Hi,
I guess it would be easier if it uses integer data, but multiplied it by 10, and then the python script divided it by 10.
for example, you want to send data that have range 0-360 from Arduino to python, and want a fractional things to be send to python over Arduino. rather than force it to send float, why not just multiplied it by 10 to add more precision, so the range could be 0,0-360,0 , but you just send 0-3600 to python, and then the python divide the value by 10 so it could be saved as float 0,0-360,0
let's discuss the details over chat
best regards,
Rijal