[ENH] Adding intervention graph classes for known graphs and equivalence classes and improve graph class documentation

[adam2392]

Signed-off-by: Adam Li [email protected]

Xref PR implementing the PsiFCI algorithm: py-why/dodiscover#111

Changes proposed in this pull request:

• Implements AugmentedGraph, IPAG and PsiPAG with minimal code-duplication
• Adds unit tests on the F-nodes
• Adds documentation details to ADMG, CPDAG and PAG that addresses some of the shortcomings described in [BUG] R8 does a check for a tail incorrectly dodiscover#114
• the main file to review is pywhy_graphs/classes/intervention.py

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-commenter]

Codecov Report

Merging #49 (aab9681) into main (4e9cef8) will increase coverage by 0.57%.
The diff coverage is 89.56%.

@@            Coverage Diff             @@
##             main      #49      +/-   ##
==========================================
+ Coverage   83.46%   84.03%   +0.57%     
==========================================
  Files          29       29              
  Lines        2147     2255     +108     
  Branches      558      578      +20     
==========================================
+ Hits         1792     1895     +103     
- Misses        253      254       +1     
- Partials      102      106       +4     

Impacted Files	Coverage Δ
pywhy_graphs/algorithms/generic.py	85.57% <ø> (ø)
pywhy_graphs/algorithms/pag.py	91.02% <ø> (ø)
pywhy_graphs/classes/admg.py	95.12% <ø> (ø)
pywhy_graphs/classes/cpdag.py	83.67% <ø> (ø)
pywhy_graphs/classes/pag.py	82.25% <ø> (ø)
pywhy_graphs/networkx/classes/mixededge.py	90.00% <62.50%> (-0.85%)	⬇️
pywhy_graphs/classes/intervention.py	91.89% <91.58%> (+91.89%)	⬆️

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[adam2392]

Docs: https://output.circle-artifacts.com/output/job/9301c837-88b2-4f0a-9390-1d778fbc2be1/artifacts/0/dev/use.html

[adam2392]

User-guide: https://output.circle-artifacts.com/output/job/cd549827-c263-4343-a1d7-12dc9455ae5e/artifacts/0/dev/user_guide.html

Intervention graphs: https://output.circle-artifacts.com/output/job/cd549827-c263-4343-a1d7-12dc9455ae5e/artifacts/0/dev/generated/pywhy_graphs.AugmentedGraph.html#pywhy_graphs.AugmentedGraph

[adam2392]

Whenever you guys get the chance, if either of you are familiar enough w/ the intervention graphs to provide a review, it would be very helpful!

The corresponding implementation of PsiFCI with known and unknown intervention targets is implemented in py-why/dodiscover#111

[adam2392]

@jaron-lee and @robertness lmk if you guys prefer to take a review before I merge. Just cuz I need it in main for dodiscover Psi-FCI PR.

However, think some reviews is good at least to get the team acquainted

[adam2392]

Ima go ahead and merge this to get PsiFCI working and get us closer to just a v0.1 for pywhy-graphs. We can open up an issue to address whether there are concerns w/ implementation.