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.

Sign up for GitHub

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

Jump to bottom

Fix defect ensemble model (with Code) and genetic algorithm #2317

Merged
merged 19 commits into from
May 16, 2024
Merged

Fix defect ensemble model (with Code) and genetic algorithm #2317

Show file tree
Hide file tree
Changes from 6 commits
Commits
Show all changes
19 commits
Select commit Hold shift + click to select a range
a0ea3b5
Closes #2304
alfoa May 8, 2024
e9feed6
added test
alfoa May 8, 2024
336bae3
Update tests/framework/Optimizers/GeneticAlgorithms/continuous/uncons…
alfoa May 8, 2024
da32086
removed trailing
alfoa May 8, 2024
70a5354
Merge branch 'alfoa/genetic_ensemble_model' of github.com:idaholab/ra…
alfoa May 8, 2024
6077f56
added missing input file
alfoa May 8, 2024
c7f27bf
Update tests/framework/Optimizers/GeneticAlgorithms/continuous/uncons…
alfoa May 8, 2024
af23869
isinstance
alfoa May 9, 2024
ee2f53f
fixed parallel execution
alfoa May 9, 2024
7fba807
readded heron and baycal?
alfoa May 9, 2024
9f077b8
removed trailing spaces
alfoa May 9, 2024
ab60b73
some cleaning
alfoa May 9, 2024
54f43a4
fixed test name and rook error message
alfoa May 9, 2024
a522493
module ensemble doc
alfoa May 10, 2024
5856f6b
Apply suggestions from code review
alfoa May 13, 2024
f83fb30
removed trailing whitespaces
alfoa May 13, 2024
91a23d9
Merge branch 'devel' into alfoa/genetic_ensemble_model
alfoa May 13, 2024
d3c7069
fix for 'cvxpy'...this is not a very maintainable way to handle such …
alfoa May 13, 2024
7ff7659
changed approach
alfoa May 13, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
12 changes: 10 additions & 2 deletions ravenframework/JobHandler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -702,7 +702,7 @@ def reAddJob(self, runner):
runner.trackTime('queue')
self.__submittedJobs.append(runner.identifier)

def addClientJob(self, args, functionToRun, identifier, metadata=None, uniqueHandler="any"):
def addClientJob(self, args, functionToRun, identifier, metadata=None, uniqueHandler="any", groupInfo = None):
"""
Method to add an internal run (function execution), without consuming
resources (free spots). This can be used for client handling (see
Expand All @@ -719,11 +719,19 @@ def addClientJob(self, args, functionToRun, identifier, metadata=None, uniqueHan
this runner. For example, if present, to retrieve this runner using the
method jobHandler.getFinished, the uniqueHandler needs to be provided.
If uniqueHandler == 'any', every "client" can get this runner.
@ In, groupInfo, dict, optional, {id:string, size:int}.
- "id": it is a special keyword attached to
this runner to identify that this runner belongs to a special set of runs that need to be
grouped together (all will be retrievable only when all the runs ended).
- "size", number of runs in this group self.__batching
NOTE: If the "size" of the group is only set the first time a job of this group is added.
Consequentially the size is immutable
@ Out, None
"""
self.addJob(args, functionToRun, identifier, metadata,
forceUseThreads = True, uniqueHandler = uniqueHandler,
clientQueue = True)
clientQueue = True, groupInfo = groupInfo)


def addFinishedJob(self, data, metadata=None, uniqueHandler="any", profile=False):
"""
Expand Down
9 changes: 8 additions & 1 deletion ravenframework/Models/EnsembleModel.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -563,7 +563,14 @@ def submit(self,myInput,samplerType,jobHandler,**kwargs):
uniqueHandler=uniqueHandler, forceUseThreads=forceThreads,
groupInfo={'id': kwargs['batchInfo']['batchId'], 'size': nRuns} if batchMode else None)
else:
jobHandler.addClientJob((self, myInput, samplerType, kwargs), self.__class__.evaluateSample, prefix, kwargs)
# for parallel strategy 2, the ensemble model works as a step => it needs the jobHandler
kw['jobHandler'] = jobHandler
# for parallel strategy 2, we need to make sure that the batchMode is set to False in the inner runs since only the
# ensemble model evaluation should be batched (THIS IS REQUIRED because the CODE does not submit runs like the other models)
kw['batchMode'] = False
jobHandler.addClientJob((self, myInput, samplerType, kw), self.__class__.evaluateSample, prefix, metadata=metadata,
uniqueHandler=uniqueHandler,
groupInfo={'id': kwargs['batchInfo']['batchId'], 'size': nRuns} if batchMode else None)

