Joe

I wish we would all just use pypi instead of conda.

If you want it in a pretty output from stdout, you could [run your server and pipe it into jq](https://github.com/uber-go/zap/issues/530). In my case I ran something like `uvicorn main:app --reload...

I did some memory profiling, and if you look in [keras_tuner/src/engine/hypermodel.py at function fit](https://github.com/keras-team/keras-tuner/blob/master/keras_tuner/engine/hypermodel.py#L125-L149), this is where memory starts leaking. ```Python def fit(self, hp, model, *args, **kwargs): """Train the model....

Try checking your gcc version. I had to update mine from gcc version 9 to 13 and change CMAKE_CXX_STANDARD from 14 to 17.