README.md

Summary of 3_Default_CatBoost

<< Go back

CatBoost

• n_jobs: -1
• learning_rate: 0.1
• depth: 6
• rsm: 1
• loss_function: Logloss
• eval_metric: Logloss
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

logloss

Training time

1.5 seconds

Metric details

	score	threshold
logloss	0.602261	nan
auc	0.732324	nan
f1	0.703361	0.32441
accuracy	0.674102	0.554565
precision	0.820755	0.801618
recall	1	0.0122554
mcc	0.350133	0.554565

Confusion matrix (at threshold=0.554565)

	Predicted as 0	Predicted as 1
Labeled as 0	617	204
Labeled as 1	322	471

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back