A python library for the markov 2 parameter RNG model

To follow along, you kind of need the paper, which isn't published yet

import markov2p as m

Compute markov params from bias and serial correlation coefficient
    p01,p10 = m.biasscc_2_p(bias,scc)

Compute probability of a symbol appearing from a generator with specific parameters
    symbol_probability = m.symbol_prob(p01,p10,symbol,bitwdith)

Compute the most probable symbol to be generated from a model with a specified set of parameters.
    mps = m.most_probable_symbol = (p01,p10,bitwidth)

Compute the per-bit entropy, the most common symbol value and the probability of the most common symbol of a model with specified parameters.

    entropy,mcv_prob,mcv = p_to_entropy(p01,p10,bitwidth)

Choose a set of parameters that generate data with a specified entropy. There are an infinite number of such parameters, unless the entropy is 1.0. So this choose one randomly from the available set of parameters that generate data with the chosen min entropy.
    p01,p10 = pick_point(desired_entropy,epsilon,entropy,bitwidth)