def __init__(self, df, depth = 3, significance = 0.05, verbose = False, priorKnowledge = None):
tetradData = pycausal.loadDiscreteData(df)
indTest = javabridge.JClassWrapper('edu.cmu.tetrad.search.IndTestChiSquare')(tetradData, significance)
cpc = javabridge.JClassWrapper('edu.cmu.tetrad.search.Cpc')(indTest)
cpc.setDepth(depth)
cpc.setVerbose(verbose)
if priorKnowledge is not None:
cpc.setKnowledge(priorKnowledge)
tetradGraph = cpc.search()
dags = javabridge.JClassWrapper('edu.cmu.tetrad.search.DagInPatternIterator')(tetradGraph)
dagGraph = dags.next()
dag = javabridge.JClassWrapper('edu.cmu.tetrad.graph.Dag')(dagGraph)
pm = javabridge.JClassWrapper('edu.cmu.tetrad.bayes.BayesPm')(dag)
est = javabridge.JClassWrapper('edu.cmu.tetrad.bayes.MlBayesEstimator')()
im = est.estimate(pm, tetradData)
self.nodes = pycausal.extractTetradGraphNodes(dag)
self.edges = pycausal.extractTetradGraphEdges(dag)
self.graph = pycausal.generatePyDotGraph(self.nodes,self.edges)
self.dag = dag
self.bayesPm = pm
self.bayesIm = im
def __init__(self,
df,
penaltydiscount=4,
maxDegree=3,
faithfulnessAssumed=True,
numofthreads=2,
verbose=False,
priorKnowledge=None):
tetradData = pycausal.loadContinuousData(df)
score = javabridge.JClassWrapper('edu.cmu.tetrad.search.SemBicScore')(
tetradData)
score.setPenaltyDiscount(
penaltydiscount
) # set to 2 if variable# <= 50 otherwise set it to 4
fgs = javabridge.JClassWrapper('edu.cmu.tetrad.search.Fgs')(score)
fgs.setMaxDegree(maxDegree)
fgs.setNumPatternsToStore(0)
fgs.setFaithfulnessAssumed(faithfulnessAssumed)
fgs.setParallelism(numofthreads)
fgs.setVerbose(verbose)
if priorKnowledge is not None:
fgs.setKnowledge(priorKnowledge)
tetradGraph = fgs.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
self.graph = pycausal.generatePyDotGraph(self.nodes, self.edges)
def __init__(self,
df,
depth=3,
significance=0.05,
verbose=False,
priorKnowledge=None):
tetradData = pycausal.loadDiscreteData(df)
indTest = javabridge.JClassWrapper(
'edu.cmu.tetrad.search.IndTestChiSquare')(tetradData, significance)
cpc = javabridge.JClassWrapper('edu.cmu.tetrad.search.Cpc')(indTest)
cpc.setDepth(depth)
cpc.setVerbose(verbose)
if priorKnowledge is not None:
cpc.setKnowledge(priorKnowledge)
tetradGraph = cpc.search()
dags = javabridge.JClassWrapper(
'edu.cmu.tetrad.search.DagInPatternIterator')(tetradGraph)
dagGraph = dags.next()
dag = javabridge.JClassWrapper('edu.cmu.tetrad.graph.Dag')(dagGraph)
pm = javabridge.JClassWrapper('edu.cmu.tetrad.bayes.BayesPm')(dag)
est = javabridge.JClassWrapper(
'edu.cmu.tetrad.bayes.MlBayesEstimator')()
im = est.estimate(pm, tetradData)
self.nodes = pycausal.extractTetradGraphNodes(dag)
self.edges = pycausal.extractTetradGraphEdges(dag)
self.graph = pycausal.generatePyDotGraph(self.nodes, self.edges)
self.dag = dag
self.bayesPm = pm
self.bayesIm = im
def __init__(self,
df,
continuous=True,
depth=3,
aggressivelyPreventCycles=False,
significance=0.05,
verbose=False,
priorKnowledge=None):
indTest = None
if (continuous):
tetradData = pycausal.loadContinuousData(df)
indTest = javabridge.JClassWrapper(
'edu.cmu.tetrad.search.IndTestFisherZ')(tetradData,
significance)
else:
tetradData = pycausal.loadDiscreteData(df)
indTest = javabridge.JClassWrapper(
'edu.cmu.tetrad.search.IndTestChiSquare')(tetradData,
significance)
cpcstable = javabridge.JClassWrapper(
'edu.cmu.tetrad.search.CpcStable')(indTest)
cpcstable.setDepth(depth)
