evidentlyai / evidently

Evidently is ​​an open-source ML and LLM observability framework. Evaluate, test, and monitor any AI-powered system or data pipeline. From tabular data to Gen AI. 100+ metrics.
https://www.evidentlyai.com/evidently-oss
Apache License 2.0
5.2k stars 586 forks source link

Question: Targetdrift doesn't detect drift when trend is reversed. #1080

Open PranavKhedkar opened 5 months ago

PranavKhedkar commented 5 months ago

Hello, I am using two pandas dataframe: data1 = { 'Sqft': [600, 700, 800, 900, 1000], 'Price': [235000, 255000, 293350, 321200, 350000] } df1 = pd.DataFrame(data1)

data2 = { 'Sqft': [600, 700, 800, 900, 1000], 'Price': [350000, 321200, 293350, 255000, 235000] } df2 = pd.DataFrame(data2)

The target here is price. The second dataframe has exact opposite trend as compared to the first. Yet, when I try to use TargetDriftPreset, drift score remains 1 and drift is not detected. Am I doing something wrong?