{\rtf1\ansi\ansicpg1252\cocoartf1671\cocoasubrtf400 {\fonttbl\f0\fnil\fcharset0 Menlo-Regular;\f1\fnil\fcharset0 Menlo-Bold;} {\colortbl;\red255\green255\blue255;\red0\green0\blue0;\red180\green36\blue25;\red46\green174\blue187; } {\*\expandedcolortbl;;\csgray\c0;\cssrgb\c76409\c21698\c12524;\cssrgb\c20196\c73240\c78250; } \paperw11900\paperh16840\margl1440\margr1440\vieww27600\viewh13460\viewkind0 \pard\tx560\tx1120\tx1680\tx2240\tx2800\tx3360\tx3920\tx4480\tx5040\tx5600\tx6160\tx6720\pardirnatural\partightenfactor0 \f0\fs18 \cf2 \CocoaLigature0 Using Theano backend.\ DataSetInfo : [dataset] : Added class "Regression"\ : Add Tree CollectionTree of type Regression with 1000000 events\ : Dataset[dataset] : Class index : 0 name : Regression\ ____________________________________________________________________________________________________\ Layer (type) Output Shape Param # Connected to \ ====================================================================================================\ dense_1 (Dense) (None, 15) 195 dense_input_1[0][0] \ ____________________________________________________________________________________________________\ dense_2 (Dense) (None, 1) 16 dense_1[0][0] \ ====================================================================================================\ Total params: 211\ ____________________________________________________________________________________________________\ Factory : Booking method: \f1\b PyKeras \f0\b0 \ : \ PyKeras : [dataset] : Create Transformation "G" with events from all classes.\ : \ : Transformation, Variable selection : \ : Input : variable 'GoodElectronsAuxDyn.calib_eta' <---> Output : variable 'GoodElectronsAuxDyn.calib_eta'\ : Input : variable 'GoodElectronsAuxDyn.calib_phi' <---> Output : variable 'GoodElectronsAuxDyn.calib_phi'\ : Input : variable 'GoodElectronsAuxDyn.calib_pt' <---> Output : variable 'GoodElectronsAuxDyn.calib_pt'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_e' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_e'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_eta' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_eta'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_phi' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_phi'\ : Input : variable 'GoodElectronsAuxDyn.f1' <---> Output : variable 'GoodElectronsAuxDyn.f1'\ : Input : variable 'GoodElectronsAuxDyn.f3' <---> Output : variable 'GoodElectronsAuxDyn.f3'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_energy' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_energy'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_x' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_x'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_y' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_y'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_z' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_z'\ PyKeras : [dataset] : Create Transformation "D" with events from all classes.\ : \ : Transformation, Variable selection : \ : Input : variable 'GoodElectronsAuxDyn.calib_eta' <---> Output : variable 'GoodElectronsAuxDyn.calib_eta'\ : Input : variable 'GoodElectronsAuxDyn.calib_phi' <---> Output : variable 'GoodElectronsAuxDyn.calib_phi'\ : Input : variable 'GoodElectronsAuxDyn.calib_pt' <---> Output : variable 'GoodElectronsAuxDyn.calib_pt'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_e' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_e'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_eta' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_eta'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_phi' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_phi'\ : Input : variable 'GoodElectronsAuxDyn.f1' <---> Output : variable 'GoodElectronsAuxDyn.f1'\ : Input : variable 'GoodElectronsAuxDyn.f3' <---> Output : variable 'GoodElectronsAuxDyn.f3'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_energy' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_energy'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_x' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_x'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_y' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_y'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_z' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_z'\ : Load model from file: timingRegressionKeras.h5\ Factory : \f1\b Train all methods \f0\b0 \ DataSetFactory : [dataset] : Number of events in input trees\ : \ : Number of training and testing events\ : ---------------------------------------------------------------------------\ : Regression -- training events : 990000\ : Regression -- testing events : 10000\ : Regression -- training and testing events: 1000000\ : \ DataSetInfo : Correlation matrix (Regression):\ : ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\ : GoodElectronsAuxDyn.calib_eta GoodElectronsAuxDyn.calib_phi GoodElectronsAuxDyn.calib_pt GoodElectronsAuxDyn.caloCluster_e GoodElectronsAuxDyn.caloCluster_eta GoodElectronsAuxDyn.caloCluster_phi GoodElectronsAuxDyn.f1 GoodElectronsAuxDyn.f3 GoodElectronsAuxDyn.maxEcell_energy GoodElectronsAuxDyn.maxEcell_x GoodElectronsAuxDyn.maxEcell_y GoodElectronsAuxDyn.maxEcell_z\ : GoodElectronsAuxDyn.calib_eta: +1.000 +0.001 -0.002 +0.007 +1.000 +0.001 +0.007 -0.002 +0.005 +0.004 +0.002 +0.994\ : GoodElectronsAuxDyn.calib_phi: +0.001 +1.000 -0.007 +0.001 +0.001 +0.991 -0.017 +0.016 +0.009 -0.011 +0.777 +0.001\ : GoodElectronsAuxDyn.calib_pt: -0.002 -0.007 +1.000 +0.360 -0.002 -0.007 -0.109 +0.008 +0.398 +0.002 -0.009 -0.002\ : GoodElectronsAuxDyn.caloCluster_e: +0.007 +0.001 +0.360 +1.000 +0.007 +0.001 +0.058 +0.359 +0.758 -0.001 -0.000 +0.007\ : GoodElectronsAuxDyn.caloCluster_eta: +1.000 +0.001 -0.002 +0.007 +1.000 +0.001 +0.007 -0.002 +0.005 +0.004 +0.002 +0.994\ : GoodElectronsAuxDyn.caloCluster_phi: +0.001 +0.991 -0.007 +0.001 +0.001 +1.000 -0.017 +0.016 +0.009 -0.012 +0.777 +0.001\ : GoodElectronsAuxDyn.f1: +0.007 -0.017 -0.109 +0.058 +0.007 -0.017 +1.000 -0.274 -0.278 +0.001 -0.019 +0.006\ : GoodElectronsAuxDyn.f3: -0.002 +0.016 +0.008 +0.359 -0.002 +0.016 -0.274 +1.000 +0.417 -0.002 +0.018 -0.001\ : GoodElectronsAuxDyn.maxEcell_energy: +0.005 +0.009 +0.398 +0.758 +0.005 +0.009 -0.278 +0.417 +1.000 +0.001 +0.010 +0.004\ : GoodElectronsAuxDyn.maxEcell_x: +0.004 -0.011 +0.002 -0.001 +0.004 -0.012 +0.001 -0.002 +0.001 +1.000 -0.001 +0.004\ : GoodElectronsAuxDyn.maxEcell_y: +0.002 +0.777 -0.009 -0.000 +0.002 +0.777 -0.019 +0.018 +0.010 -0.001 +1.000 +0.001\ : GoodElectronsAuxDyn.maxEcell_z: +0.994 +0.001 -0.002 +0.007 +0.994 +0.001 +0.006 -0.001 +0.004 +0.004 +0.001 +1.000\ : ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\ DataSetFactory : [dataset] : \ : \ Factory : [dataset] : Create Transformation "G" with events from all classes.