cpcstable.setAggressivelyPreventCycles(aggressivelyPreventCycles)
cpcstable.setVerbose(verbose)
if priorKnowledge is not None:
cpcstable.setKnowledge(priorKnowledge)
tetradGraph = cpcstable.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
self.graph = pycausal.generatePyDotGraph(self.nodes, self.edges)
def __init__(self,
df,
continuous=True,
depth=3,
significance=0.05,
verbose=False,
priorKnowledge=None):
indTest = None
if (continuous):
tetradData = pycausal.loadContinuousData(df)
indTest = javabridge.JClassWrapper(
'edu.cmu.tetrad.search.IndTestFisherZ')(tetradData,
significance)
else:
tetradData = pycausal.loadDiscreteData(df)
indTest = javabridge.JClassWrapper(
'edu.cmu.tetrad.search.IndTestChiSquare')(tetradData,
significance)
ccd = javabridge.JClassWrapper('edu.cmu.tetrad.search.Ccd')(indTest)
ccd.setDepth(depth)
ccd.setVerbose(verbose)
if priorKnowledge is not None:
ccd.setKnowledge(priorKnowledge)
tetradGraph = ccd.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
def __init__(self,
df,
structurePrior=1.0,
samplePrior=1.0,
maxDegree=3,
faithfulnessAssumed=True,
numofthreads=2,
verbose=False,
priorKnowledge=None):
tetradData = pycausal.loadDiscreteData(df)
score = javabridge.JClassWrapper('edu.cmu.tetrad.search.BDeuScore')(
tetradData)
score.setStructurePrior(structurePrior)
score.setSamplePrior(samplePrior)
fgs = javabridge.JClassWrapper('edu.cmu.tetrad.search.Fgs')(score)
fgs.setMaxDegree(maxDegree)
fgs.setNumPatternsToStore(0)
fgs.setFaithfulnessAssumed(faithfulnessAssumed)
fgs.setParallelism(numofthreads)
fgs.setVerbose(verbose)
if priorKnowledge is not None:
fgs.setKnowledge(priorKnowledge)
tetradGraph = fgs.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
self.graph = pycausal.generatePyDotGraph(self.nodes, self.edges)
def __init__(self, seed = None, numNodes = 10, numEdges = 10):
if seed is not None:
RandomUtil = javabridge.static_call("edu/cmu/tetrad/util/RandomUtil","getInstance","()Ledu/cmu/tetrad/util/RandomUtil;")
javabridge.call(RandomUtil, "setSeed", "(J)V", seed)
dag = None
initEdges = -1
while initEdges < numEdges:
graph = javabridge.static_call("edu/cmu/tetrad/graph/GraphUtils","randomGraph","(IIIIIIZ)Ledu/cmu/tetrad/graph/Graph;",numNodes,0,numEdges,30,15,15,False)
dag = javabridge.JClassWrapper("edu.cmu.tetrad.graph.Dag")(graph)
initEdges = dag.getNumEdges()
self.nodes = pycausal.extractTetradGraphNodes(dag)
self.edges = pycausal.extractTetradGraphEdges(dag)
self.graph = pycausal.generatePyDotGraph(self.nodes,self.edges)
self.dag = dag
def __init__(self, df, continuous = True, depth = 3, significance = 0.05, verbose = False, priorKnowledge = None):
indTest = None
if(continuous):
tetradData = pycausal.loadContinuousData(df)
indTest = javabridge.JClassWrapper('edu.cmu.tetrad.search.IndTestFisherZ')(tetradData, significance)
else:
tetradData = pycausal.loadDiscreteData(df)
indTest = javabridge.JClassWrapper('edu.cmu.tetrad.search.IndTestChiSquare')(tetradData, significance)
ccd = javabridge.JClassWrapper('edu.cmu.tetrad.search.Ccd')(indTest)
ccd.setDepth(depth)
ccd.setVerbose(verbose)
if priorKnowledge is not None:
ccd.setKnowledge(priorKnowledge)
tetradGraph = ccd.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
def __init__(self, seed=None, numNodes=10, numEdges=10):
if seed is not None:
RandomUtil = javabridge.static_call(
"edu/cmu/tetrad/util/RandomUtil", "getInstance",
"()Ledu/cmu/tetrad/util/RandomUtil;")
