Need an example of Python code for learning NN with reinforcement. Does anyone have?

Question

Python

I'm trying to figure out learning with a teacher, but somehow there is not much information in Russian. In such a situation, it will be easier for me to figure it out through the code.
Interested in an example code for training a neural network with reinforcement. Preferably in conjunction with Python + Tensorflow. There is a lot of code on the web, but it's all supervised learning. In this case, the part of more interest is how to make your own function for calculating the network error in the learning algorithm with subsequent adjustment of the weights.
Does anyone have something ready or a link on the topic to something Russian?

Answer 1

Here is an open RL course that takes place in shad, there are many examples: https://github.com/yandexdataschool/Practical_RL

Answer 2

Here I found another good example for parsing and learning on it
https://www.machinelearningmastery.ru/reinforcemen...