I’m totally new with tensorflow [I know it exists, I know what I can do with it, and that’s all]. I use RPi3 with hass.io installation, so it’s too weak to use tensorflow on it, but I thought I’ll play a little and test some options.
my idea is, to install tensorflow using docker, on a remote machine [I mean: machine in my LAN, but not the HA raspberry], and use it with HA still sitting on RPi.
I know that you will probably say “hey, just install HA+tensorflow on the same machine” - it is a possibility, probably will think about it, but only if it’s worth for me - and that’s what I want to test using my current HA config.
so like in the topic - is it possible? by setting some path/url that HA will search for tensorflow there?