Python module loading tends to be quite slow. Loading Tensorflow on our Heroku production environment takes 10 seconds (5 on my laptop). A client of mine runnin

Python module loading tends to be quite slow. Loading Tensorflow on our Heroku production environment takes 10 seconds (5 on my laptop). A client of mine running Rust and Python on Embedded Linux device found that loading numpy/pandas modules took over 5 seconds (laptop was under 1), and their computation took just x00 milliseconds. 10x overhead... So the spawning of a short-lived subprocess approach has massive overhead, only suitable for multi-second workloads.