Appendix Z — chapter5.analysis.reports

chapter5.analysis.reports

Provides functions to process the reports generated in the evaluation of classification models.

Functions

Name Description
extract_metrics Extracts metrics of interest from a classification report
metric_increment_summary Creates a summary with metrics increments/decrements in the provided reports to compare
metrics_summary Creates a summary of the classification reports generated by the evaluation of several models.

extract_metrics

chapter5.analysis.reports.extract_metrics(report)

Extracts metrics of interest from a classification report

Parameters

Name Type Description Default
report dict | list[dict] Classification report or a list of reports. required

Returns

Type Description
float Accuracy obtained by the model.
float Precision obtained by the model.
float Recall obtained by the model.
float F1-score obtained by the model.

metric_increment_summary

chapter5.analysis.reports.metric_increment_summary(comparisons)

Creates a summary with metrics increments/decrements in the provided reports to compare different approaches

Parameters

Name Type Description Default
comparisons dict Dict where each entry indicates which reports to compare. required

Returns

Type Description
pandas.DataFrame DataFrame containing the comparisons.

metrics_summary

chapter5.analysis.reports.metrics_summary(reports, labels)

Creates a summary of the classification reports generated by the evaluation of several models.

Parameters

Name Type Description Default
reports list[dict] List of classification reports. required

Returns

Type Description
pandas.DataFrame DataFrame summarizing the results.