Processing /mnt/build/workspace/root-makedoc-v608/rootspi/rdoc/src/v6-08-00-patches/tutorials/tmva/TMVAMultipleBackgroundExample.C...
Start Test TMVAGAexample
========================
... event: 0 (2000)
... event: 1000 (2000)
======> EVENT:0
var1 = -1.14361
var2 = -0.822373
var3 = -0.395426
var4 = -0.529427
created tree: TreeS
... event: 0 (2000)
... event: 1000 (2000)
======> EVENT:0
var1 = -1.54361
var2 = -1.42237
var3 = -1.39543
var4 = -2.02943
created tree: TreeB0
... event: 0 (2000)
... event: 1000 (2000)
======> EVENT:0
var1 = -1.54361
var2 = -0.822373
var3 = -0.395426
var4 = -2.02943
created tree: TreeB1
======> EVENT:0
var1 = 0.463304
var2 = 1.37192
var3 = -1.16769
var4 = -1.77551
created tree: TreeB2
created data file: tmva_example_multiple_background.root
========================
--- Training
<HEADER> DataSetInfo : [datasetBkg0] : Added class "Signal"
: Add Tree TreeS of type Signal with 2000 events
<HEADER> DataSetInfo : [datasetBkg0] : Added class "Background"
: Add Tree TreeB0 of type Background with 2000 events
<HEADER> Factory : Booking method: BDTG
:
: the option *InverseBoostNegWeights* does not exist for BoostType=Grad --> change
: to new default for GradBoost *Pray*
<HEADER> DataSetFactory : [datasetBkg0] : Number of events in input trees
:
:
: Number of training and testing events
: ---------------------------------------------------------------------------
: Signal -- training events : 1000
: Signal -- testing events : 1000
: Signal -- training and testing events: 2000
: Background -- training events : 1000
: Background -- testing events : 1000
: Background -- training and testing events: 2000
:
<HEADER> DataSetInfo : Correlation matrix (Signal):
: ----------------------------------------
: var1 var2 var3 var4
: var1: +1.000 +0.380 +0.597 +0.819
: var2: +0.380 +1.000 +0.706 +0.744
: var3: +0.597 +0.706 +1.000 +0.853
: var4: +0.819 +0.744 +0.853 +1.000
: ----------------------------------------
<HEADER> DataSetInfo : Correlation matrix (Background):
: ----------------------------------------
: var1 var2 var3 var4
: var1: +1.000 +0.406 +0.621 +0.837
: var2: +0.406 +1.000 +0.696 +0.727
: var3: +0.621 +0.696 +1.000 +0.853
: var4: +0.837 +0.727 +0.853 +1.000
: ----------------------------------------
<HEADER> DataSetFactory : [datasetBkg0] :
:
<HEADER> Factory : Train all methods
<HEADER> Factory : [datasetBkg0] : Create Transformation "I" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg0] : Create Transformation "D" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg0] : Create Transformation "P" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg0] : Create Transformation "G" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg0] : Create Transformation "D" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: -0.0033185 1.0233 [ -4.0592 3.5808 ]
: var2: 0.010677 1.0686 [ -3.6891 3.7877 ]
: var3: -0.018220 1.1329 [ -3.6148 4.5640 ]
: var4: 0.14122 1.3261 [ -4.8486 5.0412 ]
: -----------------------------------------------------------
: Preparing the Decorrelation transformation...
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: -0.10624 1.0000 [ -3.3999 2.8598 ]
: var2: -0.051494 1.0000 [ -3.7359 3.5041 ]
: var3: -0.14575 1.0000 [ -3.4614 3.4198 ]
: var4: 0.28275 1.0000 [ -3.2217 2.9133 ]
: -----------------------------------------------------------
: Preparing the Principle Component (PCA) transformation...
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: 1.0532e-09 2.0470 [ -7.1768 7.9217 ]
: var2: 1.8738e-10 0.80759 [ -2.8767 2.4675 ]
: var3: 2.1503e-10 0.52103 [ -1.5564 1.7353 ]
: var4:-4.1175e-11 0.34211 [ -1.0644 1.1223 ]
: -----------------------------------------------------------
: Preparing the Gaussian transformation...
: Preparing the Decorrelation transformation...
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: 0.023076 1.0000 [ -2.7354 6.4668 ]
: var2: 0.021818 1.0000 [ -2.9124 6.8653 ]
: var3: 0.012974 1.0000 [ -3.1214 6.1526 ]
: var4: 0.0065905 1.0000 [ -2.6929 8.7981 ]
: -----------------------------------------------------------
: Ranking input variables (method unspecific)...
<HEADER> IdTransformation : Ranking result (top variable is best ranked)
: -----------------------------------
: Rank : Variable : Separation
: -----------------------------------
: 1 : Variable 4 : 3.265e-01
: 2 : Variable 3 : 1.925e-01
: 3 : Variable 2 : 1.007e-01
: 4 : Variable 1 : 4.412e-02
: -----------------------------------
<HEADER> Factory : Train method: BDTG for Classification
:
<HEADER> BDTG : #events: (reweighted) sig: 1000 bkg: 1000
: #events: (unweighted) sig: 1000 bkg: 1000
: Training 1000 Decision Trees ... patience please
: Elapsed time for training with 2000 events: 1.19 sec
<HEADER> BDTG : [datasetBkg0] : Evaluation of BDTG on training sample (2000 events)
: Elapsed time for evaluation of 2000 events: 0.294 sec
: Creating xml weight file: datasetBkg0/weights/TMVAMultiBkg0_BDTG.weights.xml
: Creating standalone class: datasetBkg0/weights/TMVAMultiBkg0_BDTG.class.C
: TMVASignalBackground0.root:/datasetBkg0/Method_BDTG/BDTG
<HEADER> Factory : Training finished
:
: Ranking input variables (method specific)...
<HEADER> BDTG : Ranking result (top variable is best ranked)
: --------------------------------------
: Rank : Variable : Variable Importance
: --------------------------------------
: 1 : var1 : 2.706e-01
: 2 : var2 : 2.594e-01
: 3 : var3 : 2.369e-01
: 4 : var4 : 2.330e-01
: --------------------------------------
<HEADER> Factory : === Destroy and recreate all methods via weight files for testing ===
:
<HEADER> Factory : Test all methods
<HEADER> Factory : Test method: BDTG for Classification performance
:
<HEADER> BDTG : [datasetBkg0] : Evaluation of BDTG on testing sample (2000 events)
: Elapsed time for evaluation of 2000 events: 0.248 sec
<HEADER> Factory : Evaluate all methods
<HEADER> Factory : Evaluate classifier: BDTG
:
<HEADER> BDTG : [datasetBkg0] : Loop over test events and fill histograms with classifier response...
:
<HEADER> TFHandler_BDTG : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: 0.024055 1.0671 [ -3.6592 3.2749 ]
: var2: -0.0060579 1.0733 [ -3.6952 3.2955 ]
: var3: -0.0091467 1.1473 [ -4.5727 3.9796 ]
: var4: 0.14407 1.3591 [ -4.7970 4.2221 ]
: -----------------------------------------------------------
:
: Evaluation results ranked by best signal efficiency and purity (area)
: -------------------------------------------------------------------------------------------------------------------
: DataSet MVA
: Name: Method: ROC-integ
: datasetBkg0 BDTG : 0.945
: -------------------------------------------------------------------------------------------------------------------
:
: Testing efficiency compared to training efficiency (overtraining check)
: -------------------------------------------------------------------------------------------------------------------
: DataSet MVA Signal efficiency: from test sample (from training sample)
: Name: Method: @B=0.01 @B=0.10 @B=0.30
: -------------------------------------------------------------------------------------------------------------------
: datasetBkg0 BDTG : 0.345 (0.745) 0.855 (0.917) 0.961 (0.967)
: -------------------------------------------------------------------------------------------------------------------
:
<HEADER> Dataset:datasetBkg0 : Created tree 'TestTree' with 2000 events
:
<HEADER> Dataset:datasetBkg0 : Created tree 'TrainTree' with 2000 events
:
<HEADER> Factory : Thank you for using TMVA!
: For citation information, please visit: http://tmva.sf.net/citeTMVA.html
<HEADER> DataSetInfo : [datasetBkg1] : Added class "Signal"
: Add Tree TreeS of type Signal with 2000 events
<HEADER> DataSetInfo : [datasetBkg1] : Added class "Background"
: Add Tree TreeB1 of type Background with 2000 events
<HEADER> Factory : Booking method: BDTG
:
: the option *InverseBoostNegWeights* does not exist for BoostType=Grad --> change
: to new default for GradBoost *Pray*
<HEADER> DataSetFactory : [datasetBkg1] : Number of events in input trees
:
:
: Number of training and testing events
: ---------------------------------------------------------------------------
: Signal -- training events : 1000
: Signal -- testing events : 1000
: Signal -- training and testing events: 2000
: Background -- training events : 1000
: Background -- testing events : 1000
: Background -- training and testing events: 2000
:
<HEADER> DataSetInfo : Correlation matrix (Signal):
: ----------------------------------------
: var1 var2 var3 var4
: var1: +1.000 +0.380 +0.597 +0.819
: var2: +0.380 +1.000 +0.706 +0.744
: var3: +0.597 +0.706 +1.000 +0.853
: var4: +0.819 +0.744 +0.853 +1.000
: ----------------------------------------
<HEADER> DataSetInfo : Correlation matrix (Background):
: ----------------------------------------
: var1 var2 var3 var4
: var1: +1.000 +0.406 +0.621 +0.837
: var2: +0.406 +1.000 +0.696 +0.727
: var3: +0.621 +0.696 +1.000 +0.853
: var4: +0.837 +0.727 +0.853 +1.000
: ----------------------------------------
<HEADER> DataSetFactory : [datasetBkg1] :
:
<HEADER> Factory : Train all methods
<HEADER> Factory : [datasetBkg1] : Create Transformation "I" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg1] : Create Transformation "D" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg1] : Create Transformation "P" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg1] : Create Transformation "G" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg1] : Create Transformation "D" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: -0.0033185 1.0233 [ -4.0592 3.5808 ]
: var2: 0.31068 1.0243 [ -3.0891 3.7877 ]
: var3: 0.48178 1.0287 [ -2.6800 4.5640 ]
: var4: 0.14122 1.3261 [ -4.8486 5.0412 ]
: -----------------------------------------------------------
: Preparing the Decorrelation transformation...
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: -0.15694 1.0000 [ -3.6768 2.9345 ]
: var2: 0.16935 1.0000 [ -3.6322 3.8796 ]
: var3: 0.54902 1.0000 [ -2.9826 3.9034 ]
: var4: -0.036260 1.0000 [ -3.2175 2.7306 ]
: -----------------------------------------------------------
: Preparing the Principle Component (PCA) transformation...
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1:-1.2426e-09 1.9131 [ -6.8557 7.4957 ]
: var2:-8.5571e-10 0.84772 [ -2.9068 2.6663 ]
: var3:-4.4057e-10 0.56544 [ -1.8166 1.9083 ]
: var4: 3.3708e-10 0.46340 [ -1.4760 1.4296 ]
: -----------------------------------------------------------
: Preparing the Gaussian transformation...
: Preparing the Decorrelation transformation...
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: 0.020335 1.0000 [ -2.8168 6.5920 ]
: var2: 0.012545 1.0000 [ -3.1166 7.0214 ]
: var3: 0.018226 1.0000 [ -3.1675 6.8889 ]
: var4: 0.012229 1.0000 [ -2.1606 6.7157 ]
: -----------------------------------------------------------
: Ranking input variables (method unspecific)...
<HEADER> IdTransformation : Ranking result (top variable is best ranked)
: -----------------------------------
: Rank : Variable : Separation
: -----------------------------------
: 1 : Variable 4 : 3.265e-01
: 2 : Variable 1 : 4.412e-02
: 3 : Variable 2 : 1.180e-02
: 4 : Variable 3 : 9.585e-03
: -----------------------------------
<HEADER> Factory : Train method: BDTG for Classification
:
<HEADER> BDTG : #events: (reweighted) sig: 1000 bkg: 1000
: #events: (unweighted) sig: 1000 bkg: 1000
: Training 1000 Decision Trees ... patience please
: Elapsed time for training with 2000 events: 1.11 sec
<HEADER> BDTG : [datasetBkg1] : Evaluation of BDTG on training sample (2000 events)
: Elapsed time for evaluation of 2000 events: 0.286 sec
: Creating xml weight file: datasetBkg1/weights/TMVAMultiBkg1_BDTG.weights.xml
: Creating standalone class: datasetBkg1/weights/TMVAMultiBkg1_BDTG.class.C
: TMVASignalBackground1.root:/datasetBkg1/Method_BDTG/BDTG
<HEADER> Factory : Training finished
:
: Ranking input variables (method specific)...
<HEADER> BDTG : Ranking result (top variable is best ranked)
: --------------------------------------
: Rank : Variable : Variable Importance
: --------------------------------------
: 1 : var1 : 2.627e-01
: 2 : var3 : 2.590e-01
: 3 : var4 : 2.393e-01
: 4 : var2 : 2.391e-01
: --------------------------------------
<HEADER> Factory : === Destroy and recreate all methods via weight files for testing ===
:
<HEADER> Factory : Test all methods
<HEADER> Factory : Test method: BDTG for Classification performance
:
<HEADER> BDTG : [datasetBkg1] : Evaluation of BDTG on testing sample (2000 events)
: Elapsed time for evaluation of 2000 events: 0.17 sec
<HEADER> Factory : Evaluate all methods
<HEADER> Factory : Evaluate classifier: BDTG
:
<HEADER> BDTG : [datasetBkg1] : Loop over test events and fill histograms with classifier response...
:
<HEADER> TFHandler_BDTG : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: 0.024055 1.0671 [ -3.6592 3.2749 ]
: var2: 0.29394 1.0317 [ -3.0952 3.7877 ]
: var3: 0.49085 1.0206 [ -3.5727 4.5640 ]
: var4: 0.14407 1.3591 [ -4.7970 4.2221 ]
: -----------------------------------------------------------
:
: Evaluation results ranked by best signal efficiency and purity (area)
: -------------------------------------------------------------------------------------------------------------------
: DataSet MVA
: Name: Method: ROC-integ
: datasetBkg1 BDTG : 0.995
: -------------------------------------------------------------------------------------------------------------------
:
: Testing efficiency compared to training efficiency (overtraining check)
: -------------------------------------------------------------------------------------------------------------------
: DataSet MVA Signal efficiency: from test sample (from training sample)
: Name: Method: @B=0.01 @B=0.10 @B=0.30
: -------------------------------------------------------------------------------------------------------------------
: datasetBkg1 BDTG : 0.935 (0.975) 0.996 (0.994) 1.000 (1.000)
: -------------------------------------------------------------------------------------------------------------------
:
<HEADER> Dataset:datasetBkg1 : Created tree 'TestTree' with 2000 events
:
<HEADER> Dataset:datasetBkg1 : Created tree 'TrainTree' with 2000 events
:
<HEADER> Factory : Thank you for using TMVA!
: For citation information, please visit: http://tmva.sf.net/citeTMVA.html
<HEADER> DataSetInfo : [datasetBkg2] : Added class "Signal"
: Add Tree TreeS of type Signal with 2000 events
<HEADER> DataSetInfo : [datasetBkg2] : Added class "Background"
: Add Tree TreeB2 of type Background with 2000 events
<HEADER> Factory : Booking method: BDTG
:
: the option *InverseBoostNegWeights* does not exist for BoostType=Grad --> change
: to new default for GradBoost *Pray*
<HEADER> DataSetFactory : [datasetBkg2] : Number of events in input trees
:
:
: Number of training and testing events
: ---------------------------------------------------------------------------
: Signal -- training events : 1000
: Signal -- testing events : 1000
: Signal -- training and testing events: 2000
: Background -- training events : 1000
: Background -- testing events : 1000
: Background -- training and testing events: 2000
:
<HEADER> DataSetInfo : Correlation matrix (Signal):
: ----------------------------------------
: var1 var2 var3 var4
: var1: +1.000 +0.380 +0.597 +0.819
: var2: +0.380 +1.000 +0.706 +0.744
: var3: +0.597 +0.706 +1.000 +0.853
: var4: +0.819 +0.744 +0.853 +1.000
: ----------------------------------------
<HEADER> DataSetInfo : Correlation matrix (Background):
: ----------------------------------------
: var1 var2 var3 var4
: var1: +1.000 -0.656 -0.025 -0.016
: var2: -0.656 +1.000 +0.009 +0.030
: var3: -0.025 +0.009 +1.000 -0.050
: var4: -0.016 +0.030 -0.050 +1.000
: ----------------------------------------
<HEADER> DataSetFactory : [datasetBkg2] :
:
<HEADER> Factory : Train all methods
<HEADER> Factory : [datasetBkg2] : Create Transformation "I" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg2] : Create Transformation "D" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg2] : Create Transformation "P" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg2] : Create Transformation "G" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> Factory : [datasetBkg2] : Create Transformation "D" with events from all classes.
:
<HEADER> : Transformation, Variable selection :
: Input : variable 'var1' <---> Output : variable 'var1'
: Input : variable 'var2' <---> Output : variable 'var2'
: Input : variable 'var3' <---> Output : variable 'var3'
: Input : variable 'var4' <---> Output : variable 'var4'
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: 0.29422 0.92562 [ -3.1287 3.5808 ]
: var2: 0.62585 0.95260 [ -3.0891 3.7877 ]
: var3: 0.24297 1.0792 [ -2.6800 4.5640 ]
: var4: 0.45321 1.1847 [ -3.3486 5.0412 ]
: -----------------------------------------------------------
: Preparing the Decorrelation transformation...
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: 0.26528 1.0000 [ -3.1160 3.3765 ]
: var2: 0.62361 1.0000 [ -3.2368 3.5512 ]
: var3: 0.058659 1.0000 [ -2.3761 3.1130 ]
: var4: 0.26989 1.0000 [ -2.3587 3.1806 ]
: -----------------------------------------------------------
: Preparing the Principle Component (PCA) transformation...
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1:-6.0070e-10 1.4738 [ -5.8142 7.2302 ]
: var2: 3.7398e-12 0.93490 [ -2.2584 2.5744 ]
: var3:-9.0618e-10 0.83936 [ -2.6693 2.7596 ]
: var4:-2.4955e-10 0.76279 [ -2.1164 2.1269 ]
: -----------------------------------------------------------
: Preparing the Gaussian transformation...
: Preparing the Decorrelation transformation...
<HEADER> TFHandler_Factory : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: 0.023311 1.0000 [ -2.3372 5.5472 ]
: var2: 0.023695 1.0000 [ -2.6181 5.6037 ]
: var3: 0.016566 1.0000 [ -2.6997 4.9155 ]
: var4: 0.016440 1.0000 [ -3.1160 5.2002 ]
: -----------------------------------------------------------
: Ranking input variables (method unspecific)...
<HEADER> IdTransformation : Ranking result (top variable is best ranked)
: -----------------------------------
: Rank : Variable : Separation
: -----------------------------------
: 1 : Variable 2 : 3.099e-01
: 2 : Variable 4 : 1.650e-01
: 3 : Variable 1 : 1.311e-01
: 4 : Variable 3 : 1.178e-01
: -----------------------------------
<HEADER> Factory : Train method: BDTG for Classification
:
<HEADER> BDTG : #events: (reweighted) sig: 1000 bkg: 1000
: #events: (unweighted) sig: 1000 bkg: 1000
: Training 1000 Decision Trees ... patience please
: Elapsed time for training with 2000 events: 1.08 sec
<HEADER> BDTG : [datasetBkg2] : Evaluation of BDTG on training sample (2000 events)
: Elapsed time for evaluation of 2000 events: 0.257 sec
: Creating xml weight file: datasetBkg2/weights/TMVAMultiBkg2_BDTG.weights.xml
: Creating standalone class: datasetBkg2/weights/TMVAMultiBkg2_BDTG.class.C
: TMVASignalBackground2.root:/datasetBkg2/Method_BDTG/BDTG
<HEADER> Factory : Training finished
:
: Ranking input variables (method specific)...
<HEADER> BDTG : Ranking result (top variable is best ranked)
: --------------------------------------
: Rank : Variable : Variable Importance
: --------------------------------------
: 1 : var4 : 3.036e-01
: 2 : var2 : 2.392e-01
: 3 : var1 : 2.359e-01
: 4 : var3 : 2.214e-01
: --------------------------------------
<HEADER> Factory : === Destroy and recreate all methods via weight files for testing ===
:
<HEADER> Factory : Test all methods
<HEADER> Factory : Test method: BDTG for Classification performance
:
<HEADER> BDTG : [datasetBkg2] : Evaluation of BDTG on testing sample (2000 events)
: Elapsed time for evaluation of 2000 events: 0.171 sec
<HEADER> Factory : Evaluate all methods
<HEADER> Factory : Evaluate classifier: BDTG
:
<HEADER> BDTG : [datasetBkg2] : Loop over test events and fill histograms with classifier response...
:
<HEADER> TFHandler_BDTG : Variable Mean RMS [ Min Max ]
: -----------------------------------------------------------
: var1: 0.32064 0.94395 [ -3.6592 3.2749 ]
: var2: 0.62991 0.92896 [ -3.0952 3.2955 ]
: var3: 0.25730 1.0986 [ -3.5727 3.9796 ]
: var4: 0.46324 1.2003 [ -3.2970 4.2221 ]
: -----------------------------------------------------------
:
: Evaluation results ranked by best signal efficiency and purity (area)
: -------------------------------------------------------------------------------------------------------------------
: DataSet MVA
: Name: Method: ROC-integ
: datasetBkg2 BDTG : 0.980
: -------------------------------------------------------------------------------------------------------------------
:
: Testing efficiency compared to training efficiency (overtraining check)
: -------------------------------------------------------------------------------------------------------------------
: DataSet MVA Signal efficiency: from test sample (from training sample)
: Name: Method: @B=0.01 @B=0.10 @B=0.30
: -------------------------------------------------------------------------------------------------------------------
: datasetBkg2 BDTG : 0.812 (0.936) 0.962 (0.970) 0.998 (0.996)
: -------------------------------------------------------------------------------------------------------------------
:
<HEADER> Dataset:datasetBkg2 : Created tree 'TestTree' with 2000 events
:
<HEADER> Dataset:datasetBkg2 : Created tree 'TrainTree' with 2000 events
:
<HEADER> Factory : Thank you for using TMVA!
: For citation information, please visit: http://tmva.sf.net/citeTMVA.html
========================
--- Application & create combined tree
: Booking "BDT method" of type "BDT" from datasetBkg0/weights/TMVAMultiBkg0_BDTG.weights.xml.
<HEADER> DataSetInfo : [Default] : Added class "Signal"
<HEADER> DataSetInfo : [Default] : Added class "Background"
: Booked classifier "BDTG" of type: "BDT"
: Booking "BDT method" of type "BDT" from datasetBkg1/weights/TMVAMultiBkg1_BDTG.weights.xml.
<HEADER> DataSetInfo : [Default] : Added class "Signal"
<HEADER> DataSetInfo : [Default] : Added class "Background"
: Booked classifier "BDTG" of type: "BDT"
: Booking "BDT method" of type "BDT" from datasetBkg2/weights/TMVAMultiBkg2_BDTG.weights.xml.
<HEADER> DataSetInfo : [Default] : Added class "Signal"
<HEADER> DataSetInfo : [Default] : Added class "Background"
: Booked classifier "BDTG" of type: "BDT"
--- Select signal sample
--- Processing: 2000 events
--- ... Processing event: 0
--- ... Processing event: 1000
--- End of event loop: Real time 0:00:00, CP time 0.500
--- Select background 0 sample
--- Processing: 2000 events
--- ... Processing event: 0
--- ... Processing event: 1000
--- End of event loop: Real time 0:00:00, CP time 0.540
--- Select background 1 sample
--- Processing: 2000 events
--- ... Processing event: 0
--- ... Processing event: 1000
--- End of event loop: Real time 0:00:00, CP time 0.540
--- Select background 2 sample
--- Processing: 2000 events
--- ... Processing event: 0
--- ... Processing event: 1000
--- End of event loop: Real time 0:00:00, CP time 0.530
--- Created root file: "tmva_example_multiple_backgrounds__applied.root" containing the MVA output histograms
==> Application of readers is done! combined tree created
========================
--- maximize significance
Classifier ranges (defined by the user)
range: -1 1
range: -1 1
range: -1 1
<HEADER> FitterBase : <GeneticFitter> Optimisation, please be patient ... (inaccurate progress timing for GA)
: Elapsed time: 62.4 sec
======================
Efficiency : 0.897
Purity : 0.881572
True positive weights : 1794
False positive weights: 241
Signal weights : 2000
cutValue[0] = -0.232085;
cutValue[1] = 0.168626;
cutValue[2] = -0.0985904;