Refactor to be current with kneed v0.4.1

[arvkevi]

This PR fixes #931

I have made the following changes:

1. Refactored utils/kneed.py to be current with kneed v0.4.1.

TODOs and questions

Questions for the @DistrictDataLabs/team-oz-maintainers:

• The issue raised by @Kautumn06 is still unresolved. I show a quick experiment with different random_state values. Should a separate issue be created for this?

CHECKLIST

• Is the commit message formatted correctly?

• Included a sample plot to visually illustrate your changes?
• Do all of your functions and methods have docstrings?
• Have you added/updated unit tests where appropriate?
• Have you updated the baseline images if necessary?
• Have you run the unit tests using pytest?
• Is your code style correct (are you using PEP8, pyflakes)?
• Have you documented your new feature/functionality in the docs?

[lwgray]

@arvkevi Thank you so much for this PR we really appreciate your contributions

[rebeccabilbro]

Thank you so much @arvkevi for helping to keep our port of kneed up-to-date and for resolving #931! I'll get this merged in now!

Just to loop in @Kautumn06 — here's what the new elbow looks like for a synthetic dataset:

from sklearn.cluster import KMeans
from sklearn.datasets import make_blobs

from yellowbrick.cluster import KElbowVisualizer


X, y = make_blobs(n_samples=1000, n_features=12, centers=8, random_state=12)

model = KMeans()
visualizer = KElbowVisualizer(model, k=(4,12))

visualizer.fit(X)
visualizer.poof()

And for our new NFL dataset:

[Kautumn06]

This looks great @arvkevi! I really appreciate your help to get this fixed, and thank you also for taking the time to create and work through the checklist!

[arvkevi]

Thanks, looking forward to the release 😄