Closed mateuszbaran closed 1 year ago
Merging #160 (c35a1be) into master (71f85b6) will decrease coverage by
0.12%
. The diff coverage is100.00%
.:exclamation: Current head c35a1be differs from pull request most recent head 9d68f5b. Consider uploading reports for the commit 9d68f5b to get more accurate results
@@ Coverage Diff @@
## master #160 +/- ##
==========================================
- Coverage 99.96% 99.84% -0.12%
==========================================
Files 19 20 +1
Lines 2539 2553 +14
==========================================
+ Hits 2538 2549 +11
- Misses 1 4 +3
Files Changed | Coverage Δ | |
---|---|---|
src/DefaultManifold.jl | 100.00% <100.00%> (ø) |
|
src/maintypes.jl | 100.00% <100.00%> (ø) |
... and 1 file with indirect coverage changes
:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more
While I think this is quite technical and should not concern beginners, it might be nice for advanced usages. A small benchmark illustrating when which is nicer might be nice as well (if that can be well-illustrated)
Yes, it's definitely not for beginners. Benchmarks would be hard at the moment because all good examples live in Manifolds.jl. I think rotation matrices would make a great example for a benchmark.
Sure let's just keep that a bit in mind to explain this feature with a benchmark somewhen. I think I see advantages of both and disadvantages, but having a small tutorial on that would be great – sure probably only in Manifolds.jl
Sure, I will add a clear explanation with an example in the Manifolds.jl PR that converts key manifolds to AbstractSize
.
Basic infrastructure for solving https://github.com/JuliaManifolds/Manifolds.jl/issues/625 .