(*, axis=0, how='any', thresh=None, subset=None)


Removes rows with missing values (NaN, None and SQL NULL).


  • axis (int) – only axis=0 is supported. This means rows that contain missing values are dropped.

  • how (str) – determines when a row is removed. Supported values:

    • ’any’: rows with at least one missing value are removed
    • ’all’: rows with all missing values are removed
  • thresh (Optional[int]) – determines the least amount of non-missing values a row needs to have in order to be kept

  • subset (Optional[Union[str, Sequence[str]]]) – series label or sequence of labels to be considered for missing values. If subset is None, all DataFrame’s series labels will be used. In case subset is an empty list, a copy from the DataFrame will be returned.


a new dataframe with dropped rows.

Return type