ] }, { "id": 1133151, "author_name": "NguyenThanhNhan", "author_url": "", "post_date": "2020-12-31T01:42:33.600000", "content": "Yes, you can use YOLOv5 (https://github.com/ultralytics/yolov5)
", "votes": 4, "replies": [ { "id": 1133189, "author_name": "DUC NGUYEN HUYNH", "author_url": "", "post_date": "2020-12-31T02:35:06.473000", "content": "Currently, I'm using YOLOv4 (root version \"https://github.com/AlexeyAB/darknet\") for many tasks, so i have a question:
\nPersonally I prefer Yolov5 because it is implemented in PyTorch which is easier to handle and work with. You might look at this blog from roboflow which presents comparison of these two models in different aspects.
", "votes": 9, "replies": [] }, { "id": 1142159, "author_name": "Sirish Somanchi", "author_url": "", "post_date": "2021-01-07T07:48:09.320000", "content": "@andy2709 Earlier Yolo was not allowed because of GPLv3 license.
\nHere is Kaggle's official reply in Global Wheat Detection competition where all Yolo submissions were disqualified.
\nI see the Rules sections [8.C] are similar for both competitions - so want to avoid Global Wheat Detection situation where participants had to re-write their code using a different model.
", "votes": 3, "replies": [] }, { "id": 1220175, "author_name": "", "author_url": "", "post_date": "2021-02-27T17:13:22.450000", "content": "", "votes": 0, "replies": [] } ] }, { "id": 1149978, "author_name": "Matthias", "author_url": "", "post_date": "2021-01-12T09:19:36.180000", "content": "@juliaelliott and @andy2709 : As stated by @sirishks there is a clear contradiction between the clearance of yolov5 in this thread and the competition rules section 8.c:
\nCould you please give a clear go for GPL3 license in this competition.
\nOtherwise a lot of people will now use yolov5 and will be very angry later on if you decide not to allow it :-)
", "votes": 1, "replies": [] }, { "id": 1133015, "author_name": "Mohammed Aliy", "author_url": "", "post_date": "2020-12-30T21:22:05.147000", "content": "I think the competition is among the classification problems. As indicated on the sample submission csv file, only the labels are required - no bounding box values.
", "votes": 1, "replies": [ { "id": 1133027, "author_name": "raddar", "author_url": "", "post_date": "2020-12-30T21:35:23.103000", "content": "The training data is in bounding box format. detectors produces box probabilities. You can use max(box probability) in your sample_submission file
", "votes": 1, "replies": [] }, { "id": 1133045, "author_name": "Mohammed Aliy", "author_url": "", "post_date": "2020-12-30T21:56:03.267000", "content": "You are right - looked the ddescription and the files again. Yeah, the competition is all about object detection. YOLO series are the ideal candidates.
", "votes": 2, "replies": [] } ] }, { "id": 1132962, "author_name": "raddar", "author_url": "", "post_date": "2020-12-30T20:19:43.430000", "content": "Of course! I will be using them for sure
", "votes": 1, "replies": [] } ], "raw_markdown_by_id": { "3241833": "https://github.com/ultralytics/ultralytics", "1132946": "Are YOLO Models allowed for this competition?", "1150581": "Allowances for to use of external data and enforcement of winning models compliance is ultimately at the discretion of the host. As @andy2709 has clarified below, this host permits the use of YOLOv5.", "1133151": "Yes, you can use YOLOv5 (https://github.com/ultralytics/yolov5)", "1149978": "@juliaelliott and @andy2709 : As stated by @sirishks there is a clear contradiction between the clearance of yolov5 in this thread and the competition rules section 8.c:\n\nCould you please give a clear go for GPL3 license in this competition.\n\nOtherwise a lot of people will now use yolov5 and will be very angry later on if you decide not to allow it :-)", "1133015": "I think the competition is among the classification problems. As indicated on the sample submission csv file, only the labels are required - no bounding box values. ", "1132962": "Of course! I will be using them for sure" } }