Naive inverse cumulative distribution sampler
Creates a cumulative distribution naively from any function y(x). Then the function sample_n(N) creates a sample of N uniformly sampled values between 0 and 1 (from numpy.random.uniform), checks index where the cumulative distribution is closest to each value, then saves the x value associated with the selected index. In this way, a distribution of N values is built up that has a distribution matched to the input function y.
Usage examples can be found in the jupyter notebook included in this repo