Hey I’m trying to setup a bayesian sensor for occupancy in a room I don’t have a motion sensor for, but I think I may be setting this up incorrectly. It’s hard to wrap my head around it. First here is my sensor:
https://hastebin.com/isonosenik.pl
The way I understood it is basically the prob_given_true just adds and adds the more things that are true (so say the computer is on and the fan is on, my max probability would be 0.5) and that the prior would essentially take a probability of that.
But when I assume it is false, it’s true. I looked at it through templating and I got this:
<template state binary_sensor.office_bay_occupancy=on; observations=[{'prob_true': 0.1, 'prob_false': 0.0}, {'prob_true': 0.4, 'prob_false': 0.0}, {'prob_true': 0.1, 'prob_false': 0.0}], probability=1.0, probability_threshold=0.9, friendly_name=Office Bay Occupancy @ 2018-10-01T18:47:36.198996-04:00>
Which makes no sense to me. How am I getting a probability of 1.0? Any help would be appreciated!