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

Add six Krylov processes #618

Merged
merged 17 commits into from
Sep 28, 2022
Merged

Add six Krylov processes #618

merged 17 commits into from
Sep 28, 2022

Conversation

amontoison
Copy link
Member

@amontoison amontoison commented Aug 30, 2022

@amontoison amontoison requested review from dpo and tmigot August 30, 2022 18:01
@codecov
Copy link

codecov bot commented Aug 30, 2022

Codecov Report

Base: 97.97% // Head: 98.06% // Increases project coverage by +0.08% 🎉

Coverage data is based on head (8752a04) compared to base (0604212).
Patch coverage: 100.00% of modified lines in pull request are covered.

❗ Current head 8752a04 differs from pull request most recent head 56e4c48. Consider uploading reports for the commit 56e4c48 to get more accurate results

Additional details and impacted files
@@            Coverage Diff             @@
##             main     #618      +/-   ##
==========================================
+ Coverage   97.97%   98.06%   +0.08%     
==========================================
  Files          36       37       +1     
  Lines        6072     6349     +277     
==========================================
+ Hits         5949     6226     +277     
  Misses        123      123              
Impacted Files Coverage Δ
src/krylov_processes.jl 100.00% <100.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

@github-actions
Copy link
Contributor

github-actions bot commented Aug 30, 2022

Package name latest stable
CaNNOLeS.jl
DCISolver.jl
FletcherPenaltySolver.jl
JSOSolvers.jl
LLSModels.jl
Percival.jl
RipQP.jl

Copy link
Member

@tmigot tmigot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It looks good to me, thanks!

@@ -0,0 +1,282 @@
### Notations
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Similar to the other pages in the doc, I think it misses Krylov processes as a title followed by one sentence explaining what is here.

Copy link
Member

@dpo dpo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This will be of great help. I have the impression that the code is for the complex case, the algorithms are for the real case, and the description is a mixture :-).

```math
\begin{align*}
A V_k &= V_k T_k + \beta_{k+1} v_{k+1} e_k^T = V_{k+1} T_{k+1,k}, \\
B U_k &= U_k S_k + \bar{\gamma}_{k+1} u_{k+1} e_k^T = U_{k+1} S_{k+1,k}, \\
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

A bit confused here. $S_k$ should be $T_k^T$.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Do you mean that $S_k$ arises from the formulation for complex numbers? If so, you shouldn't be using transposed matrices everywhere, but transposed conjugates.

\bar{\gamma}_2 & \bar{\alpha}_2 & \ddots & \\
& \ddots & \ddots & \bar{\beta}_k \\
& & \bar{\gamma}_k & \bar{\alpha}_k
\end{bmatrix},
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks like $S_k = T_k^*$ ?!

```

!!! note
We also have the relations $T_k = S_k^T$, $\enspace T_{k,k+1} = \begin{bmatrix} T_{k} & \gamma_{k+1} e_k \end{bmatrix} = S_{k+1,k}^T \enspace$ and $\enspace S_{k,k+1} = \begin{bmatrix} S_{k} & \bar{\beta}_{k+1} e_k \end{bmatrix} = T_{k+1,k}^T$.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

$T_k = S_k^T$ only in the real case.

@github-actions
Copy link
Contributor

github-actions bot commented Sep 7, 2022

Package name latest stable
CaNNOLeS.jl
DCISolver.jl
FletcherPenaltySolver.jl
JSOSolvers.jl
LLSModels.jl
Percival.jl
RipQP.jl

@amontoison
Copy link
Member Author

@tmigot @dpo
It's ready for a second round of review.

@github-actions
Copy link
Contributor

github-actions bot commented Sep 7, 2022

Package name latest stable
CaNNOLeS.jl
DCISolver.jl
FletcherPenaltySolver.jl
JSOSolvers.jl
LLSModels.jl
Percival.jl
RipQP.jl

@github-actions
Copy link
Contributor

github-actions bot commented Sep 7, 2022

Package name latest stable
CaNNOLeS.jl
DCISolver.jl
FletcherPenaltySolver.jl
JSOSolvers.jl
LLSModels.jl
Percival.jl
RipQP.jl

@github-actions
Copy link
Contributor

github-actions bot commented Sep 7, 2022

Package name latest stable
CaNNOLeS.jl
DCISolver.jl
FletcherPenaltySolver.jl
JSOSolvers.jl
LLSModels.jl
Percival.jl
RipQP.jl


## Symmetric Lanczos

![symmetric_lanczos](./graphics/symmetric_lanczos.png)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Out of curiosity, does MathJax allow us to typeset algorithms?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Also, specify that $A = A^H$.

Copy link
Member Author

@amontoison amontoison Sep 15, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Out of curiosity, does MathJax allow us to typeset algorithms?

I will try pseudocode.js.

@amontoison
Copy link
Member Author

With #637, we should be able to run the processes on GPUs.

@amontoison amontoison merged commit 426557d into main Sep 28, 2022
@amontoison amontoison deleted the krylov_processes branch September 28, 2022 04:15
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants