Missing Values

Table of contents

1. Missing Data
   1. isnull() == isna() (alias) $\longleftrightarrow$ notnull() == notna() (alias)
   2. fillna()
   3. replace()

import pandas as pd

Missing Data

isnull() == isna() (alias) $\longleftrightarrow$ notnull() == notna() (alias)

DataFrame.isnull()          # return : DataFrame
Series.isnull()             # return : Series
Index.isnull()              # return : numpy.ndarray[bool]
pandas.isnull(obj)          # return : bool or array-like of bool

pandas document - pandas.DataFrame.isna

return : boolean_mask

Q. How many missing values in df.column_name? The following 3 are equivalent.

df.column_name.isna().sum()             # equivalent
pd.isna(df.coloumn_name).sum()          # equivalent
len(df[df.coloumn_name.isna()])         # equivalent

fillna()

DataFrame.fillna(value, method, axis, inplace)
Series.fillna(value, method, axis, inplace)
Index.fillna(value)         # value : scalar only

pandas document - pandas.DataFrame.fillna

1. value : scalar, dict, Series, or DataFrame
   • scalar : value to use to fill holes
   • dict, Series, or DataFrame : specify values to use for each index (for a Series) or column (for a DataFrame)
   • list : unable
2. method : {‘backfill’, ‘bfill’, ‘pad’, ‘ffill’, None}, default None
   • pad / ffill: propagate last valid observation forward to next valid
   • backfill / bfill: use next valid observation to fill gap.
3. axis : {0 or ‘index’, 1 or ‘columns’}
4. inplace : bool, default False

replace()

DataFrame.replace(to_replace, value, inplace)
Series.replace(to_replace, value, inplace)

pandas document - pandas.DataFrame.replace

1. to_replace : str, regex, list, dict, Series, int, float, or None
   • how to find the values that will be replaced
2. value : scalar, dict, list, str, regex, default None
   • value to replace any values matching to_replace with
3. inplace : bool, default False