Learning BN as probabilistic classifier

Learning a Bayesian network can be used to obtain a classifier for one of the nodes of the model. For more about classifier, see pyAgrum.skbn.

Creative Commons License

aGrUM

interactive online version

In [1]:
import sys
import os

import numpy as np

import pyAgrum as gum
import pyAgrum.lib.notebook as gnb

%matplotlib inline
from pyAgrum.lib.bn2roc import showROC
from pyAgrum.lib.bn2roc import showPR
from pyAgrum.lib.bn2roc import showROC_PR

SIZE_LEARN=10000
SIZE_VALID=2000
In [2]:
bn=gum.loadBN("res/alarm.dsl")
bn
Out[2]:
G VENTLUNG VENTLUNG EXPCO2 EXPCO2 VENTLUNG->EXPCO2 VENTALV VENTALV VENTLUNG->VENTALV MINVOL MINVOL VENTLUNG->MINVOL ARTCO2 ARTCO2 CATECHOL CATECHOL ARTCO2->CATECHOL ARTCO2->EXPCO2 HRBP HRBP PULMEMBOLUS PULMEMBOLUS SHUNT SHUNT PULMEMBOLUS->SHUNT PAP PAP PULMEMBOLUS->PAP HR HR HR->HRBP CO CO HR->CO HRSAT HRSAT HR->HRSAT HREKG HREKG HR->HREKG VENTTUBE VENTTUBE VENTTUBE->VENTLUNG PRESS PRESS VENTTUBE->PRESS PCWP PCWP VENTMACH VENTMACH VENTMACH->VENTTUBE SAO2 SAO2 SAO2->CATECHOL ERRCAUTER ERRCAUTER ERRCAUTER->HRSAT ERRCAUTER->HREKG STROKEVOLUME STROKEVOLUME STROKEVOLUME->CO DISCONNECT DISCONNECT DISCONNECT->VENTTUBE BP BP CO->BP ANAPHYLAXIS ANAPHYLAXIS TPR TPR ANAPHYLAXIS->TPR ERRLOWOUTPUT ERRLOWOUTPUT ERRLOWOUTPUT->HRBP INTUBATION INTUBATION INTUBATION->VENTLUNG INTUBATION->SHUNT INTUBATION->VENTALV INTUBATION->PRESS INTUBATION->MINVOL HISTORY HISTORY HYPOVOLEMIA HYPOVOLEMIA HYPOVOLEMIA->STROKEVOLUME LVEDVOLUME LVEDVOLUME HYPOVOLEMIA->LVEDVOLUME MINVOLSET MINVOLSET MINVOLSET->VENTMACH PVSAT PVSAT PVSAT->SAO2 INSUFFANESTH INSUFFANESTH INSUFFANESTH->CATECHOL LVEDVOLUME->PCWP CVP CVP LVEDVOLUME->CVP FIO2 FIO2 FIO2->PVSAT KINKEDTUBE KINKEDTUBE KINKEDTUBE->VENTLUNG KINKEDTUBE->PRESS CATECHOL->HR SHUNT->SAO2 VENTALV->ARTCO2 VENTALV->PVSAT TPR->BP TPR->CATECHOL LVFAILURE LVFAILURE LVFAILURE->STROKEVOLUME LVFAILURE->HISTORY LVFAILURE->LVEDVOLUME
In [3]:
gum.generateSample(bn,SIZE_LEARN,"out/learn.csv",show_progress=True,with_labels=True)
gum.generateSample(bn,SIZE_VALID,"out/train.csv",show_progress=True,with_labels=True)
out/learn.csv: 100%|███████████████████████████████████████|
Log2-Likelihood : -152179.3188844174
out/train.csv: 100%|███████████████████████████████████████|
Log2-Likelihood : -29946.558051794076

Out[3]:
-29946.558051794076

Learning a BN from learn.csv

In [4]:
# Learning a BN from the database
learner=gum.BNLearner("out/train.csv")

bn2=learner.useMIIC().learnBN()
currentTime=learner.currentTime()
In [5]:
gnb.flow.add(gnb.getBN(bn2,size="9"),f"Learned with {SIZE_LEARN} lines in {currentTime:.3f}s")
gnb.flow.display()

