New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
New YOLOv5 support #19
Conversation
README.md
Outdated
This method allows you to use the latest Yolov5 version found on [Torch Hub](https://pytorch.org/hub/ultralytics_yolov5/). The `github` flag sets the model path on torch hub which is `ultralytics/yolov5` by default. The `type` flag sets the type of model like yolov5s, yolov5l, custom, etc. The default is `yolov5s`. The `sourceType` flag determines where to load the weights from which is either local or github. The default is local which means you have to provide pretrained weights. | ||
```bash | ||
|
||
python clip_object_tracker.py --weights models/yolov5s.pt --source "data/video/cars.mp4" --detection-engine hub --github ultralytics/yolov5 --type yolov5s --sourceType github --info |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
testing custom models and posting ex here
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@maxhs2014 any idea where we were on this? |
@Jacobsolawetz last I checked this worked perfectly fine but models trained on the older version of yolov5 in our notebooks before did not work because of conflicting yolov5 code. Now that we are completely on a suitable version of yolov5 it should work fine (I can test again), but the same thing could happen in the future if roboflow notebook models are on an older version than torch hub. |
@Jacobsolawetz - what's the next step? |
No description provided.