Closed TimQuelch closed 3 years ago
Following the general idea, that we have nice default implementations, that work (flawlessly) but might not be the most efficient one (for a special manifold, where a specific one is for some reason more efficient), I think this is a very good idea.
Merging #68 (8745a12) into master (98e5c87) will not change coverage. The diff coverage is
100.00%
.
@@ Coverage Diff @@
## master #68 +/- ##
=======================================
Coverage 99.69% 99.69%
=======================================
Files 10 10
Lines 1320 1320
=======================================
Hits 1316 1316
Misses 4 4
Impacted Files | Coverage Δ | |
---|---|---|
src/ManifoldsBase.jl | 99.40% <100.00%> (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 98e5c87...8745a12. Read the comment docs.
Could you just bump the version to 0.10.7
? Then we would be good to go (i.e. merge).
I bumped the version. I also force-pushed to reword the commit message to be more correct
Due to numerical errors the inner product can sometimes return an infinitesimal negative number (e.g. -1e-26). This negative number causes the sqrt to fail with a domain error
This fix ensures that the result of the inner product of X and X is always greater than or equal to zero (which it should be by definition)