Skip to content

Summary of 1_Default_Xgboost

<< Go back

Extreme Gradient Boosting (Xgboost)

  • n_jobs: -1
  • objective: multi:softprob
  • eta: 0.075
  • max_depth: 6
  • 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

17.6 seconds

Metric details

0 1 2 3 4 5 accuracy macro avg weighted avg logloss
precision 0.757305 0.76361 0.773113 0.955914 0.978316 0.987243 0.86384 0.86925 0.86925 0.330851
recall 0.917386 0.780635 0.699803 0.935018 0.962963 0.887234 0.86384 0.86384 0.86384 0.330851
f1-score 0.829695 0.772029 0.734634 0.945351 0.970579 0.93457 0.86384 0.864476 0.864476 0.330851
support 6296 6296 6296 6296 6296 6296 0.86384 37776 37776 0.330851

Confusion matrix

Predicted as 0 Predicted as 1 Predicted as 2 Predicted as 3 Predicted as 4 Predicted as 5
Labeled as 0 5775.86 77.6326 205.726 114.508 56.2836 65.9877
Labeled as 1 314.324 4914.88 1028.7 9.52496 28.5749 0
Labeled as 2 408.992 1443.87 4405.96 30.9843 0 6.19685
Labeled as 3 363.668 0 22.7292 5886.87 22.7292 0
Labeled as 4 107.624 0 35.8746 89.6866 6062.81 0
Labeled as 5 656.391 0 0 26.7915 26.7915 5586.03

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