-
Notifications
You must be signed in to change notification settings - Fork 40
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
Adding small GPMSA verification example #614
Conversation
Codecov Report
@@ Coverage Diff @@
## dev #614 +/- ##
=========================================
- Coverage 75.1% 75.07% -0.03%
=========================================
Files 310 311 +1
Lines 23288 23528 +240
=========================================
+ Hits 17490 17664 +174
- Misses 5798 5864 +66
Continue to review full report at Codecov.
|
7eb3e0d
to
6f3a169
Compare
This test checks that the pdf (both prior and likelihood) evaluations between matlab and queso are consistent (to within 1e-5).
6f3a169
to
f937c27
Compare
|
||
# ----- | ||
# GPMSA Options with data precision adjusted per Brian Williams | ||
# ----- |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Don't a few of these just match the default options?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes. I left them there just to be explicit.
test/test_gpmsa/scalar_pdf_small.C
Outdated
// | ||
// Regarding experimental data, the user should transformed it so that it has | ||
// zero mean and variance one. | ||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These comments are now outdated, right?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I removed them.
test/test_gpmsa/scalar_pdf_small.C
Outdated
for (unsigned int j = 0; j < point.sizeLocal(); ++j) | ||
solution_data >> point[j]; | ||
|
||
double log_likelihood; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why all the intermediate variables?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Addressed.
test/test_gpmsa/scalar_pdf_small.C
Outdated
} | ||
solution_data.close(); | ||
|
||
double initial_diff_prior = expected_log_priors[0] - computed_log_priors[0]; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is because our log priors are correct only up to a constant, but we expect it to be the same constant at every evaluations? We should have a comment about that.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I forgot to address this comment in this PR. I did it in #615, though.
10ef1b6
to
48fa850
Compare
Should be good to do. |
Labelled as 'do not merge' until we iron out a potential issue with the normalisation of the scenario variables.