javabridge.call(RandomUtil, "setSeed", "(J)V", seed)
dag = None
initEdges = -1
while initEdges < numEdges:
graph = javabridge.static_call(
"edu/cmu/tetrad/graph/GraphUtils", "randomGraph",
"(IIIIIIZ)Ledu/cmu/tetrad/graph/Graph;", numNodes, 0, numEdges,
30, 15, 15, False)
dag = javabridge.JClassWrapper("edu.cmu.tetrad.graph.Dag")(graph)
initEdges = dag.getNumEdges()
self.nodes = pycausal.extractTetradGraphNodes(dag)
self.edges = pycausal.extractTetradGraphEdges(dag)
self.graph = pycausal.generatePyDotGraph(self.nodes, self.edges)
self.dag = dag
def __init__(self, df, penaltydiscount = 4, maxDegree = 3, faithfulnessAssumed = True, numofthreads = 2, verbose = False, priorKnowledge = None):
tetradData = pycausal.loadContinuousData(df)
score = javabridge.JClassWrapper('edu.cmu.tetrad.search.SemBicScore')(tetradData)
score.setPenaltyDiscount(penaltydiscount) # set to 2 if variable# <= 50 otherwise set it to 4
fges = javabridge.JClassWrapper('edu.cmu.tetrad.search.Fgs')(score)
fges.setMaxDegree(maxDegree)
fges.setNumPatternsToStore(0)
fges.setFaithfulnessAssumed(faithfulnessAssumed)
fges.setParallelism(numofthreads)
fges.setVerbose(verbose)
if priorKnowledge is not None:
fges.setKnowledge(priorKnowledge)
tetradGraph = fges.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
self.graph = pycausal.generatePyDotGraph(self.nodes,self.edges)
def __init__(self, df, penaltydiscount = 2, maxInDegree = 3, maxPathLength = -1, significance = 0.05, completeRuleSetUsed = False, faithfulnessAssumed = True, verbose = False, priorKnowledge = None):
tetradData = pycausal.loadContinuousData(df)
indTest = javabridge.JClassWrapper('edu.cmu.tetrad.search.IndTestFisherZ')(tetradData, significance)
score = javabridge.JClassWrapper('edu.cmu.tetrad.search.SemBicScore')(tetradData)
score.setPenaltyDiscount(penaltydiscount) # set to 2 if variable# <= 50 otherwise set it to 4
gfci = javabridge.JClassWrapper('edu.cmu.tetrad.search.GFci')(indTest, score)
gfci.setMaxIndegree(maxInDegree)
gfci.setMaxPathLength(maxPathLength)
gfci.setCompleteRuleSetUsed(completeRuleSetUsed)
gfci.setFaithfulnessAssumed(faithfulnessAssumed)
gfci.setVerbose(verbose)
if priorKnowledge is not None:
gfci.setKnowledge(priorKnowledge)
tetradGraph = gfci.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
def __init__(self, df, continuous = True, depth = 3, aggressivelyPreventCycles = False, significance = 0.05, verbose = False, priorKnowledge = None):
indTest = None
if(continuous):
tetradData = pycausal.loadContinuousData(df)
indTest = javabridge.JClassWrapper('edu.cmu.tetrad.search.IndTestFisherZ')(tetradData, significance)
else:
tetradData = pycausal.loadDiscreteData(df)
indTest = javabridge.JClassWrapper('edu.cmu.tetrad.search.IndTestChiSquare')(tetradData, significance)
cpcstable = javabridge.JClassWrapper('edu.cmu.tetrad.search.CpcStable')(indTest)
cpcstable.setDepth(depth)
cpcstable.setAggressivelyPreventCycles(aggressivelyPreventCycles)
cpcstable.setVerbose(verbose)
if priorKnowledge is not None:
cpcstable.setKnowledge(priorKnowledge)
tetradGraph = cpcstable.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
self.graph = pycausal.generatePyDotGraph(self.nodes,self.edges)
def __init__(self, df, structurePrior = 1.0, samplePrior = 1.0, maxDegree = 3, faithfulnessAssumed = True, numofthreads = 2, verbose = False, priorKnowledge = None):
tetradData = pycausal.loadDiscreteData(df)
