My M1 Mac arrives in the next couple days. I'm planning on running things through the rosetta x86 emulator until the developer community bridges the gap for my favorite brew packages and ML libraries to run natively. We'll see how this goes in practice.

Edit: I’ve received my M1 MacBook and am running all the Data Science libraries I need on it natively with MiniForge—they are building for arm64 chips. I have to install the occasional thing (like openCV) from source, but MiniForge handles most dependencies that I normally get from PyPi/pip.