Visualizing quantiles of acceptance

You know how quantile() works to compute a threshold, and you've seen an example of what it does to split the loans into accepted and rejected. What does this threshold look like for the test set, and how can you visualize it?

To check this, you can create a histogram of the probabilities and add a reference line for the threshold. With this, you can visually show where the threshold exists in the distribution.

The model predictions clf_gbt_preds have been loaded into the workspace.

This exercise is part of the course

Credit Risk Modeling in Python

View Course

Exercise instructions

Create a histogram of the predicted probabilities clf_gbt_preds.
Calculate the threshold for an acceptance rate of 85% using quantile(). Store this value as threshold.
Plot the histogram again, except this time add a reference line using .axvline().

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

# Plot the predicted probabilities of default
plt.____(____, color = 'blue', bins = 40)

# Calculate the threshold with quantile
____ = np.____(____, ____)

# Add a reference line to the plot for the threshold
plt.____(x = ____, color = 'red')
plt.____()

Edit and Run Code