ff3 is a library for creating statistical reports and visualizations based on sample datasets. It can help you to detect domain drifts on your data and create low-dimensional projections.
pip install ff3In this section you'll find examples on how to use ff3 library to create drift reports.
ff3.DriftReport creates a statistical report based on a comparison between given datasets distributions. It compares them by a set of statistical tests and finds deviations.
import pandas as pd
from ff3 import DataDrift
df1 = pd.read_csv("yesterday-data.csv")
df2 = pd.read_csv("today-data.csv")
drift = DriftReport(d1, d2)
drift.eval()
drift.detailed_report() # create an interactive report
In case you want to find out relationship between two distinct features, you can create a bivariate report.
drift.bivariate_report("passenger_count", "tip_amount")
You can de/serialize DriftReport instance if required.
serialized = drift.to_dict()
drift2 = DriftReport.from_dict(serialized)
drift2.detailed_report()