G VENTLUNG VENTLUNG EXPCO2 EXPCO2 VENTLUNG->EXPCO2 VENTALV VENTALV VENTLUNG->VENTALV ARTCO2 ARTCO2 ARTCO2->VENTALV HRBP HRBP PULMEMBOLUS PULMEMBOLUS PAP PAP PULMEMBOLUS->PAP HR HR HR->HRBP CO CO HR->CO HRSAT HRSAT HR->HRSAT HREKG HREKG HR->HREKG VENTTUBE VENTTUBE VENTTUBE->VENTLUNG PRESS PRESS VENTTUBE->PRESS PCWP PCWP VENTMACH VENTMACH VENTMACH->VENTTUBE SAO2 SAO2 CATECHOL CATECHOL SAO2->CATECHOL ERRCAUTER ERRCAUTER ERRCAUTER->HRSAT ERRCAUTER->HREKG STROKEVOLUME STROKEVOLUME STROKEVOLUME->CO BP BP DISCONNECT DISCONNECT DISCONNECT->VENTTUBE CO->BP ANAPHYLAXIS ANAPHYLAXIS TPR TPR ANAPHYLAXIS->TPR HISTORY HISTORY LVFAILURE LVFAILURE HISTORY->LVFAILURE INTUBATION INTUBATION SHUNT SHUNT INTUBATION->SHUNT INTUBATION->PRESS ERRLOWOUTPUT ERRLOWOUTPUT ERRLOWOUTPUT->HRBP HYPOVOLEMIA HYPOVOLEMIA HYPOVOLEMIA->STROKEVOLUME LVEDVOLUME LVEDVOLUME HYPOVOLEMIA->LVEDVOLUME MINVOLSET MINVOLSET MINVOLSET->VENTMACH PVSAT PVSAT PVSAT->SAO2 LVEDVOLUME->PCWP CVP CVP LVEDVOLUME->CVP INSUFFANESTH INSUFFANESTH FIO2 FIO2 FIO2->PVSAT KINKEDTUBE KINKEDTUBE KINKEDTUBE->VENTLUNG KINKEDTUBE->PRESS CATECHOL->HR SHUNT->PULMEMBOLUS SHUNT->SAO2 EXPCO2->ARTCO2 VENTALV->INTUBATION VENTALV->PVSAT MINVOL MINVOL VENTALV->MINVOL HREKG->HRSAT TPR->BP LVFAILURE->STROKEVOLUME LVFAILURE->LVEDVOLUME
Learned with 10000 lines in 1.057s
In [6]:
import pyAgrum.lib.bn_vs_bn as bnvsbn
gnb.flow.add(gnb.getBNDiff(bn,bn2,size="8!"),"Diff with MIIC")
gnb.flow.add(bnvsbn.graphDiffLegend())
gnb.flow.display()
G KINKEDTUBE KINKEDTUBE VENTLUNG VENTLUNG KINKEDTUBE->VENTLUNG PRESS PRESS KINKEDTUBE->PRESS HYPOVOLEMIA HYPOVOLEMIA STROKEVOLUME STROKEVOLUME HYPOVOLEMIA->STROKEVOLUME LVEDVOLUME LVEDVOLUME HYPOVOLEMIA->LVEDVOLUME INTUBATION INTUBATION SHUNT SHUNT INTUBATION->SHUNT INTUBATION->VENTLUNG MINVOL MINVOL INTUBATION->MINVOL INTUBATION->PRESS VENTALV VENTALV MINVOLSET MINVOLSET VENTMACH VENTMACH MINVOLSET->VENTMACH PULMEMBOLUS PULMEMBOLUS PAP PAP PULMEMBOLUS->PAP INSUFFANESTH INSUFFANESTH CATECHOL CATECHOL INSUFFANESTH->CATECHOL ERRLOWOUTPUT ERRLOWOUTPUT HRBP HRBP ERRLOWOUTPUT->HRBP ERRCAUTER ERRCAUTER HRSAT HRSAT ERRCAUTER->HRSAT HREKG HREKG ERRCAUTER->HREKG FIO2 FIO2 PVSAT PVSAT FIO2->PVSAT LVFAILURE LVFAILURE LVFAILURE->STROKEVOLUME LVFAILURE->LVEDVOLUME HISTORY HISTORY DISCONNECT DISCONNECT VENTTUBE VENTTUBE DISCONNECT->VENTTUBE ANAPHYLAXIS ANAPHYLAXIS TPR TPR ANAPHYLAXIS->TPR CO CO STROKEVOLUME->CO TPR->CATECHOL BP BP TPR->BP PCWP PCWP LVEDVOLUME->PCWP CVP CVP LVEDVOLUME->CVP VENTMACH->VENTTUBE SHUNT->PULMEMBOLUS SAO2 SAO2 SHUNT->SAO2 HISTORY->LVFAILURE VENTTUBE->VENTLUNG VENTTUBE->PRESS VENTLUNG->MINVOL VENTLUNG->VENTALV EXPCO2 EXPCO2 VENTLUNG->EXPCO2 VENTALV->INTUBATION VENTALV->MINVOL ARTCO2 ARTCO2 VENTALV->PVSAT ARTCO2->VENTALV ARTCO2->CATECHOL PVSAT->SAO2 SAO2->CATECHOL EXPCO2->ARTCO2 HR HR CATECHOL->HR HR->HRBP HR->HRSAT HR->CO HR->HREKG CO->BP HREKG->HRSAT
Diff with MIIC
G a->b overflow c->d Missing e->f reversed g->h Correct
In [7]:
bn3=learner.useGreedyHillClimbing().useNMLCorrection().useScoreBDeu().learnBN()
gnb.flow.add(gnb.getBNDiff(bn,bn3,size="8!"),"Diff with GHC/NMD/BDEU")
gnb.flow.add(bnvsbn.graphDiffLegend())
gnb.flow.display()
