results1.txt

glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_1


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.034264075410084235
RMSE: 0.1851055790895678
LogLoss: 0.12085960655990972
Null degrees of freedom: 5830
Residual degrees of freedom: 5530
Null deviance: 2686.033133694482
Residual deviance: 1409.4647317016672
AIC: 2011.4647317016672
AUC: 0.942604407491897
pr_auc: 0.6293341059778711
Gini: 0.885208814983794

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.3153655624027554: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.12 %, avg score:  6.12 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.051913320602323294
RMSE: 0.2278449485995318
LogLoss: 0.19882087614699245
Null degrees of freedom: 5830
Residual degrees of freedom: 5530
Null deviance: 2688.148162640199
Residual deviance: 2318.6490576262177
AIC: 2920.6490576262177
AUC: 0.8632391575556054
pr_auc: 0.352901159741942
Gini: 0.7264783151112109

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1917528206408285: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.12 %, avg score: 966.07 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_19


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.012437300639165106
RMSE: 0.1115226463063225
LogLoss: 0.05586303959938421
Mean Per-Class Error: 0.020429911094872688
AUC: 0.9982852476028775
pr_auc: 0.9787117715554463
Gini: 0.996570495205755

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.28823036815699593: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.12 %, avg score:  6.16 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.04135704110695338
RMSE: 0.20336430637393912
LogLoss: 0.14938488295194158
Mean Per-Class Error: 0.17461640038644743
AUC: 0.8995117199859933
pr_auc: 0.4384180726926545
Gini: 0.7990234399719867

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1730570983047975: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  4.99 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04563211753485938
RMSE: 0.21361675387211412
LogLoss: 0.1674845391760666
Mean Per-Class Error: 0.19519546949214472
AUC: 0.8767118100437107
pr_auc: 0.4143699806611364
Gini: 0.7534236200874214

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.14976154569692646: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.12 %, avg score:  4.64 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_462


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 60,783 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.0002025634470533075
RMSE: 0.014232478598378692
LogLoss: 0.0014725022017782006
Mean Per-Class Error: 9.134088417983577e-05
AUC: 0.9999994882863631
pr_auc: 0.9159585512373423
Gini: 0.9999989765727262

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.3817001420578375: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.12 %, avg score:  6.18 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.05283576911382561
RMSE: 0.22986032522779049
LogLoss: 0.33316835515994925
Mean Per-Class Error: 0.18103760689530057
AUC: 0.8645187232893587
pr_auc: 0.36215044818841974
Gini: 0.7290374465787175

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1668261091492872: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.52 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.06600013891723396
RMSE: 0.2569049219404602
LogLoss: 0.3501334670817284
Mean Per-Class Error: 0.22408963585434183
AUC: 0.8364317594045292
pr_auc: 0.30618984138647304
Gini: 0.6728635188090584

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.16336558024892478: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.12 %, avg score:  7.37 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_501


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.2861207498667428
RMSE: 0.5349025610957034
LogLoss: 4.741192193919239
Mean Per-Class Error: 0.21970527341371326
AUC: 0.8037294201568096
pr_auc: 0.03258001686143964
Gini: 0.6074588403136192

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999986437458381: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.12 %, avg score: 33.68 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.2796090055615876
RMSE: 0.5287806781280757
LogLoss: 4.567464844338186
Mean Per-Class Error: 0.2387424007830199
AUC: 0.7859086469702276
pr_auc: 0.03234253068083473
Gini: 0.5718172939404551

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999945381831505: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 32.48 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.2833825989676502
RMSE: 0.5323369224162928
LogLoss: 4.782142831626247
Mean Per-Class Error: 0.23075752039946407
AUC: 0.7954442646623867
pr_auc: 0.03494579064413157
Gini: 0.5908885293247734

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999983158041506: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.12 %, avg score: 33.20 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.924928
dtype: float64
GBM acc:predict    0.932063
dtype: float64
DL acc:predict    0.930165
dtype: float64
B acc:predict    0.82716
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_520


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.036475221620232826
RMSE: 0.19098487275235393
LogLoss: 0.12544952889433192
Null degrees of freedom: 11689
Residual degrees of freedom: 11389
Null deviance: 5391.982001605795
Residual deviance: 2933.0099855494796
AIC: 3535.0099855494796
AUC: 0.9455649412574875
pr_auc: 0.5772143817032235
Gini: 0.8911298825149749

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.28955329569327665: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score:  6.13 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04538024404092814
RMSE: 0.2130263928271052
LogLoss: 0.1620893443905556
Null degrees of freedom: 11689
Residual degrees of freedom: 11389
Null deviance: 5393.446028375481
Residual deviance: 3789.648871851189
AIC: 4391.648871851189
AUC: 0.9062731764197172
pr_auc: 0.4258373760254989
Gini: 0.8125463528394343

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.17895500608203735: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score: 597.89 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_538


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.019961407264670796
RMSE: 0.14128484442667869
LogLoss: 0.07730650019788868
Mean Per-Class Error: 0.04883299835363608
AUC: 0.9901488260585353
pr_auc: 0.9016023399484042
Gini: 0.9802976521170705

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.24719113358179037: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score:  6.17 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03951350363529628
RMSE: 0.19878003832200122
LogLoss: 0.14303040245744278
Mean Per-Class Error: 0.1654334243200256
AUC: 0.9064720964768321
pr_auc: 0.4666585003225704
Gini: 0.8129441929536643

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.18097685296677044: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.54 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04169489484066804
RMSE: 0.20419327814761198
LogLoss: 0.14795680808706776
Mean Per-Class Error: 0.1600097157457998
AUC: 0.9130260264798559
pr_auc: 0.4867419562941152
Gini: 0.8260520529597117

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.16396773163522013: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score:  5.44 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_967


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 121,842 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.0008267592451034052
RMSE: 0.028753421450384044
LogLoss: 0.0034874703973230043
Mean Per-Class Error: 0.001414490890858544
AUC: 0.9999816586182095
pr_auc: 0.9017162450363572
Gini: 0.9999633172364191

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.3003916100298748: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.01 %, avg score:  5.95 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04908488518747141
RMSE: 0.22155108933939235
LogLoss: 0.3290042203203713
Mean Per-Class Error: 0.15289730035311844
AUC: 0.8917533993443896
pr_auc: 0.41918212180418674
Gini: 0.7835067986887791

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.02115714734086097: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  4.65 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.04934972335118933
RMSE: 0.22214797624824165
LogLoss: 0.2517597022206859
Mean Per-Class Error: 0.15671539919017663
AUC: 0.9056982264442416
pr_auc: 0.4312641605563829
Gini: 0.8113964528884832

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19532600985052362: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score:  5.38 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_1020


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.1542671107507001
RMSE: 0.39276852056994094
LogLoss: 1.8513166665925391
Mean Per-Class Error: 0.16853495221757075
AUC: 0.8809800167547045
pr_auc: 0.0652931645181444
Gini: 0.761960033509409

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999927983757287: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score: 20.50 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.1550359843651464
RMSE: 0.3937460912379276
LogLoss: 1.9246534620593279
Mean Per-Class Error: 0.18058144293362477
AUC: 0.8661501027291789
pr_auc: 0.06364639365044591
Gini: 0.7323002054583578

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999947086303693: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 20.10 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.155106569663931
RMSE: 0.39383571405337403
LogLoss: 1.876245851271284
Mean Per-Class Error: 0.17838434671841275
AUC: 0.8717750975165236
pr_auc: 0.0653529692844937
Gini: 0.7435501950330472

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999933046087028: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score: 20.30 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.936466
dtype: float64
GBM acc:predict    0.934264
dtype: float64
DL acc:predict    0.927964
dtype: float64
B acc:predict    0.921057
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_1039


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.038841934233576986
RMSE: 0.19708357169885313
LogLoss: 0.1350181354250104
Null degrees of freedom: 17592
Residual degrees of freedom: 17292
Null deviance: 8076.169251610593
Residual deviance: 4750.748113064416
AIC: 5352.748113064416
AUC: 0.9324544089973231
pr_auc: 0.5208686013777714
Gini: 0.8649088179946463

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19859171807063816: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score:  6.09 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04410895764981131
RMSE: 0.21002132665472645
LogLoss: 0.15591904903508025
Null degrees of freedom: 17592
Residual degrees of freedom: 17292
Null deviance: 8077.663876724361
Residual deviance: 5486.167659348339
AIC: 6088.167659348339
AUC: 0.9076170129920581
pr_auc: 0.42270525644706386
Gini: 0.8152340259841162

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19800148401607492: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score: 439.65 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_1057


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.02576340846452519
RMSE: 0.16050983915176412
LogLoss: 0.09537895358893542
Mean Per-Class Error: 0.07101948831292959
AUC: 0.980314403107036
pr_auc: 0.816487554945469
Gini: 0.960628806214072

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.29751196843290245: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score:  6.12 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03967489887783791
RMSE: 0.19918558903153086
LogLoss: 0.14295892832937174
Mean Per-Class Error: 0.16723456003645087
AUC: 0.9109504136589728
pr_auc: 0.4645275901822439
Gini: 0.8219008273179456

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2090360066062496: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.70 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04225952240466633
RMSE: 0.20557121005789292
LogLoss: 0.15045577779637592
Mean Per-Class Error: 0.160346493653035
AUC: 0.9082515231631926
pr_auc: 0.46067043595346047
Gini: 0.8165030463263852

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.13391111752940213: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score:  5.60 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_1426


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 183,148 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.0008131592451520644
RMSE: 0.02851594720769528
LogLoss: 0.0038237907159489925
Mean Per-Class Error: 0.0012353624808756347
AUC: 0.999958257326115
pr_auc: 0.8873601374070946
Gini: 0.99991651465223

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.4113586824024038: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.83 %, avg score:  5.93 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04948752248594388
RMSE: 0.22245791171802337
LogLoss: 0.29208128200116873
Mean Per-Class Error: 0.13860201830140362
AUC: 0.9019552294847509
pr_auc: 0.427408720168853
Gini: 0.8039104589695019

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.17922675508626557: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.01 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.05142110261149639
RMSE: 0.22676221601381566
LogLoss: 0.2417046412591304
Mean Per-Class Error: 0.16071350732265677
AUC: 0.9034137748248047
pr_auc: 0.4224292157349248
Gini: 0.8068275496496093

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.18448950988314664: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score:  5.78 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_1481


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.14864622303128738
RMSE: 0.38554665480495015
LogLoss: 1.719548298148039
Mean Per-Class Error: 0.17622768331034133
AUC: 0.872089327513513
pr_auc: 0.07352224353261681
Gini: 0.744178655027026

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999932957355422: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score: 19.73 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.148680150958049
RMSE: 0.38559065206258436
LogLoss: 1.7290081428950734
Mean Per-Class Error: 0.16657451978855087
AUC: 0.8757910357716923
pr_auc: 0.06604112178030136
Gini: 0.7515820715433845

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999913958330208: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 19.83 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.14896893707237843
RMSE: 0.38596494280229443
LogLoss: 1.7342898976528052
Mean Per-Class Error: 0.1802904117057711
AUC: 0.8661997744616304
pr_auc: 0.07456294922981647
Gini: 0.7323995489232609

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999907703805865: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score: 19.64 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.927661
dtype: float64
GBM acc:predict    0.937301
dtype: float64
DL acc:predict    0.937149
dtype: float64
B acc:predict    0.925232
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_1500


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.03991758528674311
RMSE: 0.1997938569794955
LogLoss: 0.1384481775770503
Null degrees of freedom: 23199
Residual degrees of freedom: 22899
Null deviance: 10874.995094881593
Residual deviance: 6423.995439575134
AIC: 7025.995439575134
AUC: 0.9315255788153693
pr_auc: 0.5244277137644696
Gini: 0.8630511576307387

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20940518117286813: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.27 %, avg score:  6.27 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04381384589484784
RMSE: 0.2093175718730939
LogLoss: 0.1534008852158109
Null degrees of freedom: 23199
Residual degrees of freedom: 22899
Null deviance: 10878.446031226727
Residual deviance: 7117.801074013629
AIC: 7719.801074013629
AUC: 0.9137769317748126
pr_auc: 0.4559362765281299
Gini: 0.8275538635496251

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.24807332194156417: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.27 %, avg score: 376.92 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_1518


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.028803156204226382
RMSE: 0.16971492628589385
LogLoss: 0.10508324057553159
Mean Per-Class Error: 0.08492911037731155
AUC: 0.9729166320969628
pr_auc: 0.7687509539777678
Gini: 0.9458332641939255

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.26365472678489854: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.27 %, avg score:  6.30 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.0386404431696093
RMSE: 0.1965717252547001
LogLoss: 0.13883804879511494
Mean Per-Class Error: 0.1543634587880911
AUC: 0.9175915808480748
pr_auc: 0.4876443056852089
Gini: 0.8351831616961496

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.18437598892677529: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.07 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04155604738602806
RMSE: 0.20385300435860165
LogLoss: 0.14752613069669374
Mean Per-Class Error: 0.15425555031412996
AUC: 0.9173365761691079
pr_auc: 0.499469458078755
Gini: 0.8346731523382158

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.18297273434066919: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.27 %, avg score:  5.88 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_1857


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 265,204 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.0010079817881440952
RMSE: 0.03174872892170796
LogLoss: 0.005185916646167417
Mean Per-Class Error: 0.002947305132059208
AUC: 0.9995644406270131
pr_auc: 0.9536283199822119
Gini: 0.9991288812540262

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21717298166950347: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.29 %, avg score:  6.14 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04552927205100707
RMSE: 0.21337589379076322
LogLoss: 0.3043091152143173
Mean Per-Class Error: 0.1599059827617485
AUC: 0.8819545228261283
pr_auc: 0.43373857581309433
Gini: 0.7639090456522566

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.06298196228132207: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  4.64 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.04422834330090539
RMSE: 0.21030535728056332
LogLoss: 0.20480405785087238
Mean Per-Class Error: 0.15929522053100642
AUC: 0.9122599736559102
pr_auc: 0.5053651635299075
Gini: 0.8245199473118203

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.11646114665410885: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.27 %, avg score:  4.78 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_1922


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.1744754936919775
RMSE: 0.41770263788008033
LogLoss: 1.9778057286649964
Mean Per-Class Error: 0.16891745070755293
AUC: 0.8769929651640105
pr_auc: 0.057227008515648665
Gini: 0.753985930328021

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999875844365684: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.27 %, avg score: 23.14 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.17964566170362833
RMSE: 0.423846271310281
LogLoss: 2.0573530685342116
Mean Per-Class Error: 0.16995572309107243
AUC: 0.8734550016664487
pr_auc: 0.05511804251335961
Gini: 0.7469100033328975

