Implementation:Evidentlyai Evidently Dataset As Dataframe

Knowledge Sources	Evidently Evidently Datasets Guide
Domains	Data_Engineering, LLM_Evaluation
Last Updated	2026-02-14 12:00 GMT

Overview

Concrete method for extracting the underlying pandas DataFrame from an Evidently Dataset provided by the Evidently library.

Description

Dataset.as_dataframe() returns the underlying pandas DataFrame including all computed descriptor columns. It is an abstract method on Dataset implemented by PandasDataset.

Usage

Call on a Dataset object to retrieve the full DataFrame with computed descriptor columns.

Code Reference

Source Location

• Repository: evidently
• File: src/evidently/core/datasets.py
• Lines: L1297-1304 (abstract), L1622 (PandasDataset implementation)

Signature

class Dataset:
    @abstractmethod
    def as_dataframe(self) -> pd.DataFrame:
        """Get the underlying pandas.DataFrame with all computed descriptors."""

Import

from evidently import Dataset
# as_dataframe() is called on a Dataset instance

I/O Contract

Inputs

Name	Type	Required	Description
(none)			No parameters required

Outputs

Name	Type	Description
return value	pd.DataFrame	Full DataFrame with original + computed descriptor columns

Usage Examples

from evidently import Dataset, DataDefinition
from evidently.descriptors import Sentiment, NegativityLLMEval

dataset = Dataset.from_pandas(
    df,
    data_definition=DataDefinition(),
    descriptors=[
        Sentiment("response"),
        NegativityLLMEval("response"),
    ],
)

# Extract DataFrame with computed scores
eval_df = dataset.as_dataframe()
print(eval_df[["response", "Sentiment", "Negativity"]].head())

# Use for database storage
for _, row in eval_df.iterrows():
    insert_into_db(row["Sentiment"], row["Negativity"])

Related Pages

Implements Principle

• Principle:Evidentlyai_Evidently_Dataset_Score_Extraction