G KINKEDTUBE KINKEDTUBE VENTLUNG VENTLUNG PRESS PRESS KINKEDTUBE->PRESS HYPOVOLEMIA HYPOVOLEMIA LVFAILURE LVFAILURE HYPOVOLEMIA->LVFAILURE STROKEVOLUME STROKEVOLUME HYPOVOLEMIA->STROKEVOLUME LVEDVOLUME LVEDVOLUME INTUBATION INTUBATION SHUNT SHUNT INTUBATION->SHUNT INTUBATION->VENTLUNG MINVOL MINVOL INTUBATION->MINVOL INTUBATION->PRESS VENTALV VENTALV INTUBATION->VENTALV MINVOLSET MINVOLSET VENTMACH VENTMACH PULMEMBOLUS PULMEMBOLUS PAP PAP PULMEMBOLUS->PAP INSUFFANESTH INSUFFANESTH CATECHOL CATECHOL INSUFFANESTH->CATECHOL ERRLOWOUTPUT ERRLOWOUTPUT HRBP HRBP ERRCAUTER ERRCAUTER HRSAT HRSAT ERRCAUTER->HRSAT HREKG HREKG FIO2 FIO2 PVSAT PVSAT FIO2->PVSAT HISTORY HISTORY LVFAILURE->HISTORY DISCONNECT DISCONNECT VENTTUBE VENTTUBE DISCONNECT->VENTTUBE ANAPHYLAXIS ANAPHYLAXIS TPR TPR STROKEVOLUME->LVFAILURE HR HR STROKEVOLUME->HR CO CO STROKEVOLUME->CO TPR->ANAPHYLAXIS BP BP TPR->BP LVEDVOLUME->HYPOVOLEMIA LVEDVOLUME->LVFAILURE LVEDVOLUME->STROKEVOLUME PCWP PCWP LVEDVOLUME->PCWP CVP CVP LVEDVOLUME->CVP VENTMACH->MINVOLSET VENTMACH->VENTTUBE SHUNT->KINKEDTUBE SHUNT->PULMEMBOLUS SAO2 SAO2 SHUNT->SAO2 VENTTUBE->KINKEDTUBE VENTTUBE->VENTLUNG VENTTUBE->PRESS VENTLUNG->KINKEDTUBE VENTLUNG->MINVOL VENTLUNG->VENTALV EXPCO2 EXPCO2 VENTLUNG->EXPCO2 ARTCO2 ARTCO2 VENTALV->ARTCO2 VENTALV->PVSAT ARTCO2->EXPCO2 ARTCO2->CATECHOL PVSAT->SAO2 SAO2->CATECHOL SAO2->HR SAO2->HRBP CATECHOL->TPR HR->ERRLOWOUTPUT HR->ERRCAUTER HR->CATECHOL HR->HRSAT HR->HREKG HRBP->ERRLOWOUTPUT HRBP->HR HRBP->CO CO->HR CO->BP HREKG->ERRCAUTER
Diff with GHC/NMD/BDEU
G a->b overflow c->d Missing e->f reversed g->h Correct
In [8]:
bn4=learner.useGreedyHillClimbing().useNMLCorrection().useScoreBDeu().setInitialDAG(bn2.dag()).learnBN()
gnb.flow.add(gnb.getBNDiff(bn,bn4,size="8!"),"Diff with GHC/NMD/BDEU with intial DAG from MIIC")
gnb.flow.add(bnvsbn.graphDiffLegend())
gnb.flow.display()
G KINKEDTUBE KINKEDTUBE VENTLUNG VENTLUNG KINKEDTUBE->VENTLUNG PRESS PRESS KINKEDTUBE->PRESS HYPOVOLEMIA HYPOVOLEMIA STROKEVOLUME STROKEVOLUME HYPOVOLEMIA->STROKEVOLUME LVEDVOLUME LVEDVOLUME HYPOVOLEMIA->LVEDVOLUME INTUBATION INTUBATION SHUNT SHUNT INTUBATION->SHUNT INTUBATION->VENTLUNG MINVOL MINVOL INTUBATION->MINVOL INTUBATION->PRESS VENTALV VENTALV INTUBATION->VENTALV MINVOLSET MINVOLSET VENTMACH VENTMACH MINVOLSET->VENTMACH PULMEMBOLUS PULMEMBOLUS PAP PAP PULMEMBOLUS->PAP INSUFFANESTH INSUFFANESTH CATECHOL CATECHOL INSUFFANESTH->CATECHOL ERRLOWOUTPUT ERRLOWOUTPUT HRBP HRBP ERRLOWOUTPUT->HRBP ERRCAUTER ERRCAUTER HRSAT HRSAT ERRCAUTER->HRSAT HREKG HREKG ERRCAUTER->HREKG FIO2 FIO2 PVSAT PVSAT FIO2->PVSAT LVFAILURE LVFAILURE LVFAILURE->STROKEVOLUME LVFAILURE->LVEDVOLUME HISTORY HISTORY DISCONNECT DISCONNECT VENTTUBE VENTTUBE DISCONNECT->VENTTUBE ANAPHYLAXIS ANAPHYLAXIS TPR TPR CO CO STROKEVOLUME->CO TPR->ANAPHYLAXIS BP BP TPR->BP PCWP PCWP LVEDVOLUME->PCWP CVP CVP LVEDVOLUME->CVP VENTMACH->VENTTUBE SHUNT->PULMEMBOLUS SAO2 SAO2 SHUNT->SAO2 HISTORY->LVFAILURE VENTTUBE->VENTLUNG VENTTUBE->PRESS VENTLUNG->MINVOL VENTLUNG->VENTALV EXPCO2 EXPCO2 VENTLUNG->EXPCO2 ARTCO2 ARTCO2 VENTALV->ARTCO2 VENTALV->PVSAT ARTCO2->EXPCO2 ARTCO2->CATECHOL PVSAT->SAO2 SAO2->CATECHOL CATECHOL->TPR HR HR CATECHOL->HR HR->HRBP HR->HRSAT HR->CO HR->HREKG CO->BP
Diff with GHC/NMD/BDEU with intial DAG from MIIC
G a->b overflow c->d Missing e->f reversed g->h Correct
In [9]:
print(bn2.names())
{'VENTLUNG', 'ARTCO2', 'HRBP', 'PULMEMBOLUS', 'HR', 'VENTTUBE', 'PCWP', 'VENTMACH', 'SAO2', 'ERRCAUTER', 'STROKEVOLUME', 'BP', 'DISCONNECT', 'CO', 'ANAPHYLAXIS', 'HISTORY', 'INTUBATION', 'ERRLOWOUTPUT', 'HYPOVOLEMIA', 'MINVOLSET', 'PVSAT', 'LVEDVOLUME', 'INSUFFANESTH', 'FIO2', 'HRSAT', 'KINKEDTUBE', 'CATECHOL', 'SHUNT', 'EXPCO2', 'VENTALV', 'CVP', 'HREKG', 'PRESS', 'PAP', 'TPR', 'MINVOL', 'LVFAILURE'}
In [10]:
gnb.showInference(bn2,evs={},size="14")
../_images/notebooks_32-Learning_learningClassifier_13_0.svg

