Summary of 3_Linear

<< Go back

Logistic Regression (Linear)

Validation

Optimized metric

logloss

Training time

7.8 seconds

Metric details

0 1 2 3 accuracy macro avg weighted avg logloss
precision 0 0 0.618321 0.630389 0.623658 0.312177 0.589272 0.801042
recall 0 0 0.694631 0.622571 0.623658 0.329301 0.623658 0.801042
f1-score 0 0 0.654258 0.626456 0.623658 0.320179 0.605353 0.801042
support 282 19 2682 2419 0.623658 5402 5402 0.801042

Confusion matrix

Predicted as 0 Predicted as 1 Predicted as 2 Predicted as 3
Labeled as 0 0 0 222 60
Labeled as 1 0 0 15 4
Labeled as 2 0 0 1863 819
Labeled as 3 0 0 913 1506

Learning curves

Learning curves

Coefficients

Coefficients learner #1

0 1 2 3
intercept -0.446662 -3.34533 2.02469 1.7673
filtre -0.0888187 0.154754 -0.0178255 -0.0481096
latitude -0.0141252 -0.0735754 0.0444655 0.043235
longitude 0.0300214 -0.0624727 0.0400118 -0.00756055
has_agrement -0.134032 -0.441637 0.126474 0.449195
dept -0.0291757 0.0783951 -0.0435639 -0.00565552
year 0.254794 -0.0158806 -0.186856 -0.0520575
month 0.282456 -0.071908 -0.101168 -0.10938
weekday -0.0993311 0.106422 -0.00852444 0.00143376
count_controls_dept -0.175346 -0.359413 0.243545 0.291214
score_controls_dept -0.232151 -0.375072 0.0273233 0.5799
count_controls_filtre 0.0762235 0.049562 -0.0335539 -0.0922316
score_controls_filtre -0.261644 -0.427335 0.161154 0.527825
count_controls_activite -0.00271059 -0.75941 0.22235 0.539771
score_controls_activite -0.136814 -0.30772 0.0218038 0.42273
count_controls_wday -0.0903018 0.269135 -0.102005 -0.0768278
score_controls_wday -0.0495039 0.0616961 -0.0118467 -0.000345442

Permutation-based Importance

Permutation-based Importance

SHAP Importance

SHAP Importance

SHAP Dependence plots

Dependence 0 (Fold 1)

SHAP Dependence from fold 1

Dependence 1 (Fold 1)

SHAP Dependence from fold 1

Dependence 2 (Fold 1)

SHAP Dependence from fold 1

Dependence 3 (Fold 1)

SHAP Dependence from fold 1

SHAP Decision plots

Worst decisions for selected sample 1 (Fold 1)

SHAP worst decisions from Fold 1

Worst decisions for selected sample 2 (Fold 1)

SHAP worst decisions from Fold 1

Worst decisions for selected sample 3 (Fold 1)

SHAP worst decisions from Fold 1

Worst decisions for selected sample 4 (Fold 1)

SHAP worst decisions from Fold 1

Best decisions for selected sample 1 (Fold 1)

SHAP best decisions from Fold 1

Best decisions for selected sample 2 (Fold 1)

SHAP best decisions from Fold 1

Best decisions for selected sample 3 (Fold 1)

SHAP best decisions from Fold 1

Best decisions for selected sample 4 (Fold 1)

SHAP best decisions from Fold 1

<< Go back