-
Notifications
You must be signed in to change notification settings - Fork 39
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
Bayesian Opt input structure fix #356
Bayesian Opt input structure fix #356
Conversation
growth/shrink factors in GradientDescent now toggleable by users in heron input
src/Cases.py
Outdated
@@ -545,6 +545,13 @@ def _specs_optimizer(cls): | |||
Bayesian Optimizer.""", default='ExpectedImprovement') | |||
bayesian_opt.addSub(acquisitionSub) | |||
|
|||
#======== Kernel for BO========# | |||
seedSub = InputData.parameterInputFactory('initialSeed', contentType=InputTypes.IntegerType, |
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'm fine with initialSeed
; if at some point you want to change this to seed
while making other changes, I slightly prefer 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.
eh, might as well change it now. it'll be seed
for the HERON input, but I think it has to still be initialSeed
in the outer.xml unless we modify the RAVEN input structure
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.
sure, that's fine. We intentionally split HERON syntax from RAVEN syntax most of the time, since much of the purpose of the HERON input was to make it more accessible than the RAVEN running RAVEN inputs that Aaron originally wrote. This is not a big deal, but I like the cleanness of the change. Thanks!
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.
Just one commented line to consider.
src/Cases.py
Outdated
modelSelection.addSub(modelMethod) | ||
optimizer.addSub(modelSelection) | ||
|
||
# optimizer.addSub(modelSelection) |
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.
do we want to keep this line?
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.
Nope! will delete
Checklist passes, merging. We've had enough little changes as well as big fixes like this one, it might be time for a submodule update in RAVEN; @GabrielSoto-INL would you mind opening one? |
Pull Request Description
What issue does this change request address?
#350 , #355
What are the significant changes in functionality due to this change request?
Fixes a previous issue in
template_driver.py
where if the user requests usage of theBayesianOpt
optimization algorithm/strategy, it still writes aGradientDescent
node into theouter.xml
.There are modifications to the XML input node under the
Cases -> optimization_settings -> algorithm
node. The<algorithm>
node accepts subnodes rather than a string for either<GradientDescent>
or<BayesianOpt>
. Each now has algorithm-specific settings that can be modified by the user (all available in the HERON user manual). Default is still<BayesianOpt>
.Some features available in the
outer.xml
for<GradientDescent>
are now able to be modified in the heron input script as described in [TASK] Access GradientDescent Features in HERON input script #355.Updated some of the scripts for checking XMLs; regolding 'optimization_settings' test to change from Gradient Descent to Bayesian Opt results (larger tolerance, shorter duration)
For Change Control Board: Change Request Review
The following review must be completed by an authorized member of the Change Control Board.