[ENH] Extended isolation forest

[adam2392]

Fixes #98

Depends on the completion of #75

Basically replicates the figures from: https://github.com/sahandha/eif/blob/master/Notebooks/EIF.ipynb.
The actual implementation is trivial because it is just a case of extremely randomized oblique trees.

Before submitting

• I've read and followed all steps in the Making a pull request
  section of the CONTRIBUTING docs.
• I've updated or added any relevant docstrings following the syntax described in the
  Writing docstrings section of the CONTRIBUTING docs.
• If this PR fixes a bug, I've added a test that will fail without my fix.
• If this PR adds a new feature, I've added tests that sufficiently cover my new functionality.

After submitting

• All GitHub Actions jobs for my pull request have passed.

[codecov]

Codecov Report

Patch coverage: 96.47% and project coverage change: +0.69% 🎉

Comparison is base (8770f55) 86.99% compared to head (d6fb214) 87.68%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #101      +/-   ##
==========================================
+ Coverage   86.99%   87.68%   +0.69%     
==========================================
  Files          26       28       +2     
  Lines        2153     2323     +170     
==========================================
+ Hits         1873     2037     +164     
- Misses        280      286       +6     

Files Changed	Coverage Δ
sktree/tests/test_eiforest.py	96.27% <96.27%> (ø)
sktree/__init__.py	80.76% <100.00%> (+0.76%)	⬆️
sktree/ensemble/__init__.py	100.00% <100.00%> (ø)
sktree/ensemble/_eiforest.py	100.00% <100.00%> (ø)

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[adam2392]

@SUKI-O you might be interested to see how your work in #75 enables this new algorithm to be easily added!

"Extended Isolation Forest". The algorithm we add is a slight variation, but essentially almost the same. If you're interested, feel free to read the paper in the GH issue and then if you want, you can implement an example showing how it performs relative to IsolationForest on Gaussian blob data, and sinusoidal data.

If you're interested in looking at other GH issues to solve, though we can take a look at other issues that might be of interest as well!

[SUKI-O]

Can I get started with this issue or tracking constant feature #115 is a higher priority?

[adam2392]

We need to finish the thread here first: #75 (comment)

[adam2392]

Docs rendering: https://output.circle-artifacts.com/output/job/75ed96fb-f94e-4ff9-9bf4-8a5b095d95b1/artifacts/0/dev/auto_examples/plot_extended_isolation_forest.html#sphx-glr-auto-examples-plot-extended-isolation-forest-py

[adam2392]

@SUKI-O may you review this PR and code if you have some time?

[PSSF23]

Figures lgtm

[adam2392]

Okay in this goes! Feel free to look at the code and open up a GH issue if you have any questions/comments/concerns