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999910682583553: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 23.40 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.17513256221705759
RMSE: 0.41848842542782183
LogLoss: 1.9901579084239953
Mean Per-Class Error: 0.16982203121308448
AUC: 0.8736603034706402
pr_auc: 0.06039057667947188
Gini: 0.7473206069412803

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999894357658929: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.27 %, avg score: 23.14 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.931456
dtype: float64
GBM acc:predict    0.932291
dtype: float64
DL acc:predict    0.937301
dtype: float64
B acc:predict    0.915136
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_1941


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.03881187353747626
RMSE: 0.19700729310732704
LogLoss: 0.1339720378757824
Null degrees of freedom: 29453
Residual degrees of freedom: 29153
Null deviance: 13751.202935745077
Residual deviance: 7892.02480718659
AIC: 8494.02480718659
AUC: 0.9362250129990819
pr_auc: 0.5465981891554834
Gini: 0.8724500259981638

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20121101893026222: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score:  6.24 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04190258173243758
RMSE: 0.20470120110159973
LogLoss: 0.14650326134308198
Null degrees of freedom: 29453
Residual degrees of freedom: 29153
Null deviance: 13752.736262004022
Residual deviance: 8630.214119198181
AIC: 9232.214119198181
AUC: 0.921592557691255
pr_auc: 0.48742718879366265
Gini: 0.8431851153825101

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.18565156809517233: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score: 334.53 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_1959


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.029744593414002974
RMSE: 0.1724662094846494
LogLoss: 0.10800980380900395
Mean Per-Class Error: 0.09174077787602086
AUC: 0.9676545646966755
pr_auc: 0.7452777440378126
Gini: 0.9353091293933511

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2568787368316338: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score:  6.26 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.039196081478799605
RMSE: 0.1979800027245166
LogLoss: 0.1403660860287305
Mean Per-Class Error: 0.1510614645341748
AUC: 0.9160154684408368
pr_auc: 0.4790010868483477
Gini: 0.8320309368816736

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.228810924320401: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.01 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04090498875048355
RMSE: 0.20224981767725664
LogLoss: 0.14611458973378058
Mean Per-Class Error: 0.15665224702724168
AUC: 0.9161685910997875
pr_auc: 0.5017549788994642
Gini: 0.8323371821995751

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22208736197938309: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score:  5.97 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_2262


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 306,304 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.0021624681361534655
RMSE: 0.0465023454908832
LogLoss: 0.013813961296982312
Mean Per-Class Error: 0.00561207737719327
AUC: 0.9978772872967895
pr_auc: 0.863251316998858
Gini: 0.995754574593579

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.6047686425883746: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.39 %, avg score:  6.59 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.047439560032100904
RMSE: 0.21780624424497316
LogLoss: 0.2622322554545281
Mean Per-Class Error: 0.14557990305065582
AUC: 0.9087358299125431
pr_auc: 0.4235897893899065
Gini: 0.8174716598250862

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2691718385547542: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.94 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.04616786495059195
RMSE: 0.21486708670848578
LogLoss: 0.19415218993997096
Mean Per-Class Error: 0.14746555896229607
AUC: 0.9186603799317394
pr_auc: 0.5031396949496503
Gini: 0.8373207598634789

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.33504671840389477: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score:  6.68 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_2327


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.13166215153949717
RMSE: 0.3628527959648336
LogLoss: 1.4836588711979497
Mean Per-Class Error: 0.16107997706949928
AUC: 0.8889458555197505
pr_auc: 0.09398174917570676
Gini: 0.7778917110395009

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999936947334661: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score: 18.26 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.13562641996044364
RMSE: 0.3682749244252772
LogLoss: 1.5413917333798417
Mean Per-Class Error: 0.15985029384342142
AUC: 0.8876988136572853
pr_auc: 0.06973360551773108
Gini: 0.7753976273145706

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.99998283308723: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 18.58 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.1321152041097196
RMSE: 0.3634765523520322
LogLoss: 1.513693602430757
Mean Per-Class Error: 0.16612044142416282
AUC: 0.880504992181628
pr_auc: 0.08414830283496666
Gini: 0.7610099843632561

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999823052190037: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score: 18.16 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.927053
dtype: float64
GBM acc:predict    0.937984
dtype: float64
DL acc:predict    0.939198
dtype: float64
B acc:predict    0.930014
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_2346


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.039402079054335654
RMSE: 0.19849956940592
LogLoss: 0.1365288817545869
Null degrees of freedom: 34901
Residual degrees of freedom: 34601
Null deviance: 15914.03637586223
Residual deviance: 9530.262061997186
AIC: 10132.262061997186
AUC: 0.9316999007150096
pr_auc: 0.502282713612446
Gini: 0.8633998014300193

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21383977207444305: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.04 %, avg score:  6.04 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.042015523487942685
RMSE: 0.2049768852528077
LogLoss: 0.1468319604412911
Null degrees of freedom: 34901
Residual degrees of freedom: 34601
Null deviance: 15915.519212971467
Residual deviance: 10249.458166643884
AIC: 10851.458166643884
AUC: 0.9184810243090844
pr_auc: 0.45130072753311523
Gini: 0.8369620486181688

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2103395651249438: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.04 %, avg score: 305.38 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_2364


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.030794805108963198
RMSE: 0.17548448680428477
LogLoss: 0.11105190281578103
Mean Per-Class Error: 0.0992649900544964
AUC: 0.9637690104779276
pr_auc: 0.7091733009125055
Gini: 0.9275380209558552

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2719425567278518: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.04 %, avg score:  6.06 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.0390732324461714
RMSE: 0.1976695030756424
LogLoss: 0.14014226426886817
Mean Per-Class Error: 0.1468866941454915
AUC: 0.9153062257679375
pr_auc: 0.47943354808117317
Gini: 0.830612451535875

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22325071718528802: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.93 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.040267176970242985
RMSE: 0.2006668307674265
LogLoss: 0.14321009805782392
Mean Per-Class Error: 0.1562739235328292
AUC: 0.9185310611655888
pr_auc: 0.4889448770469972
Gini: 0.8370621223311776

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1797569036204227: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.04 %, avg score:  5.82 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_2631


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 363,641 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.0024407889922290816
RMSE: 0.049404341835805096
LogLoss: 0.011006152995036647
Mean Per-Class Error: 0.007921318447634218
AUC: 0.9985494159253917
pr_auc: 0.8241307292565068
Gini: 0.9970988318507834

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.6425598384226923: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score:  6.45 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.05055390145750348
RMSE: 0.22484194772662747
LogLoss: 0.2824748682716739
Mean Per-Class Error: 0.14664495361366225
AUC: 0.9120426231168074
pr_auc: 0.4153482063858558
Gini: 0.8240852462336148

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.16464301205318324: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.41 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.043093705485971925
RMSE: 0.20759023456312178
LogLoss: 0.18041948461836224
Mean Per-Class Error: 0.14629899116579947
AUC: 0.92013716828151
pr_auc: 0.5060611915810653
Gini: 0.8402743365630201

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1917320870245977: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.04 %, avg score:  5.95 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_2698


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.1590856836610372
RMSE: 0.3988554671319389
LogLoss: 1.8268435919218513
Mean Per-Class Error: 0.17231579038008693
AUC: 0.8753538126745999
pr_auc: 0.07083973802477385
Gini: 0.7507076253491998

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999932719645568: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.04 %, avg score: 21.04 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.16146566908033347
RMSE: 0.4018279097826997
LogLoss: 1.8652177280095872
Mean Per-Class Error: 0.16808592546966428
AUC: 0.875398260143188
pr_auc: 0.06375811765382408
Gini: 0.750796520286376

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999931540869496: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 21.30 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.15950132773614503
RMSE: 0.39937617322036756
LogLoss: 1.8308390943460593
Mean Per-Class Error: 0.17577795010685016
AUC: 0.870785465765709
pr_auc: 0.0676022727463182
Gini: 0.7415709315314181

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999895542911883: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.04 %, avg score: 21.01 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.932443
dtype: float64
GBM acc:predict    0.937453
dtype: float64
DL acc:predict    0.930165
dtype: float64
B acc:predict    0.921892
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_2717


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.03993266867300027
RMSE: 0.19983160078676312
LogLoss: 0.1385755423768015
Null degrees of freedom: 40949
Residual degrees of freedom: 40649
Null deviance: 18786.19204132063
Residual deviance: 11349.336920660044
AIC: 11951.336920660044
AUC: 0.9285255599992327
pr_auc: 0.5007114744204069
Gini: 0.8570511199984654

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19707778788591382: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score:  6.09 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04202961112699646
RMSE: 0.20501124634272253
LogLoss: 0.14698963855410097
Null degrees of freedom: 40949
Residual degrees of freedom: 40649
Null deviance: 18789.736466452337
Residual deviance: 12038.45139758087
AIC: 12640.45139758087
AUC: 0.9172959633464587
pr_auc: 0.46033015857327897
Gini: 0.8345919266929174

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.201733611614425: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score: 277.44 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_2735


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.032630962583081284
RMSE: 0.18064042344691644
LogLoss: 0.1172738883104788
Mean Per-Class Error: 0.10709795524824994
AUC: 0.9570021428593557
pr_auc: 0.6719071834035407
Gini: 0.9140042857187114

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2460122252161818: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score:  6.11 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.039294399676118384
RMSE: 0.19822815056423843
LogLoss: 0.14086172121595664
Mean Per-Class Error: 0.1549958655196999
AUC: 0.9144341351975058
pr_auc: 0.47269911924385516
Gini: 0.8288682703950117

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1680392840647636: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.97 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04060913377899853
RMSE: 0.2015170806135265
LogLoss: 0.14511220867754582
Mean Per-Class Error: 0.1577233947540828
AUC: 0.9153541766544578
pr_auc: 0.49127470835288983
Gini: 0.8307083533089157

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20243840017851536: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score:  5.90 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_2990


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 468,315 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.0015070006084471207
RMSE: 0.03882010572431663
LogLoss: 0.00846131580921504
Mean Per-Class Error: 0.005288230460851917
AUC: 0.9989618547925858
pr_auc: 0.9033358915232674
Gini: 0.9979237095851716

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.32858463818205536: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.55 %, avg score:  6.48 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.047514242631101704
RMSE: 0.21797761956471978
LogLoss: 0.26962049667190197
Mean Per-Class Error: 0.14672869793402543
AUC: 0.9090395876488729
pr_auc: 0.4268908009317607
Gini: 0.8180791752977459

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.0720999426330932: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.59 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.0437828597395489
RMSE: 0.20924354169137194
LogLoss: 0.1738332763380223
Mean Per-Class Error: 0.15338595152783985
AUC: 0.9173264932225501
pr_auc: 0.4952400246909007
Gini: 0.8346529864451002

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2620005455754741: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score:  6.37 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_3069


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.1432379519104443
RMSE: 0.3784679007662926
LogLoss: 1.5916296896243731
Mean Per-Class Error: 0.1647878276351411
AUC: 0.8867228739730157
pr_auc: 0.08121138524474954
Gini: 0.7734457479460315

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999915411649016: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score: 19.67 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.14480334719772298
RMSE: 0.3805303499035563
LogLoss: 1.6120801382364591
Mean Per-Class Error: 0.16058880483987337
AUC: 0.8866112449426446
pr_auc: 0.06704700111393584
Gini: 0.7732224898852893

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.999987936140446: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 19.73 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.14516779717003
RMSE: 0.38100892006622367
LogLoss: 1.690206688875229
Mean Per-Class Error: 0.17444679410798636
AUC: 0.8779247467446646
pr_auc: 0.08406737280050268
Gini: 0.7558494934893292

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999891862280454: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.09 %, avg score: 19.56 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.93138
dtype: float64
GBM acc:predict    0.927661
dtype: float64
DL acc:predict    0.929634
dtype: float64
B acc:predict    0.928571
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_3088


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04087483744466516
RMSE: 0.20217526417607362
LogLoss: 0.14260962149497028
Null degrees of freedom: 46979
Residual degrees of freedom: 46679
Null deviance: 22450.260466163316
Residual deviance: 13399.600035667409
AIC: 14001.600035667409
AUC: 0.9285521747019003
pr_auc: 0.5231175208957075
Gini: 0.8571043494038006

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.26281005257201817: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.44 %, avg score:  6.44 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.042801826056883444
RMSE: 0.20688602189825064
LogLoss: 0.14991828155247114
Null degrees of freedom: 46979
Residual degrees of freedom: 46679
Null deviance: 22453.09952561532
Residual deviance: 14086.321734670193
AIC: 14688.321734670193
AUC: 0.9194190271503276
pr_auc: 0.487326608783815
Gini: 0.8388380543006553

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23111515501481408: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.44 %, avg score: 254.04 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_3106


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.034227105325477035
RMSE: 0.18500568998135444
LogLoss: 0.12281853989577458
Mean Per-Class Error: 0.1097388482298729
AUC: 0.9537367537491352
pr_auc: 0.6769447440141
Gini: 0.9074735074982705

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22826775950375036: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.44 %, avg score:  6.47 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03874840039849935
RMSE: 0.19684613381648963
LogLoss: 0.13970253608285163
Mean Per-Class Error: 0.15122505073176085
AUC: 0.9162442359860103
pr_auc: 0.4846744121728569
Gini: 0.8324884719720207

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22957471793701528: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.18 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.041717980522691045
RMSE: 0.20424979932105453
LogLoss: 0.14840134868010785
Mean Per-Class Error: 0.15291883742785717
AUC: 0.9175196220731998
pr_auc: 0.5128690152174055
Gini: 0.8350392441463996

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1829884235874817: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.44 %, avg score:  6.29 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_3331


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 488,867 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.003401861643314642
RMSE: 0.05832548022360932
LogLoss: 0.018806533738068586
Mean Per-Class Error: 0.012533630224784709
AUC: 0.9973453588521054
pr_auc: 0.9129957143645235
Gini: 0.9946907177042108

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.5076420602853565: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.68 %, avg score:  6.75 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04571179639232159
RMSE: 0.21380317208199132
LogLoss: 0.24544829247120895
Mean Per-Class Error: 0.1412383823212997
AUC: 0.909247630056828
pr_auc: 0.4321142705831309
Gini: 0.8184952601136559

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2821743753871057: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.91 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.04319211306536109
RMSE: 0.20782712302623324
LogLoss: 0.17363235656349296
Mean Per-Class Error: 0.14677528489873193
AUC: 0.9208472312022654
pr_auc: 0.5158258658722653
Gini: 0.8416944624045308

