[MRG] Remove seaborn dependencies

[glemaitre]

Reference Issue

Fixes #257
Fixes #265

What does this implement/fix? Explain your changes.

The examples are currently relying on searborn. We should modify them to purely rely on matplotlib.
Additionally, the following examples are improved:

• ensemble methods -> better illustration of the methods
• instance threshold hardness -> better illustration of the ratio
• allknn/RENN/ENN -> the examples has been merged
• SMOTE -> merge all SMOTE together
• Nearmiss -> Merge all nearmiss together

Any other comments?

[pep8speaks]

Hello @glemaitre! Thanks for updating the PR.

• In the file examples/combine/plot_smote_enn.py, following are the PEP8 issues :

Line 62:1: E128 continuation line under-indented for visual indent
Line 63:1: E128 continuation line under-indented for visual indent
Line 64:1: E901 TokenError: EOF in multi-line statement

Comment last updated on March 23, 2017 at 12:15 Hours UTC

[codecov]

Codecov Report

Merging #264 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #264   +/-   ##
=======================================
  Coverage   98.27%   98.27%           
=======================================
  Files          58       58           
  Lines        3427     3427           
=======================================
  Hits         3368     3368           
  Misses         59       59

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update a55e37d...c20f09a. Read the comment docs.

[glemaitre]

@chkoar I think that they are nicer now:

https://548-36019880-gh.circle-artifacts.com/0/home/ubuntu/imbalanced-learn/doc/_build/html/auto_examples/index.html

You can check the results with the artifacts.

[chkoar]

@glemaitre yeap indeed they are!

I think that the legend for the next examples has to be fixed because overlaps wight X dimension

under-sampling-plot-enn-renn-allknn
under-sampling-plot-nearmiss
plot-instance-hardness-threshold
ver-sampling-plot-smote

[glemaitre]

Thanks, locally I didn't have this issue. I will check those.

[glemaitre]

@chkoar Check the artifacts but I think that this is fine this time.

[glemaitre]

https://580-36019880-gh.circle-artifacts.com/0/home/ubuntu/imbalanced-learn/doc/_build/html/auto_examples/index.html

[glemaitre]

https://584-36019880-gh.circle-artifacts.com/0/home/ubuntu/imbalanced-learn/doc/_build/html/auto_examples/index.html