Summary of 3_Linear

<< Go back

Logistic Regression (Linear)

Validation

Optimized metric

logloss

Training time

8.2 seconds

Metric details

0 1 2 3 4 accuracy macro avg weighted avg logloss
precision 0.520548 0.333333 0.666667 0.380952 0.666667 0.491429 0.513633 0.476675 1.08111
recall 0.808511 0.28 0.689655 0.235294 0.4 0.491429 0.482692 0.491429 1.08111
f1-score 0.633333 0.304348 0.677966 0.290909 0.5 0.491429 0.481311 0.468777 1.08111
support 47 50 29 34 15 0.491429 175 175 1.08111

Confusion matrix

Predicted as 0 Predicted as 1 Predicted as 2 Predicted as 3 Predicted as 4
Labeled as 0 38 7 0 2 0
Labeled as 1 19 14 3 11 3
Labeled as 2 5 4 20 0 0
Labeled as 3 6 13 7 8 0
Labeled as 4 5 4 0 0 6

Learning curves

Learning curves

Coefficients

Coefficients learner #1

0 1 2 3 4
intercept 1.80784 2.01673 -2.72552 0.50835 -1.6074
nom_com 0.380576 0.367239 -1.3283 -0.316997 0.897483
insee_com 0.380576 0.367239 -1.3283 -0.316997 0.897483
nom_station 0.0596358 0.0612581 -0.394172 -0.690075 0.963353
code_station -0.432223 -0.38998 1.4818 -0.231621 -0.427975
typologie 0.29101 0.287097 -0.82773 0.215953 0.0336699
influence 0.117335 0.103726 -0.347627 0.27383 -0.147263
valeur -0.853461 -0.0459299 1.04733 -0.125223 -0.0227164
lat 0.225976 0.192929 -0.939005 -0.17667 0.69677
long -0.227332 -0.211188 0.572259 -0.642842 0.509103
jour_semaine_debut -0.0166062 -0.0175814 0.012963 0.0134262 0.00779841
jour_semaine_fin -0.0166062 -0.0175814 0.012963 0.0134262 0.00779841
jour_debut -0.0166062 -0.0175814 0.012963 0.0134262 0.00779841
jour_fin -0.0166062 -0.0175814 0.012963 0.0134262 0.00779841
heure_debut -0.00304425 -0.0303952 0.034434 -0.00822246 0.00722798
heure_fin -0.00304425 -0.0303952 0.034434 -0.00822246 0.00722798

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

Dependence 4 (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