def __retrieveDependentOutput(self,modelIn,listOfOutputs, typeOutputs):
"""
Expand Down
16 changes: 16 additions & 0 deletions ...ticAlgorithms/continuous/unconstrained/metaModelWithCodeAndFunctionsAndGenetic/ABCDsum.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,16 @@
# Copyright 2017 Battelle Energy Alliance, LLC
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

def run(self, Input):
self.sum = self.A + self.B + self.C + self.D
6 changes: 6 additions & 0 deletions ...rithms/continuous/unconstrained/metaModelWithCodeAndFunctionsAndGenetic/decayConstantB.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
# Copyright Nucube Energy, Inc.

def evaluate(self):
return self.decay_A+0.0001


alfoa marked this conversation as resolved.
Show resolved Hide resolved
39 changes: 39 additions & 0 deletions ...ithms/continuous/unconstrained/metaModelWithCodeAndFunctionsAndGenetic/referenceInput.xml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,39 @@
<?xml version="1.0" ?>
<AnalyticalBateman>
<totalTime>300</totalTime>
<powerHistory>1 1 1</powerHistory>
<flux>1e14 1e14 1e14</flux>
<stepDays>0 100 200 400</stepDays>
<timeSteps>100 100 100</timeSteps>
<nuclides>
<A>
<equationType>N1</equationType>
<initialMass>1.0</initialMass>
<decayConstant>$RAVEN-decay_A|10$</decayConstant>
<sigma>$RAVEN-sigma-A|10$</sigma>
<ANumber>230</ANumber>
</A>
<B>
<equationType>N2</equationType>
<initialMass>1.0</initialMass>
<decayConstant>$RAVEN-decay_B:0.000000006$</decayConstant>
<sigma>$RAVEN-sigma-B:5$</sigma>
<ANumber>200</ANumber>
</B>
<C>
<equationType>N3</equationType>
<initialMass>1.0</initialMass>
<decayConstant>$RAVEN-decay-C:0.000000008$</decayConstant>
<sigma>$RAVEN-sigma-C:3$</sigma>
<ANumber>150</ANumber>
</C>
<D>
<equationType>N4</equationType>
<initialMass>1.0</initialMass>
<decayConstant>$RAVEN-decay-D:0.000000009$</decayConstant>
<sigma>$RAVEN-sigma-D:1$</sigma>
<ANumber>100</ANumber>
</D>
</nuclides>
</AnalyticalBateman>

181 changes: 181 additions & 0 deletions ...neticAlgorithms/continuous/unconstrained/test_ensembleModel_withGA_Code_and_Functions.xml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,181 @@
<?xml version="1.0" ?>
<Simulation verbosity="debug">
<TestInfo>
<name>framework/Optimizers/GeneticAlgorithms.GAwithEnsembleModelIncludingCode</name>
<author>alfoa</author>
<created>2024-05-07</created>
<classesTested>Models.EnsembleModel, Optimizers.GeneticAlgorithm</classesTested>
<description>
This test is aimed to test the usage of the GeneticAlgorithm with an EnsembleModel, including
a model Code. This test has been created to test the modifications performed to address issue
#2304 (see \\url{https://github.com/idaholab/raven/issues/2304}).
</description>
</TestInfo>

<RunInfo>
<JobName>metaModelWithCodeAndFunctionsAndGenetic</JobName>
<Sequence>
optimize
</Sequence>
<WorkingDir>metaModelWithCodeAndFunctionsAndGenetic</WorkingDir>
<batchSize>1</batchSize>
Copy link
Collaborator

Choose a reason for hiding this comment

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

Have you tested it with different batch size?

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

ops. Not yet. let me test it

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

@wangcj05 I fixed the parallel execution and converted the test into a parallel (batchSize > 1) to test it.

