Implementation:Evidentlyai Evidently Column Statistics Classes

Knowledge Sources	Evidently Evidently Metrics
Domains	Data_Quality, Statistics
Last Updated	2026-02-14 12:00 GMT

Overview

Concrete metric classes for computing column-level statistics provided by the Evidently library.

Description

Individual metric classes for column statistics:

• MeanValue(column): Computes arithmetic mean
• CategoryCount(column, category): Counts category occurrences (count and share)
• InRangeValueCount(column, left, right): Counts values in range (count and share)
• MissingValueCount(column): Counts null/NaN values (count and share)

Usage

Import from evidently.metrics and include in Report metrics lists.

Code Reference

Source Location

• Repository: evidently
• File: src/evidently/metrics/column_statistics.py
• Lines: L264-524 (CategoryCount, InRangeValueCount, MissingValueCount, MeanValue)

Signature

class MeanValue(StatisticsMetric):
    """Column mean. Inherits column from ColumnMetric."""
    # column: str (inherited)

class CategoryCount(ColumnMetric, CountMetric):
    category: Optional[Label] = None
    categories: List[Label] = []
    def __init__(self, column: str, categories: Optional[List[Label]] = None,
                 category: Optional[Label] = None, tests: Optional[List[MetricTest]] = None,
                 share_tests: Optional[List[MetricTest]] = None):

class InRangeValueCount(ColumnMetric, CountMetric):
    left: Union[int, float]
    right: Union[int, float]

class MissingValueCount(ColumnMetric, CountMetric):
    # column: str (inherited)

Import

from evidently.metrics import MeanValue, CategoryCount, InRangeValueCount, MissingValueCount

I/O Contract

Inputs

Metric	Key Parameters	Description
MeanValue	column: str	Column to compute mean of
CategoryCount	column: str, category: Label	Column and category to count
InRangeValueCount	column: str, left: float, right: float	Column and range bounds
MissingValueCount	column: str	Column to check for missing values

Outputs

Metric	Returns	Description
MeanValue	float	Arithmetic mean
CategoryCount	(count: int, share: float)	Category count and proportion
InRangeValueCount	(count: int, share: float)	In-range count and proportion
MissingValueCount	(count: int, share: float)	Missing count and proportion

Usage Examples

from evidently import Report
from evidently.metrics import MeanValue, CategoryCount, InRangeValueCount, MissingValueCount

report = Report([
    MeanValue(column="Sentiment"),
    CategoryCount(column="Negativity", category="negative"),
    InRangeValueCount(column="text_length", left=10, right=500),
    MissingValueCount(column="review"),
])

snapshot = report.run(current_dataset, reference_dataset)

Related Pages

Implements Principle

• Principle:Evidentlyai_Evidently_Column_Statistics_Metrics