Closed banrovegrie closed 1 year ago
Merging #195 (1df336b) into master (c8564cc) will decrease coverage by
0.08%
. The diff coverage is93.54%
.
@@ Coverage Diff @@
## master #195 +/- ##
==========================================
- Coverage 94.38% 94.30% -0.09%
==========================================
Files 11 11
Lines 695 755 +60
==========================================
+ Hits 656 712 +56
- Misses 39 43 +4
Impacted Files | Coverage Δ | |
---|---|---|
procrustes/psdp.py | 94.66% <93.54%> (-0.89%) |
:arrow_down: |
Some linting issues remain. Will fix them as soon as possible. Let me know if there is anything
Resolved some of the issues, and I will fix all of them by tomorrow night. I will tag you when I finally push the updated code.
Resolved some of the issues, and I will fix all of them by tomorrow night. I will tag you when I finally push the updated code.
Thanks.
Can you push your local work to fix the problems so that we can move forward to final cleaning up/reafactorization for GSoC? Thanks. @banrovegrie
Yep, I was having a few issues regarding refactoring the code. I will get them done right away.
@FanwangM The code has been reformatted.
Should we merge this and open a new PR for fixing tests? @FanwangM
@FanwangM let me know if I need to do anymore updates before we merge this.
@FanwangM let me know if I need to do anymore updates before we merge this.
Sorry for the late reply. I am going to merge it soon after one minor fix.
@FanwangM @PaulWAyers This is the interim implementation of OptPSDP (the required third algorithm). Tests have passed on my local machine.
The algorithm appears to provide us with better error thresholds when compared to the previous algorithms. However, the threshold is quite nominal for the examples we have had in place. Optimizing the algorithm for the case when $A$ is a diagonal matrix remains.
Thanks to @abhishekmittal15 for helping me understand the MATLAB implementation provided here!