ghsanti

I think this is a problem with config files overwriting stuff. You can convert the model and pass -f path/to/model.pth

I have faced the exact same issue, also changing to `-b 64` @FateScript

I have had this error in the past with tensorflow and now with onnx runtime. Similar to canvas it seems to be because of a `.node` file @sheremet-va ```shell Error:...

@PrajwalCogniac use yolov7 or yolox this project is unmaintained and does not work

No, you cant transfer the weights to yolox or yolov7 because the Neural Net Architectures are different shape, and the weights can not be transferred to completely different shapes. The...

I recommend you to switch to https://github.com/WongKinYiu/yolov7 because it is active or you will end up days stuck fixing things, unless you have experience

i think this will error because it calls element.item() afterwards...

I've [tried something here](https://github.com/tensorflow/profiler/issues/602#issuecomment-2264427977)I don't know if it's a meaningful plot yet, using **Trace**. Seems to be ok for CPU only though. [GIST](https://gist.github.com/ghsanti/ca621a1f88778145c33665eac6807936)

@lakeyk could you share a gist/colab notebook to reproduce it? Note that the problem in most cases is: 1. Not using the right `--logdir` path 2. Having correct logdir path...