How to choose the optimal threshold for imbalanced classification

threshold_precision_recall.py

# Create the Precision-Recall curve
precision, recall, thresholds = precision_recall_curve(y_test, y_pred)

# Plot the ROC curve
df_recall_precision = pd.DataFrame({'Precision':precision[:-1],
                                    'Recall':recall[:-1],
                                    'Threshold':thresholds})
df_recall_precision.head()

# Creat a data viz
plotnine.options.figure_size = (8, 4.8)
(
    ggplot(data = df_recall_precision)+
    geom_point(aes(x = 'Recall',
                   y = 'Precision'),
               size = 0.4)+
    geom_line(aes(x = 'Recall',
                  y = 'Precision'))+
    labs(title = 'Recall Precision Curve')+
    xlab('Recall')+
    ylab('Precision')+
    theme_minimal()
)