Updated Manifold to extend Projection Visualizer

[naresh-bachwani]

This PR updates Manifold to extend Projection Visualizer as a part of issue #874.

Following changes are made to Manifold:

1. It now handles 3d plot.
2. Removed duplicate code.
3. Added new parameters like projection

Sample plot

from yellowbrick.datasets import load_credit
from yellowbrick.features import Manifold

oz = Manifold(manifold='tsne')
oz.fit_transform(*load_credit())
oz.poof(outpath="tsne_credit.png")

TODOs and questions

Still to do:

• Add tests.

• Is the commit message formatted correctly?

• Have you noted the new functionality/bugfix in the release notes of the next release?

• 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?

• Have you built the docs using make html?

[bbengfort]

Glorious code simplification! All of your hard work in the ProjectionVisualizer has paid off and now Manifold makes use of all that functionality and is so much more than it was before! It's awesome to see a plan come together in this way! Very good work!

I have a few minor requests mostly related to documentation, but then we should be good to go.

One question I have looking through the new baseline images - it appears that test_manifold_classification.png and test_manifold_classification_3d.png have sequential instead of discrete colors. I think this is because of the default of setting the colormap in DataVisualizer ... we may have to remove that because of how it changes Manifold.

Also, I want to talk about style and communication in this plot. Would you mind running the following code and posting the resulting image in the PR?

from yellowbrick.datasets import load_credit

oz = Manifold(manifold='tsne')
oz.fit_transform(*load_credit())
oz.poof(outpath="tsne_credit.png")

[bbengfort]

Thanks for making those changes!

[bbengfort]

Sorry, I couldn't see this thanks to the big diff between things!

[bbengfort]

@naresh-bachwani looks great, thank you for making those changes!