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1686488908618332: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.44 %, avg score:  5.69 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_3410


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.16860561816989705
RMSE: 0.4106161445558334
LogLoss: 1.9913776923785793
Mean Per-Class Error: 0.17489698342023796
AUC: 0.8703173458306062
pr_auc: 0.061175402752630385
Gini: 0.7406346916612123

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999901162554814: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.44 %, avg score: 22.37 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.16982803952750747
RMSE: 0.41210197709730473
LogLoss: 2.020554005510383
Mean Per-Class Error: 0.16855295298927153
AUC: 0.874433987537495
pr_auc: 0.05812502890716324
Gini: 0.74886797507499

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.999993959090103: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 22.23 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.16876337659667912
RMSE: 0.41080819928122064
LogLoss: 2.0031545759182308
Mean Per-Class Error: 0.17526408294244766
AUC: 0.8672302693668117
pr_auc: 0.05993273615044829
Gini: 0.7344605387336234

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999851535134231: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.44 %, avg score: 22.34 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.939502
dtype: float64
GBM acc:predict    0.937528
dtype: float64
DL acc:predict    0.941703
dtype: float64
B acc:predict    0.917489
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_3429


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.03929651625083986
RMSE: 0.19823348922631578
LogLoss: 0.13684840212916186
Null degrees of freedom: 52515
Residual degrees of freedom: 52215
Null deviance: 23910.555528644567
Residual deviance: 14373.461372430127
AIC: 14975.461372430127
AUC: 0.930263247080392
pr_auc: 0.5048824249764345
Gini: 0.860526494160784

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21228999375987556: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.02 %, avg score:  6.02 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04093117139690486
RMSE: 0.20231453580231168
LogLoss: 0.14327288475800942
Null degrees of freedom: 52515
Residual degrees of freedom: 52215
Null deviance: 23911.86167447227
Residual deviance: 15048.237631903248
AIC: 15650.237631903248
AUC: 0.9216448715171633
pr_auc: 0.47330432013063894
Gini: 0.8432897430343267

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2213239379081278: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.02 %, avg score: 246.36 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_3447


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03240608019845979
RMSE: 0.18001688864787044
LogLoss: 0.11659561192647215
Mean Per-Class Error: 0.10689421117659581
AUC: 0.9559465418985552
pr_auc: 0.6665394881212002
Gini: 0.9118930837971104

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.26159729306599194: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.02 %, avg score:  6.05 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03781785814885803
RMSE: 0.19446814173241342
LogLoss: 0.1358882430198127
Mean Per-Class Error: 0.14382886626643776
AUC: 0.9211795081655986
pr_auc: 0.5117458591900641
Gini: 0.8423590163311971

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.17778647458446062: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.84 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.03911968431270022
RMSE: 0.19778696699403686
LogLoss: 0.13939704202198602
Mean Per-Class Error: 0.1476259888201663
AUC: 0.9235075356647436
pr_auc: 0.5122954297331406
Gini: 0.8470150713294873

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.17568505952928284: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.02 %, avg score:  5.90 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_3686


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 546,319 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.002383213479269212
RMSE: 0.04881816751240477
LogLoss: 0.010429710556906238
Mean Per-Class Error: 0.006432214704031769
AUC: 0.9995218577940108
pr_auc: 0.8739905896342239
Gini: 0.9990437155880216

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.5000723764396792: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.04 %, avg score:  6.16 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04536830159144906
RMSE: 0.21299836053699817
LogLoss: 0.2565888550026753
Mean Per-Class Error: 0.1489971775843666
AUC: 0.9023696267576814
pr_auc: 0.45140780576606426
Gini: 0.8047392535153628

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.13170186016670216: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.30 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.04090785349147186
RMSE: 0.20225689973761551
LogLoss: 0.15635435664791203
Mean Per-Class Error: 0.14340287547603037
AUC: 0.9268547044795364
pr_auc: 0.5182786941371454
Gini: 0.8537094089590729

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.32924094702438955: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.02 %, avg score:  6.10 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_3765


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.19969085325734282
RMSE: 0.4468678252653046
LogLoss: 2.5497186501979425
Mean Per-Class Error: 0.18772733308891842
AUC: 0.8534522807494099
pr_auc: 0.04508375608714359
Gini: 0.7069045614988199

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999935434237118: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.02 %, avg score: 25.21 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.20088434132688748
RMSE: 0.44820122860930167
LogLoss: 2.591493546435532
Mean Per-Class Error: 0.1825665208348275
AUC: 0.8483423054368415
pr_auc: 0.046001151939390764
Gini: 0.696684610873683

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999920575189924: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 25.18 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.20044216706206822
RMSE: 0.44770768036975667
LogLoss: 2.5695587547994623
Mean Per-Class Error: 0.18867070969217448
AUC: 0.8502373247809948
pr_auc: 0.04497651169937981
Gini: 0.7004746495619896

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999903869902805: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.02 %, avg score: 25.22 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.935707
dtype: float64
GBM acc:predict    0.93601
dtype: float64
DL acc:predict    0.937604
dtype: float64
B acc:predict    0.896539
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_3784


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04020613080975613
RMSE: 0.2005146648246859
LogLoss: 0.13911970766111398
Null degrees of freedom: 58445
Residual degrees of freedom: 58145
Null deviance: 27095.620505757295
Residual deviance: 16261.980867922937
AIC: 16863.98086792294
AUC: 0.9308624038154364
pr_auc: 0.5045639798424029
Gini: 0.8617248076308728

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21340997458319147: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.18 %, avg score:  6.18 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.041823545856564226
RMSE: 0.20450805817024478
LogLoss: 0.14541433641441218
Null degrees of freedom: 58445
Residual degrees of freedom: 58145
Null deviance: 27097.83561123124
Residual deviance: 16997.77261215347
AIC: 17599.77261215347
AUC: 0.9230202665252764
pr_auc: 0.4737239350175519
Gini: 0.8460405330505527

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20866775962688805: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.18 %, avg score: 229.66 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_3802


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03399996588054943
RMSE: 0.18439079662648414
LogLoss: 0.1211667815845791
Mean Per-Class Error: 0.11136562025593721
AUC: 0.9532705201649886
pr_auc: 0.6481186826281945
Gini: 0.9065410403299772

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.24186876112251696: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.18 %, avg score:  6.20 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.038696208687201916
RMSE: 0.1967135193300194
LogLoss: 0.13907443244366294
Mean Per-Class Error: 0.15485748699537183
AUC: 0.9171715935890242
pr_auc: 0.48642744474207583
Gini: 0.8343431871780485

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20672832611053704: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.01 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04031482028712525
RMSE: 0.20078550816013901
LogLoss: 0.14274834064485675
Mean Per-Class Error: 0.15020146701855275
AUC: 0.9224205204983927
pr_auc: 0.508548436173052
Gini: 0.8448410409967855

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1918696154890809: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.18 %, avg score:  6.03 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_4033


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 607,749 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.0034546179394886176
RMSE: 0.058775997988027545
LogLoss: 0.016316278580382617
Mean Per-Class Error: 0.012179268180704583
AUC: 0.9969853876291319
pr_auc: 0.9174608204356666
Gini: 0.9939707752582638

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.5547466075931267: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.71 %, avg score:  5.92 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.045932738178298065
RMSE: 0.2143192436023841
LogLoss: 0.25881578577629694
Mean Per-Class Error: 0.14382570212335088
AUC: 0.9122387999881871
pr_auc: 0.44132119510171025
Gini: 0.8244775999763743

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1991612743071197: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.79 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.04108541491850167
RMSE: 0.20269537468452917
LogLoss: 0.1586762436971102
Mean Per-Class Error: 0.13619010824318456
AUC: 0.9284121916592796
pr_auc: 0.5215716907223087
Gini: 0.8568243833185591

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.17776605424271624: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.18 %, avg score:  5.76 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_4112


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.2225625801089198
RMSE: 0.4717653867219593
LogLoss: 2.555134531820335
Mean Per-Class Error: 0.17497164044325375
AUC: 0.867751167975261
pr_auc: 0.04491513751199536
Gini: 0.735502335950522

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999926947021173: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.18 %, avg score: 28.38 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.2256682027740158
RMSE: 0.47504547442746553
LogLoss: 2.6202319098155598
Mean Per-Class Error: 0.17379309035527002
AUC: 0.8658574194936527
pr_auc: 0.046589829471064576
Gini: 0.7317148389873054

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999960907855372: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 28.48 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.2246365210472528
RMSE: 0.4739583537055263
LogLoss: 2.8530931990800723
Mean Per-Class Error: 0.19108652665444725
AUC: 0.8526788314044054
pr_auc: 0.04754529110283777
Gini: 0.7053576628088107

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999897864812985: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.18 %, avg score: 28.24 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.932595
dtype: float64
GBM acc:predict    0.93434
dtype: float64
DL acc:predict    0.93935
dtype: float64
B acc:predict    0.903522
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_4131


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04000134277252373
RMSE: 0.20000335690313734
LogLoss: 0.13920197647376448
Null degrees of freedom: 64267
Residual degrees of freedom: 63967
Null deviance: 29715.18298650372
Residual deviance: 17892.46524803179
AIC: 18494.46524803179
AUC: 0.929349975663198
pr_auc: 0.5081371562737149
Gini: 0.858699951326396

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20479849401506664: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score:  6.15 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04144152808423421
RMSE: 0.2035719236148104
LogLoss: 0.14466180787042462
Null degrees of freedom: 64267
Residual degrees of freedom: 63967
Null deviance: 29721.793411095947
Residual deviance: 18594.2501364329
AIC: 19196.2501364329
AUC: 0.922332565034787
pr_auc: 0.4804154270004524
Gini: 0.844665130069574

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20228287514731288: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score: 218.09 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_4149


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03449945359040237
RMSE: 0.18574028531905074
LogLoss: 0.12386184073507658
Mean Per-Class Error: 0.1171776570613523
AUC: 0.949348029222105
pr_auc: 0.6412217887297139
Gini: 0.8986960584442101

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22237315979204272: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score:  6.18 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.038704765484049795
RMSE: 0.19673526751462178
LogLoss: 0.13927527352583063
Mean Per-Class Error: 0.15151235492403947
AUC: 0.9186065852145923
pr_auc: 0.4813873195024291
Gini: 0.8372131704291845

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19408582047118056: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.02 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04034963857319231
RMSE: 0.20087219462432404
LogLoss: 0.14397565252529043
Mean Per-Class Error: 0.1515750148147308
AUC: 0.9191997465057075
pr_auc: 0.5060481274795494
Gini: 0.8383994930114149

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1859582018944653: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score:  6.06 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_4346


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 734,397 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.002770485728618431
RMSE: 0.05263540375658223
LogLoss: 0.010642065080341438
Mean Per-Class Error: 0.005708096966722032
AUC: 0.9997981894284579
pr_auc: 0.8978447344037419
Gini: 0.9995963788569158

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.29652831704196597: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.40 %, avg score:  6.20 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04533410207210263
RMSE: 0.21291806422213835
LogLoss: 0.2655572131457466
Mean Per-Class Error: 0.15486096755276746
AUC: 0.8994512321173179
pr_auc: 0.4439074076467499
Gini: 0.7989024642346358

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.13871757939068347: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.12 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.041232893786578795
RMSE: 0.2030588431627118
LogLoss: 0.1552489311445421
Mean Per-Class Error: 0.1396436222727947
AUC: 0.9277515547161549
pr_auc: 0.537553066910347
Gini: 0.8555031094323098

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2971931175666297: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score:  6.91 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_4435


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.15262378669908488
RMSE: 0.3906709442729071
LogLoss: 1.7832085254942138
Mean Per-Class Error: 0.169108334413001
AUC: 0.8776165290117506
pr_auc: 0.06562394279371161
Gini: 0.7552330580235012

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999854330381135: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score: 20.50 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.15498057315134423
RMSE: 0.39367572080501007
LogLoss: 1.8213977066599418
Mean Per-Class Error: 0.16810448844244008
AUC: 0.8722666549944944
pr_auc: 0.05840026328853786
Gini: 0.7445333099889888

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999846724911372: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 20.56 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.15290519179127968
RMSE: 0.39103093457075705
LogLoss: 1.7903877017030323
Mean Per-Class Error: 0.17005336028027251
AUC: 0.8775161361664455
pr_auc: 0.06957907714256915
Gini: 0.755032272332891

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999900579810348: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score: 20.50 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.933202
dtype: float64
GBM acc:predict    0.934796
dtype: float64
DL acc:predict    0.937301
dtype: float64
B acc:predict    0.921208
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_4454


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04102089437223671
RMSE: 0.20253615571605163
LogLoss: 0.14205034842019013
Null degrees of freedom: 70252
Residual degrees of freedom: 69952
Null deviance: 33066.399102292606
Residual deviance: 19958.92625512723
AIC: 20560.92625512723
AUC: 0.9294689854512187
pr_auc: 0.5060018402924359
Gini: 0.8589379709024374

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19985568102191367: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.31 %, avg score:  6.31 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04231110561784485
RMSE: 0.20569663492105275
LogLoss: 0.14706123848898772
Null degrees of freedom: 70252
Residual degrees of freedom: 69952
Null deviance: 33067.49145514975
Residual deviance: 20662.98637513371
AIC: 21264.98637513371
AUC: 0.9230287930887321
pr_auc: 0.4816712515293586
Gini: 0.8460575861774642

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21372481181839278: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.31 %, avg score: 206.60 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_4472


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.0351820418021479
RMSE: 0.1875687655291997
LogLoss: 0.12581771437311975
Mean Per-Class Error: 0.11725602482985487
AUC: 0.9492834847168329
pr_auc: 0.6400490209469565
Gini: 0.8985669694336658

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2266209124758932: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.31 %, avg score:  6.33 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03845867021372097
RMSE: 0.1961088223760496
LogLoss: 0.13806785526913343
Mean Per-Class Error: 0.14963169374469998
AUC: 0.9194470343541562
pr_auc: 0.4860731493564979
Gini: 0.8388940687083124

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23008866610662848: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.13 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.040821159012762454
RMSE: 0.20204246833961037
LogLoss: 0.1453225590571018
Mean Per-Class Error: 0.1504134585605854
AUC: 0.92067355273509
pr_auc: 0.5147264625390237
Gini: 0.84134710547018

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20648627447997242: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.31 %, avg score:  6.20 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_4687


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 732,076 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.003838560563945917
RMSE: 0.0619561180509715
LogLoss: 0.017218765817387894
Mean Per-Class Error: 0.012371850148510166
AUC: 0.9982681625016838
pr_auc: 0.9349482534121113
Gini: 0.9965363250033676

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2992555687491647: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.86 %, avg score:  5.53 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04390069587973084
RMSE: 0.2095249290173626
LogLoss: 0.24275099784517976
Mean Per-Class Error: 0.14916677565381753
AUC: 0.9070305149959288
pr_auc: 0.4597924938587914
Gini: 0.8140610299918576

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20970634352690545: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  4.68 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.039712721765064615
RMSE: 0.19928051024890672
LogLoss: 0.14888938946881788
Mean Per-Class Error: 0.1345958730425253
AUC: 0.9319602141804229
pr_auc: 0.5507036051329385
Gini: 0.8639204283608457

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19156431216701283: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.31 %, avg score:  5.59 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_4786


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.11858423706285343
RMSE: 0.34436062066219686
LogLoss: 1.332857616070312
Mean Per-Class Error: 0.16098632846753325
AUC: 0.8890175147418401
pr_auc: 0.111854579345688
Gini: 0.7780350294836802

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999881665788549: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.31 %, avg score: 16.55 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.11964963072944997
RMSE: 0.3459040773530286
LogLoss: 1.347216872176963
Mean Per-Class Error: 0.15599541832081032
AUC: 0.8932374351878025
pr_auc: 0.09928068937565733
Gini: 0.786474870375605

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999939310719197: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 16.54 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.12161598154752726
RMSE: 0.3487348298457257
LogLoss: 1.3865274855301424
Mean Per-Class Error: 0.16683126739143095
AUC: 0.8847500300026797
pr_auc: 0.1137957639007962
Gini: 0.7695000600053594

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.999990859790871: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.31 %, avg score: 16.72 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.933354
dtype: float64
GBM acc:predict    0.938971
dtype: float64
DL acc:predict    0.944284
dtype: float64
B acc:predict    0.93434
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_4805


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.039352538229155264
RMSE: 0.19837474191326693
LogLoss: 0.13724160463637006
Null degrees of freedom: 76109
Residual degrees of freedom: 75809
Null deviance: 34853.154090192475
Residual deviance: 20890.91705774825
AIC: 21492.91705774825
AUC: 0.9303505323034671
pr_auc: 0.5095803250774779
Gini: 0.8607010646069342

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2162800737623707: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.07 %, avg score:  6.07 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04049949493268326
RMSE: 0.20124486312123166
LogLoss: 0.14169075302934664
Null degrees of freedom: 76109
Residual degrees of freedom: 75809
Null deviance: 34855.24577723895
Residual deviance: 21568.16642612714
AIC: 22170.16642612714
AUC: 0.9245391140499274
pr_auc: 0.48666991817637806
Gini: 0.8490782280998548

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22579084251665202: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.07 %, avg score: 203.07 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_4823


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.034830248748696876
RMSE: 0.18662863860805737
LogLoss: 0.12481311970024084
Mean Per-Class Error: 0.12119311367299823
AUC: 0.9463295646632992
pr_auc: 0.621485386574526
Gini: 0.8926591293265984

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2413970481823341: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.07 %, avg score:  6.09 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.038183772796868555
RMSE: 0.19540668565038544
LogLoss: 0.13743795226560307
Mean Per-Class Error: 0.1512663955347613
AUC: 0.9199970151583547
pr_auc: 0.49401651177790795
Gini: 0.8399940303167095

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.26401906472874737: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.91 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.03976809181941867
RMSE: 0.19941938676923734
LogLoss: 0.14203046511209366
Mean Per-Class Error: 0.1524359711949428
AUC: 0.9200946980005322
pr_auc: 0.5053311383829858
Gini: 0.8401893960010645

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19616457898544132: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.07 %, avg score:  5.99 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_5032


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 871,116 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.005178443584669376
RMSE: 0.07196140343732449
LogLoss: 0.020926043953276
Mean Per-Class Error: 0.01464455125281161
AUC: 0.996932761638644
pr_auc: 0.8813528916590876
Gini: 0.9938655232772879

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.33309218897061765: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.33 %, avg score:  6.08 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.044750181102994446
RMSE: 0.21154238606717674
LogLoss: 0.24792163839974007
Mean Per-Class Error: 0.15058599929966965
AUC: 0.9091593504647071
pr_auc: 0.4490691092769276
Gini: 0.8183187009294142

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.28909050154675664: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.19 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.03827381863007762
RMSE: 0.19563695619712965
LogLoss: 0.14106045334794498
Mean Per-Class Error: 0.13236579247416214
AUC: 0.9324975415348415
pr_auc: 0.5422978933942193
Gini: 0.8649950830696831

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23701287227870213: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.07 %, avg score:  5.79 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_5129


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.2785398711774593
RMSE: 0.5277687667695573
LogLoss: 3.613008720953664
Mean Per-Class Error: 0.18914235532583845
AUC: 0.8368174048728454
pr_auc: 0.028031306738713322
Gini: 0.6736348097456908

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999869933640717: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.07 %, avg score: 34.04 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.27897756547198943
RMSE: 0.528183268830043
LogLoss: 3.600615528619235
Mean Per-Class Error: 0.18656399373921562
AUC: 0.8398672009146483
pr_auc: 0.030732474321422416
Gini: 0.6797344018292966

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999945647599826: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 33.85 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.27635057373464156
RMSE: 0.525690568428464
LogLoss: 3.6745348008296927
Mean Per-Class Error: 0.19550055297307223
AUC: 0.8348236033089962
pr_auc: 0.03099250687323766
Gini: 0.6696472066179924

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999911624963306: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.07 %, avg score: 33.72 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.936997
dtype: float64
GBM acc:predict    0.944056
dtype: float64
DL acc:predict    0.944056
dtype: float64
B acc:predict    0.863747
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_5148


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04083393314523418
RMSE: 0.202074078360472
LogLoss: 0.14144524804349648
Null degrees of freedom: 81813
Residual degrees of freedom: 81513
Null deviance: 37997.80423124849
Residual deviance: 23144.403046861247
AIC: 23746.403046861247
AUC: 0.9284277178498215
pr_auc: 0.49610519886353593
Gini: 0.856855435699643

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1923180872741826: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score:  6.19 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04181923788107111
RMSE: 0.2044975253666193
LogLoss: 0.14546250576926387
Null degrees of freedom: 81813
Residual degrees of freedom: 81513
Null deviance: 37999.88477215276
Residual deviance: 23801.73889401314
AIC: 24403.73889401314
AUC: 0.9231616370070282
pr_auc: 0.4770483938388879
Gini: 0.8463232740140565

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.17928055196015297: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score: 191.53 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_5166


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.035857007153339386
RMSE: 0.1893594654442692
LogLoss: 0.12766967524633033
Mean Per-Class Error: 0.11905613401921955
AUC: 0.9459819070707892
pr_auc: 0.610996930467613
Gini: 0.8919638141415784

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20325317752747002: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score:  6.21 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03831873442476682
RMSE: 0.1957517162754054
LogLoss: 0.13797514223998392
Mean Per-Class Error: 0.15280300888913256
AUC: 0.9191888402782759
pr_auc: 0.48895224537385384
Gini: 0.8383776805565517

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.24244256552838414: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.05 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.040654505596422925
RMSE: 0.2016296247985968
LogLoss: 0.14456718298490756
Mean Per-Class Error: 0.15104083388066802
AUC: 0.9200812345780164
pr_auc: 0.5006833930758441
Gini: 0.8401624691560328

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21795050969949148: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score:  6.11 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_5359


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 935,526 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.00413286811965406
RMSE: 0.06428738694062824
LogLoss: 0.01899878483193085
Mean Per-Class Error: 0.012033120857216861
AUC: 0.9976962498797963
pr_auc: 0.9283155016982169
Gini: 0.9953924997595927

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.31739744648026896: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score:  5.86 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.045084956019820724
RMSE: 0.21233218319374178
LogLoss: 0.26389968910405287
Mean Per-Class Error: 0.1500653922904599
AUC: 0.9100316519780114
pr_auc: 0.4476361003268477
Gini: 0.8200633039560228

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.05620922781092502: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  4.92 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.03946856994135385
RMSE: 0.19866698251434195
LogLoss: 0.1470058335042663
Mean Per-Class Error: 0.13067889535928423
AUC: 0.9341680694079305
pr_auc: 0.5430382660145967
Gini: 0.868336138815861

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.26050847599734295: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score:  5.83 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_5460


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.1597648748611207
RMSE: 0.39970598552075837
LogLoss: 1.800973068536119
Mean Per-Class Error: 0.16894437527416029
AUC: 0.8787924334933565
pr_auc: 0.06590926181970251
Gini: 0.7575848669867129

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999823618030366: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score: 21.46 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.1600472959393258
RMSE: 0.40005911555584606
LogLoss: 1.8054665587591445
Mean Per-Class Error: 0.16420615868810406
AUC: 0.8814134543582908
pr_auc: 0.06271234600902977
Gini: 0.7628269087165815

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999918279560105: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 21.38 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.15977375903760271
RMSE: 0.3997170987556108
LogLoss: 1.8340341013816073
Mean Per-Class Error: 0.1701077517016174
AUC: 0.8770490380678078
pr_auc: 0.07077362443051488
Gini: 0.7540980761356155

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999877182776558: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score: 21.40 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.932367
dtype: float64
GBM acc:predict    0.941931
dtype: float64
DL acc:predict    0.93267
dtype: float64
B acc:predict    0.925004
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_5479


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04037555916382522
RMSE: 0.2009367043718624
LogLoss: 0.1401816663811659
Null degrees of freedom: 87566
Residual degrees of freedom: 87266
Null deviance: 41118.322082843006
Residual deviance: 24550.575959999107
AIC: 25152.575959999107
AUC: 0.9307438610415983
pr_auc: 0.5184301071302444
Gini: 0.8614877220831967

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22392300334010595: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.29 %, avg score:  6.29 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04144884274867635
RMSE: 0.20358988862091446
LogLoss: 0.14427572179575557
Null degrees of freedom: 87566
Residual degrees of freedom: 87266
Null deviance: 41120.932925594636
Residual deviance: 25267.584260977856
AIC: 25869.584260977856
AUC: 0.9256255903275631
pr_auc: 0.4978291906729281
Gini: 0.8512511806551262

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20800517647033498: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.29 %, avg score: 184.73 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_5497


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03583388405335146
RMSE: 0.18929839950023736
LogLoss: 0.12803809354880005
Mean Per-Class Error: 0.12396897753323621
AUC: 0.9457295486315248
pr_auc: 0.6240013423726738
Gini: 0.8914590972630496

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23758372199138886: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.29 %, avg score:  6.31 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03861319118816171
RMSE: 0.19650239486622473
LogLoss: 0.13907040439041685
Mean Per-Class Error: 0.15423910796478102
AUC: 0.9177621281604516
pr_auc: 0.4842411511976346
Gini: 0.8355242563209031

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23191650996379817: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.16 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04049893749625861
RMSE: 0.2012434781458982
LogLoss: 0.14425611142606065
Mean Per-Class Error: 0.15138770931289613
AUC: 0.921825503948179
pr_auc: 0.5234415007176084
Gini: 0.8436510078963579

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.200014713809405: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.29 %, avg score:  6.21 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_5692


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 912,299 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.007087486424955267
RMSE: 0.08418721057830142
LogLoss: 0.026292051501968734
Mean Per-Class Error: 0.020624527948221782
AUC: 0.9976082135631158
pr_auc: 0.9292043240321846
Gini: 0.9952164271262316

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.47686342134157433: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.29 %, avg score:  6.08 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.0448811814702642
RMSE: 0.21185179128405832
LogLoss: 0.23614433174561927
Mean Per-Class Error: 0.13892781956790468
AUC: 0.9153765752159
pr_auc: 0.4628195544088192
Gini: 0.8307531504318

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19088938443614653: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.57 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.03906230857770972
RMSE: 0.19764186949558465
LogLoss: 0.1448129738160398
Mean Per-Class Error: 0.13289275464414263
AUC: 0.9336268532348031
pr_auc: 0.5576161968720409
Gini: 0.8672537064696062

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.24959831063051996: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.29 %, avg score:  5.77 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_5793


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.1451974716463685
RMSE: 0.3810478600469612
LogLoss: 1.6998955281721189
Mean Per-Class Error: 0.17212434978882518
AUC: 0.8764117020490765
pr_auc: 0.07630893722290788
Gini: 0.752823404098153

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999876916678463: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.29 %, avg score: 19.59 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.1456714172776151
RMSE: 0.3816692511555196
LogLoss: 1.7165912401265058
Mean Per-Class Error: 0.16163392130143317
AUC: 0.8788095755407521
pr_auc: 0.06552725996103273
Gini: 0.7576191510815042

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999861553711066: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 19.59 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.145278110805027
RMSE: 0.38115365773533777
LogLoss: 1.7025852767505807
Mean Per-Class Error: 0.1733294232173892
AUC: 0.8761356508296566
pr_auc: 0.07472868725864482
Gini: 0.7522713016593132

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999886185795719: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.29 %, avg score: 19.57 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.935859
dtype: float64
GBM acc:predict    0.938743
dtype: float64
DL acc:predict    0.940565
dtype: float64
B acc:predict    0.925535
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_5812


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.03987968374127502
RMSE: 0.1996989828248382
LogLoss: 0.13881405756607007
Null degrees of freedom: 93606
Residual degrees of freedom: 93306
Null deviance: 42999.35683860839
Residual deviance: 25987.93497317424
AIC: 26589.93497317424
AUC: 0.9296619201141679
pr_auc: 0.5006765706035483
Gini: 0.8593238402283359

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19953385463617657: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.10 %, avg score:  6.10 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04078951820418706
RMSE: 0.20196415078965638
LogLoss: 0.142389528410204
Null degrees of freedom: 93606
Residual degrees of freedom: 93306
Null deviance: 43000.88310760552
Residual deviance: 26657.31317178793
AIC: 27259.31317178793
AUC: 0.9247938354844257
pr_auc: 0.48234386329371987
Gini: 0.8495876709688515

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19366248470187628: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.10 %, avg score: 179.75 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_5830


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03552360642361449
RMSE: 0.18847707134719197
LogLoss: 0.12665124378091416
Mean Per-Class Error: 0.12326765548537177
AUC: 0.9454766279300286
pr_auc: 0.6050744186951694
Gini: 0.8909532558600572

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2324274313265826: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.10 %, avg score:  6.12 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03890609628613187
RMSE: 0.19724628332653538
LogLoss: 0.13949585668394981
Mean Per-Class Error: 0.1555283908012033
AUC: 0.9178415481519295
pr_auc: 0.4800361695091005
Gini: 0.8356830963038591

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.219682400576266: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.04 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.039765857235217944
RMSE: 0.19941378396494547
LogLoss: 0.14161497279253088
Mean Per-Class Error: 0.1497435167063994
AUC: 0.9216870361221927
pr_auc: 0.5070626999125762
Gini: 0.8433740722443854

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21645666264232458: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.10 %, avg score:  6.02 %


Cross-Validation Metrics Summary: 

Scoring History: 

See the whole table with table.as_data_frame()

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_6027


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 973,932 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.00697678779845739
RMSE: 0.08352716802608233
LogLoss: 0.027795227786542927
Mean Per-Class Error: 0.021330189104345365
AUC: 0.9956276191477537
pr_auc: 0.9064404486445674
Gini: 0.9912552382955073

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23632383999285064: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.89 %, avg score:  5.33 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04218840442764552
RMSE: 0.20539816072118447
LogLoss: 0.22324953429679933
Mean Per-Class Error: 0.1310187275082162
AUC: 0.9218811568950898
pr_auc: 0.4824910904980719
Gini: 0.8437623137901795

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.15003920848911126: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  4.64 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.03831281056488766
RMSE: 0.1957365846357999
LogLoss: 0.13878943799884624
Mean Per-Class Error: 0.1283232195587649
AUC: 0.936484758777532
pr_auc: 0.5542123981737086
Gini: 0.872969517555064

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2835155106175367: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.10 %, avg score:  6.40 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_6134


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.15700829689382656
RMSE: 0.3962427247203746
LogLoss: 1.7793921968857616
Mean Per-Class Error: 0.1681337378022132
AUC: 0.8809479696328378
pr_auc: 0.06282813655112775
Gini: 0.7618959392656757

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999785352603838: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.10 %, avg score: 21.09 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.1601875181227487
RMSE: 0.4002343290158263
LogLoss: 1.828548133129078
Mean Per-Class Error: 0.16501512460395473
AUC: 0.877803167096287
pr_auc: 0.05891994029667594
Gini: 0.7556063341925741

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999846662407688: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 21.31 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.15708446857510813
RMSE: 0.3963388305164006
LogLoss: 1.7869255849601835
Mean Per-Class Error: 0.16917215140540076
AUC: 0.8803437176407448
pr_auc: 0.0663120085626574
Gini: 0.7606874352814896

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999851691348648: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.10 %, avg score: 21.06 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.933581
dtype: float64
GBM acc:predict    0.937225
dtype: float64
DL acc:predict    0.945575
dtype: float64
B acc:predict    0.919994
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_6153


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04122926925702667
RMSE: 0.20304991814090118
LogLoss: 0.1431324819804035
Null degrees of freedom: 99140
Residual degrees of freedom: 98840
Null deviance: 46733.27907921893
Residual deviance: 28380.59479203836
AIC: 28982.59479203836
AUC: 0.9278934573521395
pr_auc: 0.5033238599810942
Gini: 0.8557869147042789

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19970329709995313: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.32 %, avg score:  6.32 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.042105567166159034
RMSE: 0.2051964111922015
LogLoss: 0.1464908606507997
Null degrees of freedom: 99140
Residual degrees of freedom: 98840
Null deviance: 46734.41469043316
Residual deviance: 29046.500831561865
AIC: 29648.500831561865
AUC: 0.9235431131028659
pr_auc: 0.4870918240290606
Gini: 0.8470862262057317

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20657253897501748: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.32 %, avg score: 172.67 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_6171


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.037039004072300825
RMSE: 0.1924552001695481
LogLoss: 0.13175906434704912
Mean Per-Class Error: 0.1280197008444386
AUC: 0.9425518822554503
pr_auc: 0.5997613850420536
Gini: 0.8851037645109006

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22283437559496802: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.32 %, avg score:  6.34 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.038755167532826686
RMSE: 0.19686332195923822
LogLoss: 0.1386348393571616
Mean Per-Class Error: 0.14945091570300928
AUC: 0.9214942349312959
pr_auc: 0.48764570172009375
Gini: 0.8429884698625918

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23058281960447546: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.12 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.040906137091657996
RMSE: 0.20225265657503239
LogLoss: 0.14544448176726124
Mean Per-Class Error: 0.15216648803021193
AUC: 0.9213041967610024
pr_auc: 0.5135161138371337
Gini: 0.8426083935220048

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22977091492383664: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.32 %, avg score:  6.26 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_6358


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,100,743 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.006078903277541712
RMSE: 0.07796732185692742
LogLoss: 0.023786981862728436
Mean Per-Class Error: 0.01533385132699383
AUC: 0.9981914349782178
pr_auc: 0.880695789220477
Gini: 0.9963828699564357

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.5107570773503094: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.37 %, avg score:  6.79 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04744474361300486
RMSE: 0.2178181434431137
LogLoss: 0.2417538186309466
Mean Per-Class Error: 0.14849618826229483
AUC: 0.9110655884673312
pr_auc: 0.43025133179745306
Gini: 0.8221311769346624

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.36072861380923055: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.19 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.03915394162826063
RMSE: 0.19787354959231065
LogLoss: 0.14076187750401684
Mean Per-Class Error: 0.13244793173352398
AUC: 0.9337379968116079
pr_auc: 0.5554946275597933
Gini: 0.8674759936232157

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23779901893086355: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.32 %, avg score:  5.84 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_6467


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.19855850965998476
RMSE: 0.4455990458472558
LogLoss: 2.563163166235468
Mean Per-Class Error: 0.19155783417868832
AUC: 0.8496309060111633
pr_auc: 0.04832091308588071
Gini: 0.6992618120223266

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999908290697772: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.32 %, avg score: 25.16 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.20070534061856352
RMSE: 0.4480014962235768
LogLoss: 2.6126136521286116
Mean Per-Class Error: 0.18509076871801577
AUC: 0.84694486164257
pr_auc: 0.04498247109353655
Gini: 0.69388972328514

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.999991568276565: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 25.12 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.19905576259738672
RMSE: 0.4461566570134158
LogLoss: 2.604596622597406
Mean Per-Class Error: 0.19360140887242516
AUC: 0.8475099816609453
pr_auc: 0.04996369276435518
Gini: 0.6950199633218905

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999892902677262: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.32 %, avg score: 25.16 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.931911
dtype: float64
GBM acc:predict    0.939047
dtype: float64
DL acc:predict    0.940489
dtype: float64
B acc:predict    0.895552
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_6486


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04144343368146907
RMSE: 0.20357660396388647
LogLoss: 0.14376294051912525
Null degrees of freedom: 105159
Residual degrees of freedom: 104859
Null deviance: 49190.01602279799
Residual deviance: 30236.22164998242
AIC: 30838.22164998242
AUC: 0.9262793809367407
pr_auc: 0.4908192097626937
Gini: 0.8525587618734813

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.17094810629331095: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score:  6.25 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04231776789821058
RMSE: 0.20571282871568944
LogLoss: 0.14713076836721514
Null degrees of freedom: 105159
Residual degrees of freedom: 104859
Null deviance: 49191.61407471733
Residual deviance: 30944.54320299269
AIC: 31546.54320299269
AUC: 0.922119184538713
pr_auc: 0.4739495867165424
Gini: 0.844238369077426

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.18333578357743296: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score: 167.19 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_6504


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03695395044830288
RMSE: 0.19223410323952117
LogLoss: 0.13135784951302137
Mean Per-Class Error: 0.12687370316973
AUC: 0.9416284975422295
pr_auc: 0.5952752806366938
Gini: 0.8832569950844591

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21204700545347555: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score:  6.27 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.0388175828995004
RMSE: 0.19702178280459345
LogLoss: 0.13911424868232414
Mean Per-Class Error: 0.1518097843741958
AUC: 0.9184269146229818
pr_auc: 0.480853659741347
Gini: 0.8368538292459635

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23698909437436505: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.09 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04088181981748679
RMSE: 0.2021925315571443
LogLoss: 0.145303542766946
Mean Per-Class Error: 0.15341736556550734
AUC: 0.9195320280488012
pr_auc: 0.505621773414978
Gini: 0.8390640560976024

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19047764565239264: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score:  6.19 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_6687


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,099,677 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.005839026905252254
RMSE: 0.07641352566955836
LogLoss: 0.02320460071584795
Mean Per-Class Error: 0.015544334588630582
AUC: 0.9979518966120267
pr_auc: 0.9498363448956679
Gini: 0.9959037932240533

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.3291022753719946: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.58 %, avg score:  5.93 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.042235845400693604
RMSE: 0.20551361366268076
LogLoss: 0.21579876934446737
Mean Per-Class Error: 0.14667227071564481
AUC: 0.913062320962237
pr_auc: 0.4886388219885125
Gini: 0.8261246419244741

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.18099643902300522: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  4.55 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.03881404041836798
RMSE: 0.19701279252466825
LogLoss: 0.13941246105134975
Mean Per-Class Error: 0.13104139777398816
AUC: 0.9333064704426356
pr_auc: 0.5528720601297914
Gini: 0.8666129408852712

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.32131092312166665: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score:  6.62 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_6798


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.11556593124832557
RMSE: 0.3399498952026983
LogLoss: 1.3284450475495628
Mean Per-Class Error: 0.18592082624759787
AUC: 0.8572806203481196
pr_auc: 0.13288727155934255
Gini: 0.7145612406962392

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.99085343379706: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score: 14.88 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.11145562079717704
RMSE: 0.3338496979138622
LogLoss: 1.3112425886552415
Mean Per-Class Error: 0.17516611751205535
AUC: 0.8707066269812809
pr_auc: 0.12322311025019109
Gini: 0.7414132539625617

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9835720899304422: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 14.51 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.13507721957462565
RMSE: 0.36752852892615784
LogLoss: 1.6583722784389066
Mean Per-Class Error: 0.21623310776458393
AUC: 0.82460197142758
pr_auc: 0.11117972242303133
Gini: 0.6492039428551599

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999923783117485: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score: 16.39 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.926522
dtype: float64
GBM acc:predict    0.93973
dtype: float64
DL acc:predict    0.945195
dtype: float64
B acc:predict    0.904129
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_6817


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04031073532607068
RMSE: 0.20077533545251688
LogLoss: 0.14036801743513846
Null degrees of freedom: 110854
Residual degrees of freedom: 110554
Null deviance: 51081.188188542204
Residual deviance: 31120.993145544548
AIC: 31722.993145544548
AUC: 0.926610390002793
pr_auc: 0.49796039748517007
Gini: 0.853220780005586

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1890622149674929: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score:  6.13 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04119550839086716
RMSE: 0.20296676671531022
LogLoss: 0.14369079694859865
Null degrees of freedom: 110854
Residual degrees of freedom: 110554
Null deviance: 51084.28002309035
Residual deviance: 31857.686591473805
AIC: 32459.686591473805
AUC: 0.9222717993674934
pr_auc: 0.4806140746180576
Gini: 0.8445435987349867

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.18825044246361314: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score: 165.01 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_6835


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03628409929634902
RMSE: 0.19048385573677634
LogLoss: 0.12936251278548405
Mean Per-Class Error: 0.12701616689523731
AUC: 0.9418738670680433
pr_auc: 0.5928757825550645
Gini: 0.8837477341360866

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19454252312138648: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score:  6.15 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03829627765444463
RMSE: 0.1956943475280894
LogLoss: 0.13750774994368292
Mean Per-Class Error: 0.14651532921854105
AUC: 0.9198565799410204
pr_auc: 0.49697942380051274
Gini: 0.8397131598820409

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22766399660532666: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.03 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04007335966591753
RMSE: 0.2001833151536799
LogLoss: 0.14260571841837333
Mean Per-Class Error: 0.15216465054871464
AUC: 0.921414741848427
pr_auc: 0.5032172690482947
Gini: 0.842829483696854

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19949045599721063: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score:  6.05 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_7016


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,199,754 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.006175825358924546
RMSE: 0.07858641968511192
LogLoss: 0.030283380632109433
Mean Per-Class Error: 0.017751140745706717
AUC: 0.9955278022900398
pr_auc: 0.850224001432413
Gini: 0.9910556045800796

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.6290457741033754: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.32 %, avg score:  6.90 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04498792569328435
RMSE: 0.2121035730328095
LogLoss: 0.20883305136775987
Mean Per-Class Error: 0.14568147204374116
AUC: 0.9182409684809161
pr_auc: 0.4692676994051635
Gini: 0.8364819369618322

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.31134601872355466: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.44 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.038091589588208526
RMSE: 0.19517066784793385
LogLoss: 0.13601794584842003
Mean Per-Class Error: 0.12851636923705279
AUC: 0.9345914064557053
pr_auc: 0.5566931067250583
Gini: 0.8691828129114105

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.26984134360951484: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score:  6.47 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_7127


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.16882055514372227
RMSE: 0.41087778614050463
LogLoss: 3.9984374544408934
Mean Per-Class Error: 0.4486828534929914
AUC: 0.5538822767604392
pr_auc: 0.06477085968352676
Gini: 0.10776455352087844

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.11627708785301118: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score: 14.02 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.16617820773478223
RMSE: 0.40764961392693877
LogLoss: 3.910684703306637
Mean Per-Class Error: 0.4459137201463099
AUC: 0.5566374229531159
pr_auc: 0.0617470727845766
Gini: 0.11327484590623182

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.010801945461729112: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 13.95 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.15392346997539647
RMSE: 0.3923308170095697
LogLoss: 3.540573845132134
Mean Per-Class Error: 0.40160092391558555
AUC: 0.6035477175354959
pr_auc: 0.07384853875632759
Gini: 0.2070954350709917

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9478311922602795: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.13 %, avg score: 13.65 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.930545
dtype: float64
GBM acc:predict    0.939957
dtype: float64
DL acc:predict    0.939654
dtype: float64
B acc:predict    0.809018
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_7146


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04032480087578581
RMSE: 0.20081036047919892
LogLoss: 0.14022671031270126
Null degrees of freedom: 116920
Residual degrees of freedom: 116620
Null deviance: 54368.9252769506
Residual deviance: 32790.89439294269
AIC: 33392.89439294269
AUC: 0.9301656518486252
pr_auc: 0.5054868478470884
Gini: 0.8603313036972504

