Kiran Sanjeevan Cabeza

Hey @jokerpp. I think ideally I want to move the repo to use torchaudio instead of torhaudio_contrib. Are you getting any specific error when you try to script the module?

You'll need to either pad the input with silence, or modify the network (smaller kernel size, or stride, or padding).

@jokerpp can you confirm that you have the `metadata/UrbanSound8K.csv` file in whatever directory you've specified for the urbansounds dataset?

Yeah this was using local mode, using remote seems to be fine.

Hi @laurenyu, yes thanks! In an ideal world, we could pass some kind of `run_hyperparameters` dictionary so that we can add any flag for `docker run`, but having `shm-size` is...

I've had some progress on this. It can be solved with two steps: 1) Using the `docker run` flag propagataion introduced https://github.com/mlflow/mlflow/pull/2608 by passing: ``` mlflow run ... -A net=host...

I'm not sure the credentials point is true @AAA97AAA. If that were the case, this error would also come up when using a remote tracking server using a `conda env`,...