Peng Algorithm

[banrovegrie]

This is the PR involving the addition of the second algorithm to the PSDP module. Here, I have included a rough sketch of the algorithm.

The issues that remain have been described in discussion #179.

[banrovegrie]

Apparently, autopep8 converts some of the lambda functions into proper functions. I hope that is fine?

@FanwangM

[banrovegrie]

Tests won't pass for this commit since I have many linting issues which we can ignore for now.

[FanwangM]

This is another improvement. Thanks.

For the commit, ff04043, is is a similar operation that we have encountered in the previous PR? If you want to update the git history, we should use git rebase to have a clean history this time. If this is not what you mean for that commit, can you explain a little? Thank you.

[banrovegrie]

@FanwangM yeah, so the issue was that I had the commits added to the branch before test-psdp was added and somehow in the end I ended messing up while rebasing when i encountered a few merge conflicts.

[banrovegrie]

I think the code is ready to be reviewed btw. Once we merge this commit, I will proceed with updating the tests to include testing for this algorithm as well.

[FanwangM]

Thank you for the codes.
I have some minor comments here.

[codecov]

Codecov Report

Merging #180 (171a794) into master (ec238c4) will decrease coverage by 2.22%.
The diff coverage is 30.00%.

@@            Coverage Diff             @@
##           master     #180      +/-   ##
==========================================
- Coverage   84.42%   82.19%   -2.23%     
==========================================
  Files          13       13              
  Lines         751      775      +24     
==========================================
+ Hits          634      637       +3     
- Misses        117      138      +21     

Impacted Files	Coverage Δ
procrustes/psdp.py	74.44% <30.00%> (-22.53%)	⬇️

[FanwangM]

I noticed that you have two git merge operations. Can you try to use git rebase to avoid them. This can help us get a clean history.? You can either do (1) close this PR and open a new one; or (2) use more complicated git history editing operations to avoid these two. I would prefer the first one. Thanks. @banrovegrie

[banrovegrie]

Yeah, I will close this PR and open a new one.

[banrovegrie]

Closing this and pivoting to #187.