Greetings, I would like to programmatically interact with multiple versions of a given Python module via the reticulate package, without having to restart the R process over and over. (My understanding is reticulate does not provide such a capability by itself, as once reticulate binds to a Python instance, the connection persists for the duration of the R session.) However, it occurs to me that it might be possible to use callr to run a wrapper effectively creating a fresh instance of Python with each invocation. Before getting too far into the weeds there, I'm curious if there is anyone on the list who has already tried such an approach, and if so, was it successful, or were there obstacles/pitfalls to be aware of? I'm not asking for sample code, just curious to hear informed opinions on whether this is a fool's errand. Regards, Ben