Summary of 11_Xgboost

<< Go back

Extreme Gradient Boosting (Xgboost)

• n_jobs: 6
• objective: multi:softprob
• eta: 0.075
• 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

43.8 seconds

Metric details

	0	1	2	3	4	5	accuracy	macro avg	weighted avg	logloss
precision	0.970847	0.965429	0.973873	0.985663	0.991266	0.998435	0.980802	0.980919	0.980896	0.0583742
recall	0.987409	0.965429	0.969898	0.990226	0.992892	0.975038	0.980802	0.980149	0.980802	0.0583742
f1-score	0.979058	0.965429	0.971881	0.987939	0.992079	0.986598	0.980802	0.980497	0.980809	0.0583742
support	2462	1562	1960	1944	1829	1963	0.980802	11720	11720	0.0583742

Confusion matrix

	Predicted as 0	Predicted as 1	Predicted as 2	Predicted as 3	Predicted as 4	Predicted as 5
Labeled as 0	2431	6	5	18	0	2
Labeled as 1	2	1508	45	1	6	0
Labeled as 2	8	46	1901	1	3	1
Labeled as 3	13	2	1	1925	3	0
Labeled as 4	6	0	0	7	1816	0
Labeled as 5	44	0	0	1	4	1914

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Precision Recall Curve

<< Go back