pandas.DataFrame.all¶
- DataFrame.all(axis=None, bool_only=None, skipna=True, level=None, **kwargs)¶
Return whether all elements are True over requested axis. %(na_action)s
Parameters: axis : {0, 1}
0 for row-wise, 1 for column-wise
skipna : boolean, default True
Exclude NA/null values. If an entire row/column is NA, the result will be NA
level : int, default None
If the axis is a MultiIndex (hierarchical), count along a particular level, collapsing into a DataFrame
bool_only : boolean, default None
Only include boolean data.
Returns: any : Series (or DataFrame if level specified)