Two classifiers from the learned BN

In [11]:
print(bn2.variableFromName("HRSAT"))
print(bn2.variableFromName("INTUBATION"))
HRSAT:Labelized({HIGH|LOW|NORMAL})
INTUBATION:Labelized({ESOPHAGEAL|NORMAL|ONESIDED})
In [12]:
showROC(bn2,"out/train.csv",'HRSAT','LOW',show_progress=False)
showROC(bn2,"out/train.csv",'HRSAT','NORMAL',show_progress=False)
showROC(bn2,"out/train.csv",'HRSAT','HIGH',show_progress=False);
../_images/notebooks_32-Learning_learningClassifier_16_0.svg
../_images/notebooks_32-Learning_learningClassifier_16_1.svg
../_images/notebooks_32-Learning_learningClassifier_16_2.svg
In [13]:
showROC(bn2,"out/train.csv",'INTUBATION',"ESOPHAGEAL",show_progress=False);
../_images/notebooks_32-Learning_learningClassifier_17_0.svg
In [14]:
showPR(bn2,"out/train.csv",'HRSAT','LOW',show_progress=False);
../_images/notebooks_32-Learning_learningClassifier_18_0.svg
In [15]:
showPR(bn2,"out/train.csv",'INTUBATION',"ESOPHAGEAL",show_progress=False);
../_images/notebooks_32-Learning_learningClassifier_19_0.svg
In [16]:
showROC_PR(bn2,"out/train.csv",'HRSAT','LOW',show_progress=False);
../_images/notebooks_32-Learning_learningClassifier_20_0.svg
In [17]:
showROC_PR(bn2,"out/train.csv",'INTUBATION',"ESOPHAGEAL",show_progress=False);
../_images/notebooks_32-Learning_learningClassifier_21_0.svg
In [ ]: