Implementation:Truera Trulens Session Get Records And Feedback

Knowledge Sources	TruLens TruLens Docs
Domains	LLM_Evaluation, Data_Analysis
Last Updated	2026-02-14 08:00 GMT

Overview

Concrete tool for querying detailed records with associated feedback scores for analysis and debugging, provided by the trulens-core library.

Description

The TruSession.get_records_and_feedback method queries the database for individual records and their associated feedback scores. It returns a tuple of (DataFrame, List[str]) where the DataFrame contains records with feedback columns and the list contains feedback column names.

Usage

Call on a TruSession instance when you need record-level detail. Use filtering parameters to narrow results to specific apps, versions, or records.

Code Reference

Source Location

• Repository: trulens
• File: src/core/trulens/core/session.py
• Lines: L797-844

Signature

class TruSession:
    def get_records_and_feedback(
        self,
        app_ids: Optional[List[AppID]] = None,
        app_name: Optional[AppName] = None,
        app_version: Optional[AppVersion] = None,
        app_versions: Optional[List[AppVersion]] = None,
        run_name: Optional[RunName] = None,
        record_ids: Optional[List[RecordID]] = None,
        offset: Optional[int] = None,
        limit: Optional[int] = None,
    ) -> Tuple[pd.DataFrame, List[str]]:
        """Get records with feedback results.

        Args:
            app_ids: Filter by app IDs.
            app_name: Filter by app name.
            app_version: Filter by app version.
            app_versions: Filter by multiple app versions.
            run_name: Filter by run name.
            record_ids: Filter by record IDs.
            offset: Row offset.
            limit: Max records.

        Returns:
            Tuple of (records DataFrame, list of feedback column names).
        """

Import

from trulens.core.session import TruSession

session = TruSession()
records_df, feedback_cols = session.get_records_and_feedback()

I/O Contract

Inputs

Name	Type	Required	Description
app_ids	List[AppID]	No	Filter by app IDs
app_name	AppName	No	Filter by app name
app_version	AppVersion	No	Filter by version
record_ids	List[RecordID]	No	Filter by record IDs
limit	int	No	Max records to return
offset	int	No	Row offset

Outputs

Name	Type	Description
records_df	pd.DataFrame	Records with feedback score columns
feedback_cols	List[str]	Names of feedback columns in the DataFrame

Usage Examples

Query All Records

from trulens.core.session import TruSession

session = TruSession()
records_df, feedback_cols = session.get_records_and_feedback()
print(f"Feedback columns: {feedback_cols}")
print(records_df[feedback_cols].describe())

Query by App Name

records_df, feedback_cols = session.get_records_and_feedback(
    app_name="RAG_App",
    limit=50
)

Related Pages

Implements Principle

• Principle:Truera_Trulens_Records_And_Feedback_Query

Requires Environment

• Environment:Truera_Trulens_Python_Core_Environment