Summary of 17_Xgboost

<< Go back

Extreme Gradient Boosting (Xgboost)

• n_jobs: 6
• objective: multi:softprob
• eta: 0.05
• 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

53.0 seconds

Metric details

	0	1	2	3	4	5	accuracy	macro avg	weighted avg	logloss
precision	0.969261	0.96408	0.971341	0.984151	0.990153	0.997393	0.979266	0.979397	0.979361	0.0701505
recall	0.98619	0.962228	0.968367	0.990226	0.989612	0.974529	0.979266	0.978525	0.979266	0.0701505
f1-score	0.977653	0.963153	0.969852	0.987179	0.989882	0.985828	0.979266	0.978925	0.979274	0.0701505
support	2462	1562	1960	1944	1829	1963	0.979266	11720	11720	0.0701505

Confusion matrix

	Predicted as 0	Predicted as 1	Predicted as 2	Predicted as 3	Predicted as 4	Predicted as 5
Labeled as 0	2428	6	6	20	0	2
Labeled as 1	2	1503	50	1	6	0
Labeled as 2	8	48	1898	1	4	1
Labeled as 3	13	2	0	1925	4	0
Labeled as 4	8	0	0	9	1810	2
Labeled as 5	46	0	0	0	4	1913

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Precision Recall Curve

<< Go back