Skip to content

Summary of 2_Xgboost_GoldenFeatures

<< Go back

Extreme Gradient Boosting (Xgboost)

  • n_jobs: -1
  • objective: multi:softprob
  • eta: 0.075
  • max_depth: 8
  • min_child_weight: 5
  • 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

18.0 seconds

Metric details

0 1 2 3 4 5 accuracy macro avg weighted avg logloss
precision 0.740733 0.771806 0.776653 0.964051 0.97591 0.991523 0.863051 0.870113 0.870113 0.530179
recall 0.923551 0.76702 0.714567 0.935018 0.965812 0.87234 0.863051 0.863051 0.863051 0.530179
f1-score 0.822101 0.769406 0.744318 0.949313 0.970835 0.928121 0.863051 0.864015 0.864015 0.530179
support 6296 6296 6296 6296 6296 6296 0.863051 37776 37776 0.530179

Confusion matrix

Predicted as 0 Predicted as 1 Predicted as 2 Predicted as 3 Predicted as 4 Predicted as 5
Labeled as 0 5814.68 89.2774 186.318 108.686 56.2836 40.7571
Labeled as 1 333.374 4829.16 1066.8 28.5749 38.0998 0
Labeled as 2 439.976 1338.52 4498.91 6.19685 6.19685 6.19685
Labeled as 3 363.668 0 22.7292 5886.87 22.7292 0
Labeled as 4 161.436 0 17.9373 35.8746 6080.75 0
Labeled as 5 736.766 0 0 40.1872 26.7915 5492.26

Learning curves

Learning curves

Confusion Matrix

Confusion Matrix

Normalized Confusion Matrix

Normalized Confusion Matrix

ROC Curve

ROC Curve

Precision Recall Curve

Precision Recall Curve

<< Go back