Colorblind-friendly decision regions

[rasbt]

Description

Modification to the plot_decision_region function that makes it easier for color blind people to distinguish between the different decision regions by using the matplotlib 2.0 color panel and solid decision boundary lines.

Related issues or pull requests

Fixes #347

Pull Request requirements

• Added appropriate unit test functions in the ./mlxtend/*/tests directories
• Ran nosetests ./mlxtend -sv and make sure that all unit tests pass
• Checked the test coverage by running nosetests ./mlxtend --with-coverage
• Checked for style issues by running flake8 ./mlxtend
• Added a note about the modification or contribution to the ./docs/sources/CHANGELOG.md file
• Modify documentation in the appropriate location under mlxtend/docs/sources/ (optional)
• Checked that the Travis-CI build passed at https://travis-ci.org/rasbt/mlxtend

[rasbt]

Below are two example plots. I think the "new" matplotlib default color palette is visually pleasing and according to Coblis, it might be more color blind-friendly. However, I think the transparency is still a bit of an issue with regard to color blindness; thus, adding solid decision boundary lines (in black) might help.

[coveralls]

Coverage increased (+0.006%) to 91.732% when pulling 4883c12 on colorblind-palette into 46c3e02 on master.

[rasbt]

Just made the lines a bit thinner. I think this looks fine now :)

[jrbourbeau]

I like the line idea! Was about to comment that a thinner line might look better, but it looks like you just updated to a thinner line 👍

[rasbt]

Thanks for the feedback! According to a conversation on twitter, the colors are also distinguishable now. I think this should be good to merge