empulse.metrics#
The metrics
module contains a collection of metrics for evaluating the performance of
models in the context of customer churn, credit scoring, and acquisition.
General Metrics#
Class to create a custom value/cost-sensitive metric. |
|
Maximum Profit Measure (MP). |
|
|
|
Compute the lift score for the top fraction of predictions. |
|
Cost of a classifier. |
|
Expected cost of a classifier. |
|
Expected log cost of a classifier. |
|
Create an objective function for the Average Expected Cost (AEC) measure. |
|
Cost savings of a classifier compared to using a baseline. |
|
Expected savings of a classifier compared to a baseline. |
Customer Acquisition Metrics#
Expected Maximum Profit measure for customer Acquisition (EMPA). |
|
|
|
Maximum Profit measure for customer Acquisition (MPA). |
|
|
|
Expected cost of a classifier for customer acquisition. |
|
Create an objective function for the Expected Cost measure for customer acquisition. |
Customer Churn Metrics#
Expected Maximum Profit Measure for Customer Churn (EMPC). |
|
|
|
Maximum Profit Measure for Customer Churn (MPC). |
|
|
|
Expected Maximum Profit Measure for B2B Customer Churn (EMPB). |
|
|
|
Area Under the Expected Profit Curve (AUEPC). |
|
Expected cost of a classifier for customer churn. |
|
Create an objective function for the Expected Cost measure for customer churn. |
Credit Scoring Metrics#
Helper Functions#
Return classification threshold for given customer threshold. |