[Remove missing data from pandas DataFrame] #pandas #python

remove_missing.py

import pandas as pd


def remove_missing(df: pd.DataFrame) -> pd.DataFrame:
    """
    Remove missing data from a DataFrame.

    Iteratively remove the column or row that contains the highest proportion of
    missing data, until there is no missing data left.
    """
    n_rows, n_cols = df.shape
    n_row_na = df.isna().sum(axis=1) / n_cols
    n_col_na = df.isna().sum(axis=0) / n_rows
    row_na_max = n_row_na.max()
    col_na_max = n_col_na.max()
    if row_na_max == col_na_max == 0:
        return df
    elif row_na_max >= col_na_max:
        return remove_missing(df.iloc[n_row_na.values != row_na_max])
    else:
        return remove_missing(df.iloc[:, n_col_na.values != col_na_max])