\ : \ : Transformation, Variable selection : \ : Input : variable 'GoodElectronsAuxDyn.calib_eta' <---> Output : variable 'GoodElectronsAuxDyn.calib_eta'\ : Input : variable 'GoodElectronsAuxDyn.calib_phi' <---> Output : variable 'GoodElectronsAuxDyn.calib_phi'\ : Input : variable 'GoodElectronsAuxDyn.calib_pt' <---> Output : variable 'GoodElectronsAuxDyn.calib_pt'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_e' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_e'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_eta' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_eta'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_phi' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_phi'\ : Input : variable 'GoodElectronsAuxDyn.f1' <---> Output : variable 'GoodElectronsAuxDyn.f1'\ : Input : variable 'GoodElectronsAuxDyn.f3' <---> Output : variable 'GoodElectronsAuxDyn.f3'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_energy' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_energy'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_x' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_x'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_y' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_y'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_z' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_z'\ Factory : [dataset] : Create Transformation "D" with events from all classes.\ : \ : Transformation, Variable selection : \ : Input : variable 'GoodElectronsAuxDyn.calib_eta' <---> Output : variable 'GoodElectronsAuxDyn.calib_eta'\ : Input : variable 'GoodElectronsAuxDyn.calib_phi' <---> Output : variable 'GoodElectronsAuxDyn.calib_phi'\ : Input : variable 'GoodElectronsAuxDyn.calib_pt' <---> Output : variable 'GoodElectronsAuxDyn.calib_pt'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_e' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_e'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_eta' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_eta'\ : Input : variable 'GoodElectronsAuxDyn.caloCluster_phi' <---> Output : variable 'GoodElectronsAuxDyn.caloCluster_phi'\ : Input : variable 'GoodElectronsAuxDyn.f1' <---> Output : variable 'GoodElectronsAuxDyn.f1'\ : Input : variable 'GoodElectronsAuxDyn.f3' <---> Output : variable 'GoodElectronsAuxDyn.f3'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_energy' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_energy'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_x' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_x'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_y' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_y'\ : Input : variable 'GoodElectronsAuxDyn.maxEcell_z' <---> Output : variable 'GoodElectronsAuxDyn.maxEcell_z'\ : Preparing the Gaussian transformation...\ : Preparing the Decorrelation transformation...\ TFHandler_Factory : Variable Mean RMS [ Min Max ]\ : --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\ : GoodElectronsAuxDyn.calib_eta: 0.0028070 1.0000 [ -69.487 83.402 ]\ : GoodElectronsAuxDyn.calib_phi: 0.0028402 1.0000 [ -22.454 24.279 ]\ : GoodElectronsAuxDyn.calib_pt: 0.0020508 1.0000 [ -4.0482 5.7365 ]\ : GoodElectronsAuxDyn.caloCluster_e: 0.0020392 1.0000 [ -3.8524 5.9887 ]\ : GoodElectronsAuxDyn.caloCluster_eta: 0.0024803 1.0000 [ -71.129 81.739 ]\ : GoodElectronsAuxDyn.caloCluster_phi: 0.0025995 1.0000 [ -22.107 24.523 ]\ : GoodElectronsAuxDyn.f1: 0.00081388 1.0000 [ -4.6821 6.1259 ]\ : GoodElectronsAuxDyn.f3: -0.0052362 1.0000 [ -4.0189 5.2945 ]\ : GoodElectronsAuxDyn.maxEcell_energy: 0.0051939 1.0000 [ -5.4252 6.4975 ]\ : GoodElectronsAuxDyn.maxEcell_x: 0.0035654 1.0000 [ -3.3921 5.6874 ]\ : GoodElectronsAuxDyn.maxEcell_y: 0.0015032 1.0000 [ -3.6443 6.6325 ]\ : GoodElectronsAuxDyn.maxEcell_z: 0.00072501 1.0000 [ -8.1522 12.892 ]\ : GoodElectronsAuxDyn.maxEcell_time: -0.00094564 0.34404 [ -11.115 8.2419 ]\ : --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\ Factory : Train method: PyKeras for Regression\ : \ : \ : \f1\b ================================================================ \f0\b0 \ : \f1\b H e l p f o r M V A m e t h o d [ PyKeras ] : \f0\b0 \ : \ : Keras is a high-level API for the Theano and Tensorflow packages.\ : This method wraps the training and predictions steps of the Keras\ : Python package for TMVA, so that dataloading, preprocessing and\ : evaluation can be done within the TMVA system. To use this Keras\ : interface, you have to generate a model with Keras first. Then,\ : this model can be loaded and trained in TMVA.\ : \ : \ : \ : \f1\b ================================================================ \f0\b0 \ : \ : Preparing the Gaussian transformation...\ : Preparing the Decorrelation transformation...\ TFHandler_PyKeras : Variable Mean RMS [ Min Max ]\ : --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\ : GoodElectronsAuxDyn.calib_eta: 0.0028070 1.0000 [ -69.487 83.402 ]\ : GoodElectronsAuxDyn.calib_phi: 0.0028402 1.0000 [ -22.454 24.279 ]\ : GoodElectronsAuxDyn.calib_pt: 0.0020508 1.0000 [ -4.0482 5.7365 ]\ : GoodElectronsAuxDyn.caloCluster_e: 0.0020392 1.0000 [ -3.8524 5.9887 ]\ : GoodElectronsAuxDyn.caloCluster_eta: 0.0024803 1.0000 [ -71.129 81.739 ]\ : GoodElectronsAuxDyn.caloCluster_phi: 0.0025995 1.0000 [ -22.107 24.523 ]\ : GoodElectronsAuxDyn.f1: 0.00081388 1.0000 [ -4.6821 6.1259 ]\ : GoodElectronsAuxDyn.f3: -0.0052362 1.0000 [ -4.0189 5.2945 ]\ : GoodElectronsAuxDyn.maxEcell_energy: 0.0051939 1.0000 [ -5.4252 6.4975 ]\ : GoodElectronsAuxDyn.maxEcell_x: 0.0035654 1.0000 [ -3.3921 5.6874 ]\ : GoodElectronsAuxDyn.maxEcell_y: 0.0015032 1.0000 [ -3.6443 6.6325 ]\ : GoodElectronsAuxDyn.maxEcell_z: 0.00072501 1.0000 [ -8.1522 12.892 ]\ : GoodElectronsAuxDyn.maxEcell_time: -0.00094564 0.34404 [ -11.115 8.2419 ]\ : --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\ : Option SaveBestOnly: Only model weights with smallest validation loss will be stored\ : Option TriesEarlyStopping: Training will stop after 3 number of epochs with no improvement of validation loss\ Train on 990000 samples, validate on 10000 samples\ Epoch 1/100\ 989600/990000 [============================>.] - ETA: 0s - loss: 0.0920Epoch 00000: val_loss improved from inf to 0.08277, saving model to dataset/weights/TrainedModel_PyKeras.h5\ 990000/990000 [==============================] - 11s - loss: 0.0920 - val_loss: 0.0828\ Epoch 2/100\ 989376/990000 [============================>.] - ETA: 0s - loss: 0.0870Epoch 00001: val_loss improved from 0.08277 to 0.08254, saving model to dataset/weights/TrainedModel_PyKeras.h5\ 990000/990000 [==============================] - 11s - loss: 0.0870 - val_loss: 0.0825\ Epoch 3/100\ 989824/990000 [============================>.] - ETA: 0s - loss: 0.0865Epoch 00002: val_loss improved from 0.08254 to 0.08218, saving model to dataset/weights/TrainedModel_PyKeras.h5\ 990000/990000 [==============================] - 11s - loss: 0.0865 - val_loss: 0.0822\ Epoch 4/100\ 989984/990000 [============================>.] - ETA: 0s - loss: 0.0863Epoch 00003: val_loss did not improve\ 990000/990000 [==============================] - 11s - loss: 0.0863 - val_loss: 0.0822\ Epoch 5/100\ 989408/990000 [============================>.] - ETA: 0s - loss: 0.0863Epoch 00004: val_loss improved from 0.08218 to 0.08197, saving model to dataset/weights/TrainedModel_PyKeras.h5\ 990000/990000 [==============================] - 11s - loss: 0.0863 - val_loss: 0.0820\ Epoch 6/100\ 989728/990000 [============================>.] - ETA: 0s - loss: 0.0862Epoch 00005: val_loss improved from 0.08197 to 0.08167, saving model to dataset/weights/TrainedModel_PyKeras.h5\ 990000/990000 [==============================] - 11s - loss: 0.0862 - val_loss: 0.0817\ Epoch 7/100\ 989280/990000 [============================>.] - ETA: 0s - loss: 0.0860Epoch 00006: val_loss did not improve\ 990000/990000 [==============================] - 11s - loss: 0.0860 - val_loss: 0.0818\ Epoch 8/100\ 989568/990000 [============================>.] - ETA: 0s - loss: 0.0860Epoch 00007: val_loss did not improve\ 990000/990000 [==============================] - 11s - loss: 0.0860 - val_loss: 0.0819\ Epoch 9/100\ 989856/990000 [============================>.] - ETA: 0s - loss: 0.0860Epoch 00008: val_loss improved from 0.08167 to 0.08163, saving model to dataset/weights/TrainedModel_PyKeras.h5\ 990000/990000 [==============================] - 11s - loss: 0.0860 - val_loss: 0.0816\ Epoch 10/100\ 989216/990000 [============================>.] - ETA: 0s - loss: 0.0860Epoch 00009: val_loss did not improve\ 990000/990000 [==============================] - 11s - loss: 0.0860 - val_loss: 0.0818\ Epoch 11/100\ 989312/990000 [============================>.] - ETA: 0s - loss: 0.0860Epoch 00010: val_loss improved from 0.08163 to 0.08147, saving model to dataset/weights/TrainedModel_PyKeras.h5\ 990000/990000 [==============================] - 11s - loss: 0.0860 - val_loss: 0.0815\ Epoch 12/100\ 989184/990000 [============================>.] - ETA: 0s - loss: 0.0860Epoch 00011: val_loss did not improve\ 990000/990000 [==============================] - 11s - loss: 0.0860 - val_loss: 0.0820\ Epoch 13/100\ 989280/990000 [============================>.] - ETA: 0s - loss: 0.0860Epoch 00012: val_loss did not improve\ 990000/990000 [==============================] - 11s - loss: 0.0860 - val_loss: 0.0821\ Epoch 14/100\ 989344/990000 [============================>.] - ETA: 0s - loss: 0.0860Epoch 00013: val_loss did not improve\ 990000/990000 [==============================] - 11s - loss: 0.0860 - val_loss: 0.0816\ Epoch 15/100\ 989536/990000 [============================>.] - ETA: 0s - loss: 0.0859Epoch 00014: val_loss did not improve\ Epoch 00014: early stopping\ 990000/990000 [==============================] - 11s - loss: 0.0859 - val_loss: 0.0819\ : Elapsed time for training with 990000 events: \f1\b \cf3 184 sec \f0\b0 \cf2 \ : Dataset[dataset] : Create results for training\ : Dataset[dataset] : Evaluation of PyKeras on training sample\ : Dataset[dataset] : Elapsed time for evaluation of 990000 events: \f1\b \cf3 321 sec \f0\b0 \cf2 \ : Create variable histograms\ : Create regression target histograms\ : Create regression average deviation\ : Results created\ : Creating xml weight file: \cf4 dataset/weights/TMVARegression_PyKeras.weights.xml\cf2 \ Factory : Training finished\ : \ Factory : === Destroy and recreate all methods via weight files for testing ===\ : \ Factory : \f1\b Test all methods \f0\b0 \ Factory : Test method: PyKeras for Regression performance\ : \ : Dataset[dataset] : Create results for testing\ : Dataset[dataset] : Evaluation of PyKeras on testing sample\ : Load model from file: dataset/weights/TrainedModel_PyKeras.h5\ : Dataset[dataset] : Elapsed time for evaluation of 10000 events: \f1\b \cf3 4.11 sec \f0\b0 \cf2 \ : Create variable histograms\ : Create regression target histograms\ : Create regression average deviation\ : Results created\ Factory : \f1\b Evaluate all methods \f0\b0 \ : Evaluate regression method: PyKeras\ TFHandler_PyKeras : Variable Mean RMS [ Min Max ]\ : --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\ : GoodElectronsAuxDyn.calib_eta: 0.0030347 0.97499 [ -4.9037 13.255 ]\ : GoodElectronsAuxDyn.calib_phi: -0.0095399 1.0472 [ -20.986 24.256 ]\ : GoodElectronsAuxDyn.calib_pt: -0.0063974 0.99113 [ -3.6957 3.1964 ]\ : GoodElectronsAuxDyn.caloCluster_e: -0.012583 1.0137 [ -3.5030 3.4343 ]\ : GoodElectronsAuxDyn.caloCluster_eta: -0.0099956 0.97716 [ -8.0771 7.6003 ]\ : GoodElectronsAuxDyn.caloCluster_phi: -0.0023092 1.0494 [ -22.095 23.440 ]\ : GoodElectronsAuxDyn.f1: -0.0034749 1.0134 [ -3.3290 3.4718 ]\ : GoodElectronsAuxDyn.f3: 0.0021997 1.0004 [ -3.3688 3.3529 ]\ : GoodElectronsAuxDyn.maxEcell_energy: 0.0052004 0.99295 [ -4.8474 3.1362 ]\ : GoodElectronsAuxDyn.maxEcell_x: -0.010282 0.99532 [ -3.3516 3.6006 ]\ : GoodElectronsAuxDyn.maxEcell_y: -0.0026627 1.0032 [ -3.3126 4.0998 ]\ : GoodElectronsAuxDyn.maxEcell_z: 0.018410 0.99444 [ -8.1277 5.1151 ]\ : GoodElectronsAuxDyn.maxEcell_time: -0.0045448 0.33694 [ -1.7844 1.7162 ]\ : --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\ TFHandler_PyKeras : Variable Mean RMS [ Min Max ]\ : --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\ : GoodElectronsAuxDyn.calib_eta: 0.0030347 0.97499 [ -4.9037 13.255 ]\ : GoodElectronsAuxDyn.calib_phi: -0.0095399 1.0472 [ -20.986 24.256 ]\ : GoodElectronsAuxDyn.calib_pt: -0.0063974 0.99113 [ -3.6957 3.1964 ]\ : GoodElectronsAuxDyn.caloCluster_e: -0.012583 1.0137 [ -3.5030 3.4343 ]\ : GoodElectronsAuxDyn.caloCluster_eta: -0.0099956 0.97716 [ -8.0771 7.6003 ]\ : GoodElectronsAuxDyn.caloCluster_phi: -0.0023092 1.0494 [ -22.095 23.440 ]\ : GoodElectronsAuxDyn.f1: -0.0034749 1.0134 [ -3.3290 3.4718 ]\ : GoodElectronsAuxDyn.f3: 0.0021997 1.0004 [ -3.3688 3.3529 ]\ : GoodElectronsAuxDyn.maxEcell_energy: 0.0052004 0.99295 [ -4.8474 3.1362 ]\ : GoodElectronsAuxDyn.maxEcell_x: -0.010282 0.99532 [ -3.3516 3.6006 ]\ : GoodElectronsAuxDyn.maxEcell_y: -0.0026627 1.0032 [ -3.3126 4.0998 ]\ : GoodElectronsAuxDyn.maxEcell_z: 0.018410 0.99444 [ -8.1277 5.1151 ]\ : GoodElectronsAuxDyn.maxEcell_time: -0.0045448 0.33694 [ -1.7844 1.7162 ]\ : --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\ : \ : Evaluation results ranked by smallest RMS on test sample:\ : ("Bias" quotes the mean deviation of the regression from true target.\ : "MutInf" is the "Mutual Information" between regression and target.\ : Indicated by "_T" are the corresponding "truncated" quantities ob-\ : tained when removing events deviating more than 2sigma from average.)\ : --------------------------------------------------------------------------------------------------\ : --------------------------------------------------------------------------------------------------\ : dataset PyKeras : -0.00267 -0.00376 0.285 0.243 | 0.210 0.247\ : --------------------------------------------------------------------------------------------------\ : \ : Evaluation results ranked by smallest RMS on training sample:\ : (overtraining check)\ : --------------------------------------------------------------------------------------------------\ : DataSet Name: MVA Method: RMS RMS_T | MutInf MutInf_T\ : --------------------------------------------------------------------------------------------------\ : dataset PyKeras : -0.00737 -0.00861 0.293 0.249 | 0.126 0.148\ : --------------------------------------------------------------------------------------------------\ : \ Dataset:dataset : Created tree 'TestTree' with 10000 events\ : \ Dataset:dataset : Created tree 'TrainTree' with 990000 events\ : \ Factory : \f1\b Thank you for using TMVA! \f0\b0 \ : \f1\b For citation information, please visit: http://tmva.sf.net/citeTMVA.html \f0\b0 \ }