Remove arm_stop(0) print.

Oktober 26, 2020

Found this Youtube lecture on do-calculus in model based learning. Can be summerized that when a variable is latent (non-observable), cutting the causal dependency to one observable variable gives problems if another variable is depend on both the observable and non-observable variable, because one have to sample from the same probability distribution. This can be solved by adding another observable variable: the front-door estimand (which can use its own probability distribution for sampling).

Oktober 17, 2020

Tried to visualize alpha-beta search with the Jupyther notebook from aima python code.
Nice implementation, including reading pseudo-code. Yet, simply pip3 install -r requirements.txt fails, so installing the prerequisites manualy.
The notebook games4e.ipynb works fine (including game-trees). Checking games.ipynb, which makes it easier to define your own game. Yet, this notebook still fails on numpy.
With moving an old local version of numpy to /tmp the games.ipynb starts. Works fine, except the part of minimax_decision. Can be repaired with psource(minmax_decision) (no 'i'). The canvas goes wrong (str instead of object), even after the correction from notebook import Canvas_min_max
The visualisation of alpha_beta algorithm works fine, after a small correction from notebook import Canvas_alpha_beta (additional '_').

This TCS3200 sensor would be great for the line-following assignment of the Turtlebot 3 of the Embedded Software and Systems course of the Software Engineering master.
The sensor could be connected to Rasberry PI, which has the same PIN layout as the Turtlebot, according to this tutorial.
More detailed information on the Turtlebot can be found here.