Create Unsupervised Random Forest Usecase Example

[jshinm]

Is your feature request related to a problem? Please describe. Once FasterBIC implementation has been delivered, create an unsupervised RF usecase example using iris data. Discuss the difference among TwoMeans, FastBIC, and FasterBIC (e.g., characteristics, pros/cons)

Describe the solution you'd like Prepare either a sklearn style pytest and/or notebook that demonstrates this

Estimated completion date The prerequisite for this task is successful implementation of aforementioned three criterions. After implementation and successful build, this task should be started. Rough projected completion date is 2/17/2023

[adam2392]

In the PR addressing this, you can add two examples:

1. An example showing off the unsupervised tree functionality and comparing it to say k-means for the iris dataset. Very simple and very straightforward.
2. Another example inspired by one of the figures in https://arxiv.org/pdf/1907.02844.pdf. Create a simulation, and add noise dimensions and compare it to k-means, or even Isomap