Skip to content

Summary of 20_Xgboost_SelectedFeatures

<< Go back

Extreme Gradient Boosting (Xgboost)

  • n_jobs: 6
  • objective: multi:softprob
  • eta: 0.1
  • max_depth: 5
  • min_child_weight: 1
  • subsample: 1.0
  • colsample_bytree: 1.0
  • eval_metric: f1
  • num_class: 6
  • explain_level: 1

Validation

  • validation_type: kfold
  • k_folds: 5
  • shuffle: True
  • stratify: True
  • random_seed: 42

Optimized metric

f1

Training time

36.5 seconds

Metric details

0 1 2 3 4 5 accuracy macro avg weighted avg logloss
precision 0.971566 0.967804 0.969913 0.984151 0.99018 0.997917 0.980119 0.980255 0.980194 0.0591794
recall 0.985378 0.962228 0.970408 0.990226 0.992346 0.976057 0.980119 0.97944 0.980119 0.0591794
f1-score 0.978423 0.965008 0.970161 0.987179 0.991262 0.986866 0.980119 0.979816 0.980123 0.0591794
support 2462 1562 1960 1944 1829 1963 0.980119 11720 11720 0.0591794

Confusion matrix

Predicted as 0 Predicted as 1 Predicted as 2 Predicted as 3 Predicted as 4 Predicted as 5
Labeled as 0 2426 7 6 20 0 3
Labeled as 1 2 1503 50 1 6 0
Labeled as 2 9 42 1902 3 4 0
Labeled as 3 13 1 1 1925 4 0
Labeled as 4 4 0 2 7 1815 1
Labeled as 5 43 0 0 0 4 1916

Learning curves

Learning curves

Permutation-based Importance

Permutation-based Importance

Confusion Matrix

Confusion Matrix

Normalized Confusion Matrix

Normalized Confusion Matrix

ROC Curve

ROC Curve

Precision Recall Curve

Precision Recall Curve

<< Go back