Summary of 1_Default_LightGBM

<< Go back

LightGBM

• n_jobs: 6
• objective: multiclass
• num_leaves: 63
• learning_rate: 0.05
• feature_fraction: 0.9
• bagging_fraction: 0.9
• 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

69.4 seconds

Metric details

	0	1	2	3	4	5	accuracy	macro avg	weighted avg	logloss
precision	0.966521	0.963344	0.96738	0.983632	0.995619	0.997909	0.978754	0.979067	0.978878	0.0699315
recall	0.984972	0.959027	0.968367	0.989198	0.993986	0.972491	0.978754	0.978007	0.978754	0.0699315
f1-score	0.975659	0.961181	0.967874	0.986407	0.994802	0.985036	0.978754	0.978493	0.978768	0.0699315
support	2462	1562	1960	1944	1829	1963	0.978754	11720	11720	0.0699315

Confusion matrix

	Predicted as 0	Predicted as 1	Predicted as 2	Predicted as 3	Predicted as 4	Predicted as 5
Labeled as 0	2425	7	3	26	0	1
Labeled as 1	3	1498	56	0	4	1
Labeled as 2	9	49	1898	1	2	1
Labeled as 3	16	1	2	1923	2	0
Labeled as 4	3	0	3	4	1818	1
Labeled as 5	53	0	0	1	0	1909

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Precision Recall Curve

<< Go back