Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

linear algebra documentation #2593

Closed
ViralBShah opened this issue Mar 18, 2013 · 7 comments
Closed

linear algebra documentation #2593

ViralBShah opened this issue Mar 18, 2013 · 7 comments
Assignees
Labels
docs This change adds or pertains to documentation
Milestone

Comments

@ViralBShah
Copy link
Member

@andreasnoackjensen I have made updates to the documentation on master, but it would be great if you can review it.

Would it be ok to rename eigenfact to eigfact and the type to EigenDense, to keep the naming scheme consistent? I realized this while trying to document things.

@dmbates - Yes, I know your views on consistentcy. ;-) Jokes aside, could you also take a look the dense linear algebra documentation, and make suggestions for improvement. It would be much easier to take a crack at sparse linear algebra documentation then.

@ViralBShah
Copy link
Member Author

@nolta I have separated out the linear algebra documentation into a separate file in 271961c. It is not showing up here, and the date shown on this page is 3 days old. Am I doing something wrong?

http://docs.julialang.org/en/latest/stdlib/

@andreasnoack
Copy link
Member

Sure. Feel free to change the names. I'll have a look at the documentation.

@ViralBShah
Copy link
Member Author

Need to also document hessfact. Also document how to use Hermitian and other matrix properties.

ViralBShah added a commit that referenced this issue Mar 18, 2013
Eigen -> EigenDense
eigenfact -> eigfact
Hessenberg -> HessenbergDense

Includes doc updates for linalg as well.
@dmbates
Copy link
Member

dmbates commented Mar 18, 2013

@ViralBShah Because the sparse matrix factorizations from umfpack.jl and cholmod.jl use the same names as those in factorization.jl I suggest that they be documented in linalg.rst too and will add those.

Also, I thing the description of the \ operator is inaccurate. The Hermitian case uses the BunchKaufman factorization, not the Cholesky factor.

I will go ahead and add the documentation of the sparse matrix factorizations unless you feel they would be better addressed in the sparse.rst file.

@andreasnoack
Copy link
Member

Here are my comments to the linalg documentation

lu: I made it return the permutation matrix and not the vector. Hence P*A=L*U. This is the problem that @StefanKarpinski mentioned about which of the MATLAB versions to mimic.

qrp: Same as for lu

eigfact: chol should be inv.

svdfact: Should be: "U, S, V and Vt can be obtained from the factorization F with F[:U], F[:S], F[:V] and F[:Vt]". Maybe it should be said that 'Vt' avoids a transpose.

svdt: Removed. Use the factorization instead.

@ViralBShah
Copy link
Member Author

Could you modify the docs? All good stuff.

@andreasnoack
Copy link
Member

c01415e updates the docs

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
docs This change adds or pertains to documentation
Projects
None yet
Development

No branches or pull requests

3 participants