Mahesh Deshwal

> > Thanks for your interest. It depends on the dataset resolution. Larger input size will help with retraining details of those images with larger sizes. In our next paper,...

Yes, it is working fine with CPU. I there is an issue with the working of `GPU tensorflow` due to dependency

> @Kikedao @anguoyang @juergengunz @xuebinqin @mutsuyuki i am new to the world of ML and AL , and i don't know how to convert this .pth file into onnx file...

> @deshwalmahesh thanks for the reply , the problem was that , i was not able to create the dummy inputs for onnx conversion but i found the code that...

No Solutions to this. For now, you can use `Detectron 2` as is given in the official `Dit for object detection`

@younesbelkada I'm using `-U` for everything while installing. Actually I'm using Multi GPU, single Node with Deepspeed Zero-2 with accelerate. I'm also using Checkpointing and Gradient Accumulation. This could cause...

Can you explain a bit on the problem?

> @deshwalmahesh Hello, I am also facing the same issue with the code while running the Jupyter notebook script (Object tracking part). Also, bounding boxes were generated earlier but now...

> I am using the same model as you (yolo v7 with deep sort tracking) on a 16 GB RAM computer with a windows operating system. I also tried it...

Hey! See if [this helps you](https://github.com/deshwalmahesh/yolov7-deepsort-tracking/issues/7) You can edit the file `bridge_wrapper.py` every time or maybe take a `your_fork` -> replace your version and then `git pull your_fork.git` in `Colab`...