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19729738045907483: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score:  6.20 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.041109170495686645
RMSE: 0.20275396542530713
LogLoss: 0.14336794769223116
Null degrees of freedom: 116920
Residual degrees of freedom: 116620
Null deviance: 54373.88665594231
Residual deviance: 33525.44762424672
AIC: 34127.44762424672
AUC: 0.9259464498440091
pr_auc: 0.48954518807334607
Gini: 0.8518928996880182

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1916542278188926: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score: 160.94 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_7164


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.036365866400771045
RMSE: 0.19069836496617124
LogLoss: 0.12950011988294197
Mean Per-Class Error: 0.127641587773895
AUC: 0.9424234783212707
pr_auc: 0.5987832714421437
Gini: 0.8848469566425414

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2283773627936528: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score:  6.22 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.0382251109607209
RMSE: 0.1955124317293427
LogLoss: 0.137518162771498
Mean Per-Class Error: 0.15265376680687337
AUC: 0.9193587020263172
pr_auc: 0.49229428685466264
Gini: 0.8387174040526344

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2242379582636348: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.08 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.03993592917669962
RMSE: 0.19983975874860244
LogLoss: 0.14202052323869965
Mean Per-Class Error: 0.14892481310293082
AUC: 0.9234526561163989
pr_auc: 0.5186522534821901
Gini: 0.8469053122327979

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20479320037660326: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score:  6.14 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_7349


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,300,242 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.005941842703781265
RMSE: 0.07708334906957057
LogLoss: 0.023896327863728816
Mean Per-Class Error: 0.01772475432998366
AUC: 0.9974109757382367
pr_auc: 0.9306351051157637
Gini: 0.9948219514764733

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.3718146240632751: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.98 %, avg score:  5.74 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04299898101214549
RMSE: 0.20736195652082734
LogLoss: 0.21652357316767062
Mean Per-Class Error: 0.15155823499879761
AUC: 0.9107681062814568
pr_auc: 0.4734679634599488
Gini: 0.8215362125629135

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.12311475772274827: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.13 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.037657691219793836
RMSE: 0.19405589715284058
LogLoss: 0.1345460656130533
Mean Per-Class Error: 0.13039356387636214
AUC: 0.937328471155422
pr_auc: 0.5658021624872405
Gini: 0.8746569423108439

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.27434792920528783: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score:  6.05 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_7462


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.15762314855207307
RMSE: 0.3970178189352124
LogLoss: 1.8328175552420165
Mean Per-Class Error: 0.1678681878510454
AUC: 0.8791008087179163
pr_auc: 0.06585804477041063
Gini: 0.7582016174358326

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999862181630564: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score: 21.16 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.15885440804426085
RMSE: 0.3985654375937041
LogLoss: 1.8492490452581594
Mean Per-Class Error: 0.16586786116583907
AUC: 0.875897034565099
pr_auc: 0.060725729166570916
Gini: 0.7517940691301981

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999878426510801: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 21.08 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.15822049207847091
RMSE: 0.3977693956031194
LogLoss: 1.8630713430230896
Mean Per-Class Error: 0.17081117836116966
AUC: 0.8765484661997286
pr_auc: 0.0670047760574075
Gini: 0.7530969323994572

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.999987262046839: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score: 21.13 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.933657
dtype: float64
GBM acc:predict    0.938363
dtype: float64
DL acc:predict    0.937604
dtype: float64
B acc:predict    0.921436
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_7481


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04062521990074462
RMSE: 0.20155698921333545
LogLoss: 0.14123122592899678
Null degrees of freedom: 122918
Residual degrees of freedom: 122618
Null deviance: 57434.583863626525
Residual deviance: 34720.00211993272
AIC: 35322.00211993272
AUC: 0.9292981468465097
pr_auc: 0.5033713908680538
Gini: 0.8585962936930194

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19586397264539943: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score:  6.24 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04129948196660054
RMSE: 0.20322273978716196
LogLoss: 0.1438273769307751
Null degrees of freedom: 122918
Residual degrees of freedom: 122618
Null deviance: 57439.67482931474
Residual deviance: 35358.23468990789
AIC: 35960.23468990789
AUC: 0.925986234333905
pr_auc: 0.491095723896177
Gini: 0.8519724686678101

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.17749925352442528: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score: 156.62 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_7499


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.036826674074862294
RMSE: 0.19190277245225587
LogLoss: 0.13080404625809586
Mean Per-Class Error: 0.12660963165873684
AUC: 0.9420675209217615
pr_auc: 0.5951506209064953
Gini: 0.884135041843523

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.24850766883233213: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score:  6.26 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03809174280439992
RMSE: 0.19517106036602844
LogLoss: 0.13706575790303815
Mean Per-Class Error: 0.14959172007037058
AUC: 0.9203857828722826
pr_auc: 0.49803374534048467
Gini: 0.8407715657445651

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.24252130993748042: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.07 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.040157468444114414
RMSE: 0.20039328442868143
LogLoss: 0.14275050916414994
Mean Per-Class Error: 0.14805022161165593
AUC: 0.9233147644137457
pr_auc: 0.5172793549221598
Gini: 0.8466295288274914

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22391936083825015: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score:  6.21 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_7676


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,300,721 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.005179771210833838
RMSE: 0.07197062741725847
LogLoss: 0.022149304947257732
Mean Per-Class Error: 0.014653787959227005
AUC: 0.9979789366107454
pr_auc: 0.9420139681382057
Gini: 0.9959578732214909

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.43686669119289595: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.36 %, avg score:  6.36 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.0409020056988014
RMSE: 0.20224244287191895
LogLoss: 0.1809269330918667
Mean Per-Class Error: 0.1384093219874194
AUC: 0.9302391881230725
pr_auc: 0.4970667049706657
Gini: 0.860478376246145

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2590762007426584: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.49 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.036644367844638176
RMSE: 0.19142718679601958
LogLoss: 0.13214026383994276
Mean Per-Class Error: 0.12156011654976595
AUC: 0.9408250473686929
pr_auc: 0.5897343749295788
Gini: 0.8816500947373858

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22833575487006516: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score:  5.82 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_7789


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.21562444814802875
RMSE: 0.4643537963105597
LogLoss: 2.739964112038536
Mean Per-Class Error: 0.18408734528083004
AUC: 0.8525228188835313
pr_auc: 0.04172242619739837
Gini: 0.7050456377670626

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999920301618357: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score: 27.19 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.2179573622726197
RMSE: 0.46685903897495623
LogLoss: 2.7864143249712776
Mean Per-Class Error: 0.1819611147908924
AUC: 0.8482239137496783
pr_auc: 0.03853067251200419
Gini: 0.6964478274993566

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999902319362873: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 27.13 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.21533498330557432
RMSE: 0.46404200597098355
LogLoss: 2.786518895748108
Mean Per-Class Error: 0.18598416366713
AUC: 0.8502307123014565
pr_auc: 0.04279184631054157
Gini: 0.700461424602913

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999910652580576: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.24 %, avg score: 27.15 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.932746
dtype: float64
GBM acc:predict    0.940944
dtype: float64
DL acc:predict    0.94436
dtype: float64
B acc:predict    0.888113
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_7808


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04095200443230111
RMSE: 0.20236601600145493
LogLoss: 0.142066559697763
Null degrees of freedom: 128629
Residual degrees of freedom: 128329
Null deviance: 60023.99038957982
Residual deviance: 36548.04314784651
AIC: 37150.04314784651
AUC: 0.929043972487303
pr_auc: 0.4980215715411303
Gini: 0.8580879449746059

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19971141244488244: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.23 %, avg score:  6.23 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.041610364119588926
RMSE: 0.20398618609991442
LogLoss: 0.1445624793749045
Null degrees of freedom: 128629
Residual degrees of freedom: 128329
Null deviance: 60024.86700109555
Residual deviance: 37190.14344398794
AIC: 37792.14344398794
AUC: 0.9258092802080482
pr_auc: 0.4849288121438919
Gini: 0.8516185604160964

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.18811217476080122: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.23 %, avg score: 152.08 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_7826


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03720413755512683
RMSE: 0.19288374103362582
LogLoss: 0.13204965002949584
Mean Per-Class Error: 0.12926994104926903
AUC: 0.940444426594126
pr_auc: 0.5851600144335388
Gini: 0.8808888531882519

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20016308603141703: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.23 %, avg score:  6.25 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03797489081233644
RMSE: 0.19487147254623094
LogLoss: 0.13645736488749427
Mean Per-Class Error: 0.14391609114419635
AUC: 0.9210621184570795
pr_auc: 0.5024327211521968
Gini: 0.842124236914159

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23731478350295354: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.10 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04039032429420844
RMSE: 0.2009734417633545
LogLoss: 0.14349411193880615
Mean Per-Class Error: 0.14903024162293987
AUC: 0.9225721136653244
pr_auc: 0.5127053231965123
Gini: 0.8451442273306489

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20638414100475957: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.23 %, avg score:  6.19 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_8001


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,399,395 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.007480987354178991
RMSE: 0.08649270116130604
LogLoss: 0.03066295140522912
Mean Per-Class Error: 0.02347414697073713
AUC: 0.9952404437336391
pr_auc: 0.8961396938752548
Gini: 0.9904808874672781

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.44049000255294213: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.74 %, avg score:  5.71 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04222648405530324
RMSE: 0.20549083691323863
LogLoss: 0.19496077399692008
Mean Per-Class Error: 0.14612118246136574
AUC: 0.9204867717724686
pr_auc: 0.47931129120988214
Gini: 0.8409735435449373

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.24577933395615567: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.57 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.037500293875351885
RMSE: 0.19364992609178008
LogLoss: 0.13344541885535008
Mean Per-Class Error: 0.12722991676564244
AUC: 0.9385417448750669
pr_auc: 0.5673573678176856
Gini: 0.8770834897501338

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2481401307707861: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.23 %, avg score:  6.09 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_8118


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.16656046676130917
RMSE: 0.4081182019480498
LogLoss: 2.030862696840463
Mean Per-Class Error: 0.17749712630974157
AUC: 0.865643232319427
pr_auc: 0.05921895126038144
Gini: 0.7312864646388539

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999891196293236: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.23 %, avg score: 21.84 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.1682755925310054
RMSE: 0.41021408134168846
LogLoss: 2.044787564430561
Mean Per-Class Error: 0.1758370213178867
AUC: 0.865199857824504
pr_auc: 0.05548177559029267
Gini: 0.7303997156490081

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999875067351192: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 21.86 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.16668670138839287
RMSE: 0.4082728271491906
LogLoss: 2.0361990127252403
Mean Per-Class Error: 0.17844810960329638
AUC: 0.8649243261487709
pr_auc: 0.06006155830586913
Gini: 0.7298486522975418

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999907784723702: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.23 %, avg score: 21.83 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.933278
dtype: float64
GBM acc:predict    0.940792
dtype: float64
DL acc:predict    0.942311
dtype: float64
B acc:predict    0.91339
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_8137


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.041327585212591486
RMSE: 0.20329187197866885
LogLoss: 0.1432867911379919
Null degrees of freedom: 134515
Residual degrees of freedom: 134215
Null deviance: 62505.460199193585
Residual deviance: 38548.73199343623
AIC: 39150.73199343623
AUC: 0.9263704564862065
pr_auc: 0.48515218785257247
Gini: 0.8527409129724131

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1858235396163303: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score:  6.20 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04196980160280654
RMSE: 0.2048653255258355
LogLoss: 0.1457819538171972
Null degrees of freedom: 134515
Residual degrees of freedom: 134215
Null deviance: 62510.270402416034
Residual deviance: 39220.01059934819
AIC: 39822.01059934819
AUC: 0.9229385861362055
pr_auc: 0.47277958795528385
Gini: 0.845877172272411

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1878656980529052: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score: 148.10 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_8155


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.037549039857726264
RMSE: 0.1937757463093002
LogLoss: 0.13319256080585823
Mean Per-Class Error: 0.13070379132342747
AUC: 0.93853404052311
pr_auc: 0.571363292369596
Gini: 0.8770680810462199

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22099222808450256: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score:  6.22 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.038071617027274005
RMSE: 0.1951194942266764
LogLoss: 0.13684010886399012
Mean Per-Class Error: 0.14781336618417007
AUC: 0.9214979264315638
pr_auc: 0.5013027021144393
Gini: 0.8429958528631276

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22340309793122043: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.01 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.040724725573135466
RMSE: 0.20180368077202027
LogLoss: 0.14446112070178743
Mean Per-Class Error: 0.1517409787076749
AUC: 0.9205011015137143
pr_auc: 0.49796592175817916
Gini: 0.8410022030274287

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20714683551486016: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score:  6.15 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_8328


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,400,097 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.008316686354941934
RMSE: 0.0911958680804231
LogLoss: 0.033544417873347054
Mean Per-Class Error: 0.023399462423441375
AUC: 0.9964795034438554
pr_auc: 0.8996389492125676
Gini: 0.9929590068877108

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.629768920306969: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.98 %, avg score:  7.17 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.046577814997241576
RMSE: 0.21581894031164545
LogLoss: 0.1945589611183828
Mean Per-Class Error: 0.14587068780032997
AUC: 0.9218448219853099
pr_auc: 0.4586174569273781
Gini: 0.8436896439706199

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.3029463360987012: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  7.39 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.037933705800456116
RMSE: 0.19476577163468975
LogLoss: 0.13639201900184997
Mean Per-Class Error: 0.12760570511332625
AUC: 0.9365102929834461
pr_auc: 0.5581322252504708
Gini: 0.8730205859668922

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2705565539730893: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score:  6.08 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_8453


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.19136770631587005
RMSE: 0.43745594785746145
LogLoss: 2.395198253567365
Mean Per-Class Error: 0.18828926331613394
AUC: 0.8564429634737812
pr_auc: 0.0507584507055972
Gini: 0.7128859269475625

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999903500144172: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score: 24.44 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.19289434483232537
RMSE: 0.43919738709642314
LogLoss: 2.4002883181194377
Mean Per-Class Error: 0.18099626209230013
AUC: 0.8549407039585539
pr_auc: 0.04706898080969405
Gini: 0.7098814079171079

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999856602185679: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 24.40 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.19137006618274746
RMSE: 0.43745864511145216
LogLoss: 2.405573642126347
Mean Per-Class Error: 0.18920143646594223
AUC: 0.8546431294335676
pr_auc: 0.051032499081173834
Gini: 0.7092862588671351

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999901067611997: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.20 %, avg score: 24.41 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.930469
dtype: float64
GBM acc:predict    0.938667
dtype: float64
DL acc:predict    0.935555
dtype: float64
B acc:predict    0.90041
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_8472