Ready for re-review

</RunInfo>

<Files>
<Input name="referenceInput.xml" type="input">referenceInput.xml</Input>
</Files>

<Models>
<Code name="testModel" subType="GenericCode">
<executable>../../../../user_guide/physicalCode/analyticalbateman/AnalyticalDplMain.py</executable>
<clargs arg="python" type="prepend"/>
<clargs arg="" extension=".xml" type="input"/>
<clargs arg=" " extension=".csv" type="output"/>
<prepend>python</prepend>
</Code>
<ExternalModel ModuleToLoad="ABCDsum" name="sumOfABCD" subType="">
<inputs>A,B,C,D</inputs>
<outputs>sum</outputs>
</ExternalModel>
<EnsembleModel name="codeAndExtModel" subType="">
<Model class="Models" type="ExternalModel">
sumOfABCD
<Input class="DataObjects" type="PointSet">inputPlaceHolderForInternalObjects</Input>
<TargetEvaluation class="DataObjects" type="PointSet">sumData</TargetEvaluation>
</Model>
<Model class="Models" type="Code">
testModel
<Input class="Files" type="">referenceInput.xml</Input>
<TargetEvaluation class="DataObjects" type="PointSet">samplesMC</TargetEvaluation>
</Model>
</EnsembleModel>
</Models>

<Distributions>
<Uniform name="sigma">
<lowerBound>0</lowerBound>
<upperBound>1000</upperBound>
</Uniform>
<Uniform name="decayConstant">
<lowerBound>0.00000001</lowerBound>
<upperBound>0.0000001</upperBound>
</Uniform>
</Distributions>

<Optimizers>
<GeneticAlgorithm name="GAopt">
<samplerInit>
<limit>20</limit>
<initialSeed>42</initialSeed>
<writeSteps>every</writeSteps>
</samplerInit>

<GAparams>
<populationSize>2</populationSize>
<parentSelection>rouletteWheel</parentSelection>
<reproduction>
<crossover type="onePointCrossover">
<crossoverProb>0.8</crossoverProb>
</crossover>
<mutation type="swapMutator">
<mutationProb>0.9</mutationProb>
</mutation>
</reproduction>
<fitness type="feasibleFirst"></fitness>
<survivorSelection>fitnessBased</survivorSelection>
</GAparams>

<convergence>
<AHDp>0.0</AHDp>
</convergence>

<variable name="sigma-A">
<distribution>sigma</distribution>
</variable>
<variable name="decay_A">
<distribution>decayConstant</distribution>
</variable>
<variable name="sigma-B">
<distribution>sigma</distribution>
</variable>
<variable name="decay_B">
<function>decayConstantB</function>
</variable>

<objective>sum</objective>
<TargetEvaluation class="DataObjects" type="PointSet">finalResponses</TargetEvaluation>
<Sampler class="Samplers" type="MonteCarlo">MC_samp</Sampler>
</GeneticAlgorithm>
</Optimizers>

<Functions>
<External file="decayConstantB.py" name="decayConstantB">
<variables>decay_A</variables>
</External>
</Functions>

<Samplers>
<MonteCarlo name="MC_samp">
<samplerInit>
<limit>2</limit>
<initialSeed>20021986</initialSeed>
</samplerInit>
<variable name="sigma-A">
<distribution>sigma</distribution>
</variable>
<variable name="decay_A">
<distribution>decayConstant</distribution>
</variable>
<variable name="sigma-B">
<distribution>sigma</distribution>
</variable>
<variable name="decay_B">
<function>decayConstantB</function>
</variable>
</MonteCarlo>
</Samplers>

<Steps>
<MultiRun name="optimize">
<Input class="Files" type="">referenceInput.xml</Input>
<Input class="DataObjects" type="PointSet">inputPlaceHolderForInternalObjects</Input>
<Model class="Models" type="EnsembleModel">codeAndExtModel</Model>
<Optimizer class="Optimizers" type="GeneticAlgorithm">GAopt</Optimizer>
<SolutionExport class="DataObjects" type="PointSet">opt_export</SolutionExport>
<Output class="DataObjects" type="PointSet">finalResponses</Output>
<Output class="OutStreams" type="Print">opt_export</Output>
</MultiRun>
</Steps>

