Symmetric quad [DRAFT]

[zsteve]

The semi-smooth Newton algorithm of [Lorenz et al. 2019] can be specialised to the case of symmetric inputs, much like the approach already used for symmetric Sinkhorn algorithm in the already implemented Sinkhorn divergence. This problem can be solved faster than the non-symmetric case.

Reference:
Quadratically regularized optimal transport. DA Lorenz, P Manns, C Meyer - Applied Mathematics & Optimization, 2019

[coveralls]

Pull Request Test Coverage Report for Build 5367524588

Details

• 0 of 109 (0.0%) changed or added relevant lines in 1 file are covered.
• 2 unchanged lines in 2 files lost coverage.
• Overall coverage decreased (-13.4%) to 81.922%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/quadratic_newton_symm.jl	0	109	0.0%

Files with Coverage Reduction	New Missed Lines	%
src/entropic/sinkhorn_unbalanced.jl	1	96.08%
src/entropic/symmetric.jl	1	94.92%

Totals
Change from base Build 2897651829:	-13.4%
Covered Lines:	648
Relevant Lines:	791

---

💛 - Coveralls

[codecov-commenter]

Codecov Report

Merging #160 (c95b498) into master (de56119) will decrease coverage by 13.77%.
The diff coverage is 0.00%.

@@             Coverage Diff             @@
##           master     #160       +/-   ##
===========================================
- Coverage   95.42%   81.64%   -13.78%     
===========================================
  Files          14       15        +1     
  Lines         634      741      +107     
===========================================
  Hits          605      605               
- Misses         29      136      +107     

Impacted Files	Coverage Δ
src/quadratic_newton_symm.jl	0.00% <0.00%> (ø)

---

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 de56119...c95b498. Read the comment docs.

[davibarreira]

Are we supposed to review this? The "[Draft]" is throwing me off.