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04039146014972387
RMSE: 0.20097626762810544
LogLoss: 0.1405532120786413
Null degrees of freedom: 140719
Residual degrees of freedom: 140419
Null deviance: 65026.780273976095
Residual deviance: 39557.29600741281
AIC: 40159.29600741281
AUC: 0.927357719196126
pr_auc: 0.49829388798894836
Gini: 0.854715438392252

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19171777523536648: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score:  6.15 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04098399770369394
RMSE: 0.2024450486025626
LogLoss: 0.1428365223433161
Null degrees of freedom: 140719
Residual degrees of freedom: 140419
Null deviance: 65029.15145552158
Residual deviance: 40199.91084830288
AIC: 40801.91084830288
AUC: 0.924296043275073
pr_auc: 0.48673524532261464
Gini: 0.848592086550146

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19329799148250698: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score: 145.97 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_8490


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03693329558615364
RMSE: 0.1921803725309992
LogLoss: 0.13152620039145796
Mean Per-Class Error: 0.13082668176076828
AUC: 0.9387802261828008
pr_auc: 0.5812743557061291
Gini: 0.8775604523656015

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2322801933582987: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score:  6.17 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.038431944286552396
RMSE: 0.19604066998088024
LogLoss: 0.13774893068037172
Mean Per-Class Error: 0.1450245959389278
AUC: 0.9202404959688818
pr_auc: 0.49090450227107285
Gini: 0.8404809919377636

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22524828794028182: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.04 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.039933239675933235
RMSE: 0.1998330294919567
LogLoss: 0.14218904955387202
Mean Per-Class Error: 0.14987645288248685
AUC: 0.9218684221975585
pr_auc: 0.5134045239564735
Gini: 0.843736844395117

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19027422377627284: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score:  6.10 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_8671


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,500,248 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.00905261860080796
RMSE: 0.0951452500170553
LogLoss: 0.03661865786692964
Mean Per-Class Error: 0.028005145703075995
AUC: 0.9938198263144534
pr_auc: 0.8877968150261847
Gini: 0.9876396526289069

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.6163754957873716: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.94 %, avg score:  6.76 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04538428969050247
RMSE: 0.21303588826886063
LogLoss: 0.19985647564560088
Mean Per-Class Error: 0.13891769431002698
AUC: 0.9244537634317874
pr_auc: 0.4647333848379142
Gini: 0.8489075268635748

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.32835702082435037: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.64 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.037003838242492156
RMSE: 0.19236381739425987
LogLoss: 0.13245830320640162
Mean Per-Class Error: 0.12516637877101755
AUC: 0.9381683493562334
pr_auc: 0.5700377430255493
Gini: 0.8763366987124668

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2738182136186777: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score:  6.04 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_8798


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.1637772292377319
RMSE: 0.4046939945659336
LogLoss: 1.821641941055363
Mean Per-Class Error: 0.16998417653683662
AUC: 0.8802488949557699
pr_auc: 0.06424408328977058
Gini: 0.7604977899115397

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999875393899417: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score: 21.90 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.16406769606696758
RMSE: 0.4050527077640237
LogLoss: 1.8235170278338717
Mean Per-Class Error: 0.16390661980922316
AUC: 0.8819040547438942
pr_auc: 0.057835913661360774
Gini: 0.7638081094877884

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999860099066106: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 21.88 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.16434432414107303
RMSE: 0.40539403565059146
LogLoss: 1.8534749963632768
Mean Per-Class Error: 0.17230733704989154
AUC: 0.8783893713369493
pr_auc: 0.06460838985308678
Gini: 0.7567787426738986

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999852311439381: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score: 21.90 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.931228
dtype: float64
GBM acc:predict    0.938895
dtype: float64
DL acc:predict    0.940109
dtype: float64
B acc:predict    0.921512
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_8817


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.040255584081302274
RMSE: 0.20063794277579272
LogLoss: 0.13991742703836912
Null degrees of freedom: 145995
Residual degrees of freedom: 145695
Null deviance: 67391.61845055198
Residual deviance: 40854.76935578748
AIC: 41456.76935578748
AUC: 0.928109286869522
pr_auc: 0.5024327230584621
Gini: 0.856218573739044

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2046087159150026: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.14 %, avg score:  6.14 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.040843955878750206
RMSE: 0.20209887649056885
LogLoss: 0.14214369522589526
Null degrees of freedom: 145995
Residual degrees of freedom: 145695
Null deviance: 67392.49114296061
Residual deviance: 41504.821856399605
AIC: 42106.821856399605
AUC: 0.9252577541827718
pr_auc: 0.49002157540090063
Gini: 0.8505155083655436

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.183197297586837: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.14 %, avg score: 142.87 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_8835


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03699313073551061
RMSE: 0.1923359839850843
LogLoss: 0.13164161876580247
Mean Per-Class Error: 0.12978076022540352
AUC: 0.9389592188103613
pr_auc: 0.5777676518697761
Gini: 0.8779184376207225

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20772622310364783: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.14 %, avg score:  6.16 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03843491010230608
RMSE: 0.19604823412187644
LogLoss: 0.13762651993912975
Mean Per-Class Error: 0.14549162345853506
AUC: 0.921131993283579
pr_auc: 0.49218451434378835
Gini: 0.842263986567158

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23528828583430728: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.04 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.03988045193906729
RMSE: 0.1997009062049226
LogLoss: 0.1420812890152689
Mean Per-Class Error: 0.14996669268650953
AUC: 0.9218619089131482
pr_auc: 0.5123285751088413
Gini: 0.8437238178262965

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2204098795318872: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.14 %, avg score:  6.09 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_9008


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,599,694 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.007513350713678415
RMSE: 0.08667958648769857
LogLoss: 0.029971482934214685
Mean Per-Class Error: 0.027648496878798334
AUC: 0.9959229461711258
pr_auc: 0.8991125877688627
Gini: 0.9918458923422515

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.4552208677809972: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.89 %, avg score:  5.59 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04462224797536065
RMSE: 0.21123978786052747
LogLoss: 0.20214354735260515
Mean Per-Class Error: 0.1401070956963435
AUC: 0.9191713847555805
pr_auc: 0.45890899435633536
Gini: 0.8383427695111609

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1713869712019028: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.32 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.03677091764776293
RMSE: 0.19175744483008456
LogLoss: 0.1316928861708444
Mean Per-Class Error: 0.12150790485584495
AUC: 0.9397104939825988
pr_auc: 0.5740396536315391
Gini: 0.8794209879651975

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.25335401270873187: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.14 %, avg score:  5.90 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_9135


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.14926574485248034
RMSE: 0.38634925242904294
LogLoss: 1.6957721515597035
Mean Per-Class Error: 0.17063890319745778
AUC: 0.8793697055527068
pr_auc: 0.06839538394249513
Gini: 0.7587394111054135

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999786732990509: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.14 %, avg score: 20.14 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.1503053919952227
RMSE: 0.3876923935225228
LogLoss: 1.721098720236063
Mean Per-Class Error: 0.16081303711328898
AUC: 0.8809842910842886
pr_auc: 0.06466381308494389
Gini: 0.7619685821685771

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999901512045816: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 20.27 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.14916671878289664
RMSE: 0.3862210750113161
LogLoss: 1.7021748445438807
Mean Per-Class Error: 0.17202125293358228
AUC: 0.8802166855615324
pr_auc: 0.07443191655594603
Gini: 0.7604333711230649

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999906284398464: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.14 %, avg score: 20.11 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.934568
dtype: float64
GBM acc:predict    0.939502
dtype: float64
DL acc:predict    0.937453
dtype: float64
B acc:predict    0.925991
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_9154


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.041220721402791226
RMSE: 0.20302886839755382
LogLoss: 0.14291784461768053
Null degrees of freedom: 152230
Residual degrees of freedom: 151930
Null deviance: 70890.93242839258
Residual deviance: 43513.052807988264
AIC: 44115.052807988264
AUC: 0.9260792600556417
pr_auc: 0.4922429433947783
Gini: 0.8521585201112833

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21510496198036827: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.21 %, avg score:  6.21 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04180558061210619
RMSE: 0.20446413038013828
LogLoss: 0.14515639709765943
Null degrees of freedom: 152230
Residual degrees of freedom: 151930
Null deviance: 70891.83317095178
Residual deviance: 44194.606973147595
AIC: 44796.606973147595
AUC: 0.9231074721566838
pr_auc: 0.48016013963989157
Gini: 0.8462149443133675

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2111765020238214: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.21 %, avg score: 138.58 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_9172


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03761991406004633
RMSE: 0.19395853696098642
LogLoss: 0.13348995118015725
Mean Per-Class Error: 0.1323590209497194
AUC: 0.9380951248774595
pr_auc: 0.573497884861639
Gini: 0.876190249754919

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20976961642821243: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.21 %, avg score:  6.23 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03818140191053329
RMSE: 0.19540061901266662
LogLoss: 0.13693628400794447
Mean Per-Class Error: 0.15125869611991682
AUC: 0.9219817766452489
pr_auc: 0.4943208702193903
Gini: 0.8439635532904979

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21386899453947078: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.15 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.040420548434079405
RMSE: 0.20104862206461252
LogLoss: 0.1437021743806089
Mean Per-Class Error: 0.15168508058914987
AUC: 0.9213262703080504
pr_auc: 0.5089772284462922
Gini: 0.8426525406161007

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2177941359776476: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.21 %, avg score:  6.17 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_9345


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,600,916 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.010869085116648093
RMSE: 0.10425490452083343
LogLoss: 0.04465123808031705
Mean Per-Class Error: 0.03419731673175974
AUC: 0.9903156089803566
pr_auc: 0.9102709152482982
Gini: 0.9806312179607133

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.36074867134428384: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.35 %, avg score:  5.62 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04089310973270672
RMSE: 0.20222044835452896
LogLoss: 0.1980198437398277
Mean Per-Class Error: 0.12864161227856274
AUC: 0.9257842855997739
pr_auc: 0.5040924342509673
Gini: 0.8515685711995478

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.06962007444981483: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  4.99 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.037708008733442634
RMSE: 0.19418550083217498
LogLoss: 0.13432702176006497
Mean Per-Class Error: 0.12775645093618082
AUC: 0.9371975252330197
pr_auc: 0.565205674757595
Gini: 0.8743950504660394

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.312187335933612: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.21 %, avg score:  6.54 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_9484


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.13969491534612827
RMSE: 0.3737578298124713
LogLoss: 1.5981133060911181
Mean Per-Class Error: 0.16584606552920866
AUC: 0.881611509741162
pr_auc: 0.07776891632782673
Gini: 0.763223019482324

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.999983456776522: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.21 %, avg score: 19.15 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.1403799329004304
RMSE: 0.37467310138363336
LogLoss: 1.6138344359955124
Mean Per-Class Error: 0.16047004400268317
AUC: 0.8839873265522231
pr_auc: 0.0730356876696422
Gini: 0.7679746531044462

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999926976646405: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 19.08 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.1397509643951182
RMSE: 0.3738328027275271
LogLoss: 1.60067769076418
Mean Per-Class Error: 0.16667592117567365
AUC: 0.8809683008209894
pr_auc: 0.07916996899588213
Gini: 0.7619366016419789

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999865533634407: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.21 %, avg score: 19.14 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.935099
dtype: float64
GBM acc:predict    0.936921
dtype: float64
DL acc:predict    0.934872
dtype: float64
B acc:predict    0.930925
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_9503


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04057238411767403
RMSE: 0.20142587747773133
LogLoss: 0.14064417700184695
Null degrees of freedom: 158028
Residual degrees of freedom: 157728
Null deviance: 73083.33771802354
Residual deviance: 44451.71729484975
AIC: 45053.71729484975
AUC: 0.9291194613693957
pr_auc: 0.49409604220080655
Gini: 0.8582389227387914

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21307776326836564: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.16 %, avg score:  6.16 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.041136734669516936
RMSE: 0.20282192847302516
LogLoss: 0.14272043786911473
Null degrees of freedom: 158028
Residual degrees of freedom: 157728
Null deviance: 73086.47036512115
Residual deviance: 45107.936152036666
AIC: 45709.936152036666
AUC: 0.9264672509844177
pr_auc: 0.48320501162838775
Gini: 0.8529345019688355

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19522622545208324: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.16 %, avg score: 137.89 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_9521


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03701161629667088
RMSE: 0.19238403337249918
LogLoss: 0.13110665406759006
Mean Per-Class Error: 0.12832310779903
AUC: 0.9404321842557978
pr_auc: 0.5806683331079543
Gini: 0.8808643685115956

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22057375157867923: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.16 %, avg score:  6.18 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03793705514478755
RMSE: 0.19477436983542662
LogLoss: 0.13641454081043705
Mean Per-Class Error: 0.14352299910138333
AUC: 0.9202981888444971
pr_auc: 0.5010044954789655
Gini: 0.8405963776889942

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2543577692284743: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.07 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.03981381669456688
RMSE: 0.19953399884372305
LogLoss: 0.14121754694665012
Mean Per-Class Error: 0.1470077513689929
AUC: 0.924221787211238
pr_auc: 0.5151313728814152
Gini: 0.8484435744224761

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1997234221863952: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.16 %, avg score:  6.12 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_9698


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,601,498 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.0110861590467383
RMSE: 0.1052908307818791
LogLoss: 0.043635022258301175
Mean Per-Class Error: 0.0396094861331755
AUC: 0.9923238612175393
pr_auc: 0.9131261103755106
Gini: 0.9846477224350787

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.3407000085492153: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.53 %, avg score:  5.97 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.041009896977969196
RMSE: 0.20250900468366634
LogLoss: 0.1772836583144507
Mean Per-Class Error: 0.13308649501541991
AUC: 0.9316751817272847
pr_auc: 0.5131233920749265
Gini: 0.8633503634545694

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1052076400431283: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  5.22 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.036963214315000076
RMSE: 0.19225819700340496
LogLoss: 0.13033197189663695
Mean Per-Class Error: 0.12732122457695394
AUC: 0.9389636672872632
pr_auc: 0.5681624885067247
Gini: 0.8779273345745264

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.29508632036965426: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.16 %, avg score:  6.32 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_9835


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.1836498569439817
RMSE: 0.4285438798349379
LogLoss: 2.275676415642746
Mean Per-Class Error: 0.1819722266790753
AUC: 0.8628530241488106
pr_auc: 0.052413493745011086
Gini: 0.7257060482976212

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999901191357575: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.16 %, avg score: 23.65 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.18420303144453073
RMSE: 0.4291888062898784
LogLoss: 2.3081732546679126
Mean Per-Class Error: 0.17925081529420206
AUC: 0.8598704916234586
pr_auc: 0.05571342563047595
Gini: 0.7197409832469173