<OutStreams>
<Print name="opt_export">
<type>csv</type>
<source>opt_export</source>
<what>input,output</what>
</Print>
</OutStreams>

<DataObjects>
<PointSet name="samplesMC">
<Input>sigma-A,sigma-B,decay_A,decay_B</Input>
<Output>A,B,C,D</Output>
</PointSet>
<PointSet name="inputPlaceHolderForInternalObjects">
<Input>A,B,C,D</Input>
<Output>OutputPlaceHolder</Output>
</PointSet>
<PointSet name="sumData">
<Input>A,B,C,D</Input>
<Output>sum</Output>
</PointSet>
<PointSet name="finalResponses">
<Input>sigma-A,sigma-B,decay_A,decay_B</Input>
<Output>A,B,C,D,sum</Output>
</PointSet>
<PointSet name="opt_export">
<Input>trajID</Input>
<Output>sigma-A,sigma-B,decay_A,decay_B,sum,age,batchId,fitness,iteration,accepted,AHDp,conv_AHDp</Output>
</PointSet>
</DataObjects>

</Simulation>
39 changes: 39 additions & 0 deletions ...thms/gold/continuous/unconstrained/metaModelWithCodeAndFunctionsAndGenetic/opt_export.csv
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,39 @@
trajID,sigma-A,sigma-B,decay_A,decay_B,sum,age,batchId,fitness,iteration,accepted,AHDp,conv_AHDp
0,812.997460787,464.508307042,1.255650268e-08,0.000100012556503,3.15760954156,0,1,-3.15760954156,0,first,,False
0,759.507928453,912.545952926,8.14640085566e-08,0.000100081464009,2.90476176926,0,1,-2.90476176926,0,first,,False
0,912.545952926,812.997460787,1.255650268e-08,0.000100012556503,3.15873736407,0,2,-3.15873736407,1,accepted,299.428022512,False
0,464.508307042,759.507928453,8.14640085566e-08,0.000100081464009,2.83929690449,0,2,-2.83929690449,1,accepted,299.428022512,False
0,794.044539517,912.545952926,5.18057476338e-08,0.000100051805748,2.98393998882,1,3,-2.98393998882,2,accepted,261.653404312,False
0,794.044539517,759.507928453,7.83557135608e-08,0.000100078355714,2.9174761674,1,3,-2.9174761674,2,accepted,261.653404312,False
0,912.545952926,464.508307042,8.14640085566e-08,0.000100081464009,2.93136915288,2,4,-2.93136915288,3,accepted,379.964041465,False
0,759.507928453,759.507928453,8.14640085566e-08,0.000100081464009,2.90491165751,2,4,-2.90491165751,3,accepted,379.964041465,False
0,464.508307042,912.545952926,8.14640085566e-08,0.000100081464009,2.83915821533,3,5,-2.83915821533,4,accepted,154.98621892,False
0,794.044539517,759.507928453,7.83557135608e-08,0.000100078355714,2.9174761674,3,5,-2.9174761674,4,accepted,154.98621892,False
0,912.545952926,464.508307042,8.14640085566e-08,0.000100081464009,2.93136915288,3,6,-2.93136915288,5,accepted,484.155218746,False
0,759.507928453,464.508307042,8.14640085566e-08,0.000100081464009,2.90520071484,3,6,-2.90520071484,5,accepted,484.155218746,False
0,912.545952926,464.508307042,8.14640085566e-08,0.000100081464009,2.93136915288,4,7,-2.93136915288,6,accepted,484.155218746,False
0,759.507928453,464.508307042,8.14640085566e-08,0.000100081464009,2.90520071484,4,7,-2.90520071484,6,accepted,484.155218746,False
0,794.044539517,759.507928453,5.18057476338e-08,0.000100051805748,2.98410104597,5,8,-2.98410104597,7,accepted,329.536232476,False
0,794.044539517,912.545952926,5.18057476338e-08,0.000100051805748,2.98393998882,5,8,-2.98393998882,7,accepted,329.536232476,False
0,794.044539517,759.507928453,5.18057476338e-08,0.000100051805748,2.98410104597,6,9,-2.98410104597,8,accepted,261.553081047,False
