Summary of 16_LightGBM_SelectedFeatures

<< Go back

LightGBM

• n_jobs: 6
• objective: multiclass
• num_leaves: 95
• learning_rate: 0.1
• feature_fraction: 0.5
• bagging_fraction: 1.0
• min_data_in_leaf: 10
• metric: custom
• custom_eval_metric_name: 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

41.7 seconds

Metric details

	0	1	2	3	4	5	accuracy	macro avg	weighted avg	logloss
precision	0.970167	0.962916	0.972694	0.989236	0.997801	0.997411	0.98157	0.981704	0.981662	0.0531892
recall	0.990658	0.964149	0.963265	0.992798	0.992346	0.981151	0.98157	0.980728	0.98157	0.0531892
f1-score	0.980305	0.963532	0.967957	0.991014	0.995066	0.989214	0.98157	0.981181	0.981577	0.0531892
support	2462	1562	1960	1944	1829	1963	0.98157	11720	11720	0.0531892

Confusion matrix

	Predicted as 0	Predicted as 1	Predicted as 2	Predicted as 3	Predicted as 4	Predicted as 5
Labeled as 0	2439	10	1	9	0	3
Labeled as 1	4	1506	50	0	2	0
Labeled as 2	20	47	1888	3	1	1
Labeled as 3	13	1	0	1930	0	0
Labeled as 4	5	0	2	6	1815	1
Labeled as 5	33	0	0	3	1	1926

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Precision Recall Curve

<< Go back