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999950310468131: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 23.49 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.18379673697697765
RMSE: 0.4287152166380121
LogLoss: 2.285619486949519
Mean Per-Class Error: 0.18269196159309176
AUC: 0.8614150553548366
pr_auc: 0.05206350829139148
Gini: 0.7228301107096733

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999891065656182: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.16 %, avg score: 23.65 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.935251
dtype: float64
GBM acc:predict    0.941627
dtype: float64
DL acc:predict    0.934948
dtype: float64
B acc:predict    0.909291
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_9854


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.04100894127801221
RMSE: 0.2025066450218664
LogLoss: 0.1414550926399943
Null degrees of freedom: 163572
Residual degrees of freedom: 163272
Null deviance: 76222.0967375639
Residual deviance: 46276.46773680358
AIC: 46878.46773680358
AUC: 0.9295350467459621
pr_auc: 0.49700942440072365
Gini: 0.8590700934919242

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.19925165917633206: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.22 %, avg score:  6.22 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04157117113624225
RMSE: 0.20389009572866026
LogLoss: 0.14356768440914375
Null degrees of freedom: 163572
Residual degrees of freedom: 163272
Null deviance: 76223.26489175494
Residual deviance: 46967.59368371374
AIC: 47569.59368371374
AUC: 0.9268145387119558
pr_auc: 0.48570974680443824
Gini: 0.8536290774239117

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.18868133137136622: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.22 %, avg score: 134.41 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_9872


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03765802841173654
RMSE: 0.1940567659519671
LogLoss: 0.13340849249614473
Mean Per-Class Error: 0.13018744129905946
AUC: 0.938399037649745
pr_auc: 0.5703722394468432
Gini: 0.8767980752994899

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20595611206543524: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.22 %, avg score:  6.24 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03824785893257305
RMSE: 0.19557059833362747
LogLoss: 0.13725900032216343
Mean Per-Class Error: 0.14677774215187045
AUC: 0.9199006142656446
pr_auc: 0.4918278643277772
Gini: 0.8398012285312892

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23507847397637055: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.07 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04022775390739289
RMSE: 0.20056857657019178
LogLoss: 0.14279874438103646
Mean Per-Class Error: 0.14843468066704357
AUC: 0.9234743262159526
pr_auc: 0.5134683445897731
Gini: 0.8469486524319052

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22763946562500867: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.22 %, avg score:  6.19 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_10045


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,799,590 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.009935902557712614
RMSE: 0.0996789975757813
LogLoss: 0.04055241318376154
Mean Per-Class Error: 0.03289106212194315
AUC: 0.9924598736535345
pr_auc: 0.9021216592115254
Gini: 0.984919747307069

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.4287293361223924: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.60 %, avg score:  6.76 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04636455221309189
RMSE: 0.21532429545476722
LogLoss: 0.20113839546866968
Mean Per-Class Error: 0.1579577143917884
AUC: 0.9080217355535773
pr_auc: 0.4539523820080331
Gini: 0.8160434711071547

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.23013917737296588: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.18 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.03736831878264811
RMSE: 0.19330886886702356
LogLoss: 0.13153555517323046
Mean Per-Class Error: 0.12439726493299552
AUC: 0.941011976905926
pr_auc: 0.5731336482484054
Gini: 0.882023953811852

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2978357532182371: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.22 %, avg score:  6.54 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_10190


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.1769165539421286
RMSE: 0.42061449563956854
LogLoss: 2.1528002925070355
Mean Per-Class Error: 0.17966932832589488
AUC: 0.8634724749509818
pr_auc: 0.05575331926970876
Gini: 0.7269449499019636

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999903045627876: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.22 %, avg score: 23.00 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.17766960944211457
RMSE: 0.4215087299714142
LogLoss: 2.1613528281773333
Mean Per-Class Error: 0.1728075652551776
AUC: 0.8681906058701183
pr_auc: 0.057165054918977054
Gini: 0.7363812117402366

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999967219680699: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 22.93 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.17724327729014858
RMSE: 0.4210027046114414
LogLoss: 2.1619094693198226
Mean Per-Class Error: 0.18051271390415313
AUC: 0.8623844469011942
pr_auc: 0.05498582255237527
Gini: 0.7247688938023884

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999902401967015: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.22 %, avg score: 23.02 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.934568
dtype: float64
GBM acc:predict    0.939502
dtype: float64
DL acc:predict    0.934037
dtype: float64
B acc:predict    0.916047
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_10209


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.040778234496008904
RMSE: 0.2019362139290744
LogLoss: 0.14160832074821872
Null degrees of freedom: 169709
Residual degrees of freedom: 169409
Null deviance: 78811.6256933554
Residual deviance: 48064.6962283604
AIC: 48666.6962283604
AUC: 0.9277439592094513
pr_auc: 0.49878206067552994
Gini: 0.8554879184189026

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20112368930783536: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score:  6.19 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.04131076957965623
RMSE: 0.20325050942041015
LogLoss: 0.14365693582846537
Null degrees of freedom: 169709
Residual degrees of freedom: 169409
Null deviance: 78812.62668319754
Residual deviance: 48760.03715889771
AIC: 49362.03715889771
AUC: 0.9250604276320336
pr_auc: 0.48825058150034845
Gini: 0.8501208552640671

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.18478786872501665: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score: 131.85 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_10227


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03744343813908547
RMSE: 0.19350307010248047
LogLoss: 0.13323966017828404
Mean Per-Class Error: 0.13201295141582947
AUC: 0.9368958927504687
pr_auc: 0.5760529170048676
Gini: 0.8737917855009374

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21829651585729135: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score:  6.21 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03827087668607866
RMSE: 0.19562943716649256
LogLoss: 0.13695863967757582
Mean Per-Class Error: 0.14355348034645254
AUC: 0.9228172686273104
pr_auc: 0.4963314389709844
Gini: 0.8456345372546208

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2279733563178289: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.09 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.040040109704700864
RMSE: 0.20010024913702848
LogLoss: 0.14235124272697894
Mean Per-Class Error: 0.14840466527731566
AUC: 0.922855582919099
pr_auc: 0.515911557078792
Gini: 0.8457111658381979

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.22304470183850542: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score:  6.16 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_10402


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,800,030 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.013888149842058972
RMSE: 0.11784799464589532
LogLoss: 0.05195054249856494
Mean Per-Class Error: 0.03672595721016603
AUC: 0.9925648148115925
pr_auc: 0.8689320831959931
Gini: 0.9851296296231851

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.5213301893998424: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.10 %, avg score:  6.94 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04406221200450515
RMSE: 0.20991000930042653
LogLoss: 0.1738058200727924
Mean Per-Class Error: 0.14273365087267065
AUC: 0.922374446802317
pr_auc: 0.45391910521140005
Gini: 0.8447488936046339

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.3514762997882288: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.95 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.03693773159286793
RMSE: 0.19219191344296443
LogLoss: 0.13241701483657292
Mean Per-Class Error: 0.1275093269230576
AUC: 0.9375251350560609
pr_auc: 0.5656233316481402
Gini: 0.8750502701121219

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.34466567650817753: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score:  6.28 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_10545


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.1462411363354524
RMSE: 0.3824148746263048
LogLoss: 1.666854399038216
Mean Per-Class Error: 0.16746168559331376
AUC: 0.8830714072923452
pr_auc: 0.07233335594413286
Gini: 0.7661428145846905

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999844271995721: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score: 19.89 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.14730117224111028
RMSE: 0.38379834840852334
LogLoss: 1.6867486871339619
Mean Per-Class Error: 0.16129651817694723
AUC: 0.8815334281169973
pr_auc: 0.06183636101491159
Gini: 0.7630668562339946

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999817939445022: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 19.92 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.14709064516202425
RMSE: 0.38352398251220776
LogLoss: 1.6868156626826583
Mean Per-Class Error: 0.16952720430516588
AUC: 0.8804261502067054
pr_auc: 0.0742180484448396
Gini: 0.7608523004134109

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999878684871598: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.19 %, avg score: 19.92 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.933354
dtype: float64
GBM acc:predict    0.938288
dtype: float64
DL acc:predict    0.938743
dtype: float64
B acc:predict    0.924017
dtype: float64
glm Model Build progress: [###############################################] 100%
gbm Model Build progress: [###############################################] 100%
deeplearning Model Build progress: [######################################] 100%
naivebayes Model Build progress: [########################################] 100%
glm prediction progress: [################################################] 100%
gbm prediction progress: [################################################] 100%
deeplearning prediction progress: [#######################################] 100%
naivebayes prediction progress: [#########################################] 100%
Model Details
=============
H2OGeneralizedLinearEstimator :  Generalized Linear Modeling
Model Key:  GLM_model_python_1573972048565_10564


GLM Model: summary


ModelMetricsBinomialGLM: glm
** Reported on train data. **

MSE: 0.041182186624050696
RMSE: 0.2029339464556157
LogLoss: 0.1430858445777631
Null degrees of freedom: 175866
Residual degrees of freedom: 175566
Null deviance: 82223.33533767059
Residual deviance: 50328.15645671493
AIC: 50930.15645671493
AUC: 0.9275698597122802
pr_auc: 0.4951661583774864
Gini: 0.8551397194245605

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.21073704903929039: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score:  6.25 %


ModelMetricsBinomialGLM: glm
** Reported on cross-validation data. **

MSE: 0.041652725408528754
RMSE: 0.20408999340616568
LogLoss: 0.1449179289795636
Null degrees of freedom: 175866
Residual degrees of freedom: 175566
Null deviance: 82224.84075971034
Residual deviance: 50972.562831697825
AIC: 51574.562831697825
AUC: 0.9251426120322425
pr_auc: 0.4863320850304198
Gini: 0.8502852240644849

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.20650025567248992: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score: 129.31 %


Cross-Validation Metrics Summary: 

See the whole table with table.as_data_frame()

Scoring History: 

Model Details
=============
H2OGradientBoostingEstimator :  Gradient Boosting Machine
Model Key:  GBM_model_python_1573972048565_10582


Model Summary: 


ModelMetricsBinomial: gbm
** Reported on train data. **

MSE: 0.03790751189702913
RMSE: 0.19469851539503102
LogLoss: 0.1344180203370941
Mean Per-Class Error: 0.1313328283438775
AUC: 0.9372204185320865
pr_auc: 0.5718097652896756
Gini: 0.874440837064173

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2370041938673815: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score:  6.27 %


ModelMetricsBinomial: gbm
** Reported on validation data. **

MSE: 0.03819488969025713
RMSE: 0.19543512911003777
LogLoss: 0.13721095606524247
Mean Per-Class Error: 0.1496677649758893
AUC: 0.9209834895013732
pr_auc: 0.4891584893367558
Gini: 0.8419669790027464

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.2295447061087724: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  6.10 %


ModelMetricsBinomial: gbm
** Reported on cross-validation data. **

MSE: 0.04035185469322861
RMSE: 0.20087771079248343
LogLoss: 0.14320278799892172
Mean Per-Class Error: 0.14775500994832647
AUC: 0.9234507130285663
pr_auc: 0.5150228376261571
Gini: 0.8469014260571326

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.1964101195513644: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score:  6.23 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ODeepLearningEstimator :  Deep Learning
Model Key:  DeepLearning_model_python_1573972048565_10753


Status of Neuron Layers: predicting target, 2-class classification, bernoulli distribution, CrossEntropy loss, 100,802 weights/biases, 1.2 MB, 1,802,790 training samples, mini-batch size 1


ModelMetricsBinomial: deeplearning
** Reported on train data. **

MSE: 0.01765371684687607
RMSE: 0.13286729035724357
LogLoss: 0.06521982266074391
Mean Per-Class Error: 0.049763474269079566
AUC: 0.9889023596590671
pr_auc: 0.8596657068631212
Gini: 0.9778047193181343

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.6413834164632737: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.15 %, avg score:  8.30 %


ModelMetricsBinomial: deeplearning
** Reported on validation data. **

MSE: 0.04837110267096053
RMSE: 0.2199343144462922
LogLoss: 0.18613533870544302
Mean Per-Class Error: 0.14865070391636537
AUC: 0.9188064008505217
pr_auc: 0.4546231780180693
Gini: 0.8376128017010434

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.42407552046697283: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score:  8.33 %


ModelMetricsBinomial: deeplearning
** Reported on cross-validation data. **

MSE: 0.03692879456856405
RMSE: 0.19216866177544156
LogLoss: 0.1297339820472187
Mean Per-Class Error: 0.12095072383155703
AUC: 0.9407846090997563
pr_auc: 0.5758175546863142
Gini: 0.8815692181995125

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.27095015818068074: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score:  6.54 %


Cross-Validation Metrics Summary: 

Scoring History: 

Variable Importances: 

See the whole table with table.as_data_frame()

Model Details
=============
H2ONaiveBayesEstimator :  Naive Bayes
Model Key:  NaiveBayes_model_python_1573972048565_10906


Model Summary: 


ModelMetricsBinomial: naivebayes
** Reported on train data. **

MSE: 0.1693988088716287
RMSE: 0.41158086553146356
LogLoss: 1.9462991381926003
Mean Per-Class Error: 0.17150172657729357
AUC: 0.8765838245327645
pr_auc: 0.061223942592759636
Gini: 0.753167649065529

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999878943739797: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score: 22.49 %


ModelMetricsBinomial: naivebayes
** Reported on validation data. **

MSE: 0.17001036263560546
RMSE: 0.4123231289117862
LogLoss: 1.952159323407531
Mean Per-Class Error: 0.1663173804270327
AUC: 0.8764605157131345
pr_auc: 0.05548470009999734
Gini: 0.7529210314262691

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.999986917986372: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  5.80 %, avg score: 22.43 %


ModelMetricsBinomial: naivebayes
** Reported on cross-validation data. **

MSE: 0.16929516122665692
RMSE: 0.41145493219386364
LogLoss: 1.9518005898280624
Mean Per-Class Error: 0.17207715373982457
AUC: 0.8758193312363224
pr_auc: 0.06185868666498568
Gini: 0.7516386624726448

Confusion Matrix (Act/Pred) for max f1 @ threshold = 0.9999891990688179: 

Maximum Metrics: Maximum metrics at their respective thresholds

Gains/Lift Table: Avg response rate:  6.25 %, avg score: 22.46 %


Cross-Validation Metrics Summary: 

GLM acc:predict    0.935175
dtype: float64
GBM acc:predict    0.941324
dtype: float64
DL acc:predict    0.933961
dtype: float64
B acc:predict    0.916806
dtype: float64