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

Double-check units on the "covariance is too big" warning #658

Open
dmcdougall opened this issue Feb 21, 2018 · 1 comment
Open

Double-check units on the "covariance is too big" warning #658

dmcdougall opened this issue Feb 21, 2018 · 1 comment
Assignees

Comments

@dmcdougall
Copy link
Member

No description provided.

@briadam
Copy link
Contributor

briadam commented Feb 23, 2018

The conditional logic for this warning

if (covMatrix(i, i) >= max_val - min_val) {
// User is trying to specify a uniform proposal distribution, which
// is unsupported. Throw an error for now.
std::cerr << "Proposal variance element "
<< i
<< " is "
<< covMatrix(i, i)
<< " but domain is of size "
<< max_val - min_val
<< std::endl;
std::cerr << "QUESO does not support uniform-like proposal "
<< "distributions. Try making the proposal variance smaller"
<< std::endl;
}

is comparing units of variance (covariance diagonal) to domain bounds. Should probably instead by comparing sqrt(var) to the bounds.

@briadam briadam self-assigned this Feb 23, 2018
briadam added a commit to briadam/queso that referenced this issue Feb 23, 2018
Address libqueso#658 by using sqrt(Covariance(i,i)) to compare
standard deviation to to domain bounds.
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

No branches or pull requests

2 participants