0,464.508307042,912.545952926,8.14640085566e-08,0.000100081464009,2.83915821533,6,9,-2.83915821533,8,accepted,261.553081047,False
0,794.044539517,912.545952926,5.18057476338e-08,0.000100051805748,2.98393998882,4,10,-2.98393998882,9,accepted,329.536232476,False
0,794.044539517,912.545952926,5.18057476338e-08,0.000100051805748,2.98393998882,4,10,-2.98393998882,9,accepted,329.536232476,False
0,464.508307042,794.044539517,9.21291357633e-08,0.000100092129136,2.81925891819,5,11,-2.81925891819,10,accepted,118.501413408,False
0,464.508307042,794.044539517,9.21291357633e-08,0.000100092129136,2.81925891819,5,11,-2.81925891819,10,accepted,118.501413408,False
0,464.508307042,912.545952926,8.14640085566e-08,0.000100081464009,2.83915821533,0,12,-2.83915821533,11,accepted,118.501413408,False
0,464.508307042,912.545952926,8.14640085566e-08,0.000100081464009,2.83915821533,0,12,-2.83915821533,11,accepted,118.501413408,False
0,464.508307042,794.044539517,9.21291357633e-08,0.000100092129136,2.81925891819,1,13,-2.81925891819,12,accepted,94.054634152,False
0,464.508307042,912.545952926,8.14640085566e-08,0.000100081464009,2.83915821533,1,13,-2.83915821533,12,accepted,94.054634152,False
0,912.545952926,794.044539517,5.18057476338e-08,0.000100051805748,3.00110960171,2,14,-3.00110960171,13,accepted,448.037645884,False
0,912.545952926,794.044539517,5.18057476338e-08,0.000100051805748,3.00110960171,2,14,-3.00110960171,13,accepted,448.037645884,False
0,912.545952926,794.044539517,5.18057476338e-08,0.000100051805748,3.00110960171,3,15,-3.00110960171,14,accepted,448.037645884,False
0,912.545952926,794.044539517,5.18057476338e-08,0.000100051805748,3.00110960171,3,15,-3.00110960171,14,accepted,448.037645884,False
0,794.044539517,464.508307042,9.21291357633e-08,0.000100092129136,2.89183632446,4,16,-2.89183632446,15,accepted,369.891914497,False
0,464.508307042,794.044539517,9.21291357633e-08,0.000100092129136,2.81925891819,4,16,-2.81925891819,15,accepted,369.891914497,False
0,794.044539517,464.508307042,9.21291357633e-08,0.000100092129136,2.89183632446,3,17,-2.89183632446,16,accepted,466.03460926,False
0,794.044539517,464.508307042,9.21291357633e-08,0.000100092129136,2.89183632446,3,17,-2.89183632446,16,accepted,466.03460926,False
0,794.044539517,464.508307042,9.21291357633e-08,0.000100092129136,2.89183632446,4,18,-2.89183632446,17,accepted,466.03460926,False
0,794.044539517,464.508307042,9.21291357633e-08,0.000100092129136,2.89183632446,4,18,-2.89183632446,17,accepted,466.03460926,False
0,464.508307042,912.545952926,8.14640085566e-08,0.000100081464009,2.83915821533,5,19,-2.83915821533,18,accepted,94.054634152,False
0,464.508307042,794.044539517,9.21291357633e-08,0.000100092129136,2.81925891819,5,19,-2.81925891819,18,accepted,94.054634152,False
10 changes: 10 additions & 0 deletions tests/framework/Optimizers/GeneticAlgorithms/tests
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -384,4 +384,14 @@
rel_err = 0.001
[../]
[../]

[./GAwithEnsembleModelIncludingCode]
type = 'RavenFramework'
input = 'continuous/unconstrained/test_ensembleModel_withGA_Code_and_Functions.xml'
[./csv]
type = OrderedCSV
output = 'continuous/unconstrained/metaModelWithCodeAndFunctionsAndGenetic/opt_export.csv'
rel_err = 0.001
[../]
[../]
[]