Using Monte Carlo Simulation to generate probability measure preserving bins for DATA.

Thanks for the data visualization courses by professor Lev Manovich from the graduate center of CUNY. I have learned a lot from that fantastic course and used what I have learned in my work and research. Here is one of my practices by using what I have learned from that amazing course.

Put the orientation(-pi to pi) into different bins by using different methods

As we know, Uniform and Gaussian distributions are two commonly used distributions. When quantifying those distributions into bins, we can have either equal bin in the domain or equal probability measure within each bin. For the latter one, the consequence is we will have an unequal bin in the domain if the distribution is not uniform. It will be useful to emphasise on the data with high probability measures.

As mentioned in the title, in order to generate this plot, Monte Carlo Simulation was used. Random points are generated and then put into different bins by different methods and the color code is provided for each bin. Everything is in the code if you are interested in the details.

Here is the code used to generate this plot.

Thanks for your time in reading.

Slightly change the figure by removing the frames

Add the following several lines after line 94 of the previous code

axs[0].axis('off')
axs[1].axis('off')
axs[2].axis('off')