score = javabridge.JClassWrapper('edu.cmu.tetrad.search.BDeuScore')(tetradData)
score.setStructurePrior(structurePrior)
score.setSamplePrior(samplePrior)
fges = javabridge.JClassWrapper('edu.cmu.tetrad.search.Fgs')(score)
fges.setMaxDegree(maxDegree)
fges.setNumPatternsToStore(0)
fges.setFaithfulnessAssumed(faithfulnessAssumed)
fges.setParallelism(numofthreads)
fges.setVerbose(verbose)
if priorKnowledge is not None:
fges.setKnowledge(priorKnowledge)
tetradGraph = fges.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
self.graph = pycausal.generatePyDotGraph(self.nodes,self.edges)
def __init__(self, df, structurePrior = 1.0, samplePrior = 1.0, maxInDegree = 3, maxPathLength = -1, significance = 0.05, completeRuleSetUsed = False, faithfulnessAssumed = True, verbose = False, priorKnowledge = None):
tetradData = pycausal.loadDiscreteData(df)
indTest = javabridge.JClassWrapper('edu.cmu.tetrad.search.IndTestChiSquare')(tetradData, significance)
score = javabridge.JClassWrapper('edu.cmu.tetrad.search.BDeuScore')(tetradData)
score.setStructurePrior(structurePrior)
score.setSamplePrior(samplePrior)
gfci = javabridge.JClassWrapper('edu.cmu.tetrad.search.GFci')(indTest, score)
gfci.setMaxIndegree(maxInDegree)
gfci.setMaxPathLength(maxPathLength)
gfci.setCompleteRuleSetUsed(completeRuleSetUsed)
gfci.setFaithfulnessAssumed(faithfulnessAssumed)
gfci.setVerbose(verbose)
if priorKnowledge is not None:
gfci.setKnowledge(priorKnowledge)
tetradGraph = gfci.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
def __init__(self,
df,
penaltydiscount=2,
maxInDegree=3,
maxPathLength=-1,
significance=0.05,
completeRuleSetUsed=False,
faithfulnessAssumed=True,
verbose=False,
priorKnowledge=None):
tetradData = pycausal.loadContinuousData(df)
indTest = javabridge.JClassWrapper(
'edu.cmu.tetrad.search.IndTestFisherZ')(tetradData, significance)
score = javabridge.JClassWrapper('edu.cmu.tetrad.search.SemBicScore')(
tetradData)
score.setPenaltyDiscount(
penaltydiscount
) # set to 2 if variable# <= 50 otherwise set it to 4
gfci = javabridge.JClassWrapper('edu.cmu.tetrad.search.GFci')(indTest,
score)
gfci.setMaxIndegree(maxInDegree)
gfci.setMaxPathLength(maxPathLength)
gfci.setCompleteRuleSetUsed(completeRuleSetUsed)
gfci.setFaithfulnessAssumed(faithfulnessAssumed)
gfci.setVerbose(verbose)
if priorKnowledge is not None:
gfci.setKnowledge(priorKnowledge)
tetradGraph = gfci.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
def __init__(self,
df,
structurePrior=1.0,
samplePrior=1.0,
maxInDegree=3,
maxPathLength=-1,
significance=0.05,
completeRuleSetUsed=False,
faithfulnessAssumed=True,
verbose=False,
priorKnowledge=None):
tetradData = pycausal.loadDiscreteData(df)
indTest = javabridge.JClassWrapper(
'edu.cmu.tetrad.search.IndTestChiSquare')(tetradData, significance)
score = javabridge.JClassWrapper('edu.cmu.tetrad.search.BDeuScore')(
tetradData)
score.setStructurePrior(structurePrior)
score.setSamplePrior(samplePrior)
gfci = javabridge.JClassWrapper('edu.cmu.tetrad.search.GFci')(indTest,
score)
gfci.setMaxIndegree(maxInDegree)
gfci.setMaxPathLength(maxPathLength)
gfci.setCompleteRuleSetUsed(completeRuleSetUsed)
gfci.setFaithfulnessAssumed(faithfulnessAssumed)
gfci.setVerbose(verbose)
if priorKnowledge is not None:
gfci.setKnowledge(priorKnowledge)
tetradGraph = gfci.search()
self.nodes = pycausal.extractTetradGraphNodes(tetradGraph)
self.edges = pycausal.extractTetradGraphEdges(tetradGraph)
