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
5.16k stars 576 forks source link

Can we still customize histogram plots in the latest version? #1018

Open naveen-pit opened 6 months ago

naveen-pit commented 6 months ago

According to the v0.1.57 user guide, we can define DataDriftOptions to customize histogram plot and define QualityMetricsOptions to select chunk of data we interest.

However, DataDriftOptions and QualityMetricsOptions are no longer compatible in the latest release since we have to pass options as a subclass of Options. DataDriftOptions is inherited from pydantic.BaseModel, while QualityMetricsOptions is inherited from object.

By looking at latest user guide, we can only see customization on color schema and raw data. DataDriftOptions and QualityMetricsOptions are still there in the API Reference, so I'm pretty not sure whether we still can customize our report via DataDriftOptions and QualityMetricsOptions or there is a workaround for this.

elenasamuylova commented 6 months ago

Hi @naveen-pit,

There are currently no parameters to specify e.g. number of bins in the histograms.

SInce the API update, we divided the customizations into two groups:

However, there are currently now parameters to visually customize the bins.

We'd like to add them, but it is not on the immediate roadmap.

naveen-pit commented 6 months ago

Thank you for your reply. will keep checking the updates.