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

calculating interpolations (WSOS cone setup) takes too much memory #20

Closed
chriscoey opened this issue Aug 5, 2018 · 4 comments
Closed
Labels
help wanted Extra attention is needed speed

Comments

@chriscoey
Copy link
Collaborator

chriscoey commented Aug 5, 2018

for n > 5, even with very low degree (d = 3), calculating the approximate Fekete points and evaluating the Chebyshev polynomials on them and performing the QR takes too much memory. this is the current method. perhaps using one of the fast multivariate polynomials packages would help. maybe need a different algorithm (at least optionally).

@chriscoey chriscoey added help wanted Extra attention is needed speed labels Aug 5, 2018
@chriscoey chriscoey added this to the 0.1 milestone Aug 6, 2018
@chriscoey
Copy link
Collaborator Author

another idea is to take log of the point values and then take combinations of them multiplied by the powers, and finally take exp of all these values.

@chriscoey
Copy link
Collaborator Author

JuliaLinearAlgebra/ToeplitzMatrices.jl#38 might help with some aspect of the interpolations (possibly after construction though)

@chriscoey
Copy link
Collaborator Author

note JuliaLang/julia#29846 (comment) QR issues

@chriscoey
Copy link
Collaborator Author

closing due to #110 improving memory usage significantly. worth looking into whether there is a more efficient algorithm for doing the QR of the generalized (for multivariate polynomials) vandermonde matrix transpose.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
help wanted Extra attention is needed speed
Development

No branches or pull requests

1 participant