-
### Search before asking
- [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no simi…
-
### Search before asking
- [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no simi…
-
### Search before asking
- [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no simi…
-
### Search before asking
- [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and found no similar bug report.
### YOLOv5 Component
Detection
### Bug
Below, co…
-
I have trained my own data set on YOLOv5s which worked perfectly fine. Now I want to predict an image with the resulting weights (best.pt). My code looks like this:
`device = torch.device('cuda:1')…
-
### Search before asking
- [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no simi…
-
## Description
I tried to convert a Yolov5s ONNX int8 model to a tensorRT int8 model. But all the PointWise operations are not converted to int8 as I would expect. By default, they are converted to…
-
Hi All,
With **tensorRT 8.5**, **cuddn 8.6.0** I make an inference with **cuda 11.8**
I obtain globally in term of time processing -> **280 ms on an T4**.
Previously i was on **cuda 10** **te…
-
### Search before asking
- [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no simi…
-
### Prerequisite
- [X] I have searched [the existing and past issues](https://github.com/open-mmlab/mmyolo/issues) but cannot get the expected help.
- [X] I have read the [FAQ documentation](https://…