Pandas DataFrame cummax() Method

Example

Return the cumulative maximum value of each row:

import pandas as pd

data = [[10, 18, 11], [13, 15, 8], [9, 20, 3]]

df = pd.DataFrame(data)

print(df.cummax())

Definition and Usage

The cummax() method returns a DataFrame with the cumulative maximum values.

The cummax() method goes through the values in the DataFrame, from the top, row by row, replacing the values with the highest value yet for each column, ending up with a DataFrame where the last row contains only the highest value from each column.

If the axis parameter is set to axes='columns', the method goes through the values, column by column, and ends up with a DataFrame where the last columns contains only the highest value from each row.

Syntax

dataframe.cummax(axis, skipna, args, kwargs)

Parameters

The axis and skipna parameters are keyword arguments.

Parameter	Value	Description
axis	`0 1 'index' 'columns'`	Optional, default 0, specifies the axis to run the accumulation over.
skip_na	`True False`	Optional, default True. Set to False if the result should NOT skip NULL values
args		Optional. These arguments has no effect, but could be accepted by a NumPy function
kwargs		Optional, keyword arguments. These arguments has no effect, but could be accepted by a NumPy function

Return Value

A DataFrame object.

This function does NOT make changes to the original DataFrame object.

❮ DataFrame Reference