📚 #534
📚 #534
Conversation
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
Codecov Report
@@ Coverage Diff @@
## master #534 +/- ##
=======================================
Coverage 98.93% 98.93%
=======================================
Files 100 100
Lines 9664 9664
=======================================
Hits 9561 9561
Misses 103 103
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
|
Good idea, and also a good PR name 😄 |
…ading the package and illustrate the first few functions.
kellertuer
added
the
preview docs
|
@FedeClaudi I started a tutorial, but I am not sure this is too plain to start with, see (can also be opened as a Pluto Notebook and hopefully renders fine also in the docs on the server, it did locally already, but the server is not yet sure). Let me know what you think and whether this is a nice start (since I am still unsure how and where to start). |
|
The Pluto notebook finally renders https://juliamanifolds.github.io/Manifolds.jl/previews/PR534/tutorials/getstarted.html |
|
That's great! I really like this slower approach introducing all concepts one at the time. Great job |
|
Ok, then I will continue this one, as soon as my teaching (and my language course) allows. |
|
I wonder if there would be the same error if there was no |
|
I will try but I fear we might need that for the 3D rendering? |
|
Yes, we need GLMakie, I'm mostly curious if that's a PlutoStaticHTML.jl bug. |
|
I am not sure whether its PlutoStaticHTML or even Pluto itself (as well). edit: It is already Pluto itself. |
|
I started a discussion at fonsp/Pluto.jl#2345 |
|
I think this might be a larger issue, so what do you think about leaving the tutorial in the repository but for now not putting it in the docs? |
|
I think that's also fine, this issue with Pluto/GLMakie shouldn't block this PR. |
|
Ok, so later today I will remove the second tutorial for now and open an issue for that to keep track :) |
| @@ -7,6 +7,16 @@ in | |||
| TangentSpace | |||
| ```` | |||
|
|
|||
| # Fallback for the exponential map: Solving the corresponding ODE | |||
|
|
|||
| When additionally loading [`NLSolve.jl`](https://github.com/JuliaNLSolvers/NLsolve.jl) the following fallback for the exponential map is available. | |||
There was a problem hiding this comment.
NLSolve is for approximate inverse retraction, not the exponential map.
There was a problem hiding this comment.
Hm? I am confused. It is either approximating Log or exp but approximating a retraction would not make sense (also I think I did not change this, but to fix errors is of course always good).
There was a problem hiding this comment.
NLSolve is for approximating Log, for Exp we need OrdinaryDiffEq.
There was a problem hiding this comment.
That (to me) also sounds a little strange to have two dependencies there, but then ist log probably, yes.
Co-authored-by: Mateusz Baran <[email protected]>
|
The problem was that For a beginners tutorial I found |
|
Yes, maybe we should export that. When I use |
|
Ok, will add the export. |
|
Yes, we could make |
|
That is why I came up with the idea of the alias (or just slight renaming). Sure we should not advertise deprecated things, just as a novel user working with a product manifold, ProductRepr sounds reasonable and ArrayPartition only after some explanation (ah but I could add that explanation as well why we want to partition the array at that point). |
|
I added a small remark on the ArrayPartition. The only problem is that until releasing this version here the Manifold. has to stay upfront, because only after updating the Pluto notebook afterwards, the version used in the notebook is too old. But besides this, the PR should be finished. |
kellertuer
added
the
Ready-for-Review
I think that's fine, we can remove |
Co-authored-by: Mateusz Baran <[email protected]>
kellertuer
removed
Ready-for-Review
This starts a tutorial section, for no just with a dummy notebook, because I am still pondering about a good start.
Maybe something like generating data on the sphere and computing a mean and a second short part (or even notebook) on comparing different metric on SPDs?
One could also adapt the Benchmark and illustrate the use of different array types with that.