NumPy, Monte Carlo and Markov Decision Processes
fastdice.ipynb
This 'essay with code' aims to use NumPy to its full extent for efficient and elegant Monte Carlo simulations with generalized dice. Later we also introduce a Markov chain class, implement a dynamic programming algorithm for computing the optimal policy and briefly put the topic in the context of existing mathematical theory. It is recommended to use Jupyter Lab for its ToC extension.- the
fastdice
package organizes the classes, methods, example functions and other functions into a package, seeing how the code is very flexible and can be used for many different ideas. For a guide it is best to consult the notebook.