python: assign nan to rows of Dataframe

user_15495707 0
Viewed 666

I have a pd Dataframe and some values are nan.

What I would like to do is assign nan values to all the elements of a row (excluding the element of the first column) if that row has one nan value.

For example, given the following dataframe:

Code   1996   1997   1998
GBA    100    nan     5
JOY    120    10      30
WII    300    nan     nan

the desired output is:

Code   1996   1997   1998
GBA    nan    nan     nan
JOY    120    10      30
WII    nan    nan     nan
pandas python
1 Answers

You can first get a boolean series that says whether a row has any NaNs in it starting from first column. Then you can use boolean indexing and set those rows' columns starting from first to NaN:

has_nan = df.loc[:, "1996":].isna().any(axis=1)

df.loc[has_nan, "1996":] = np.nan

to get

>>> df

  Code   1996  1997  1998
0  GBA    NaN   NaN   NaN
1  JOY  120.0  10.0  30.0
2  WII    NaN   NaN   NaN

has_nan is

0     True
1    False
2     True
dtype: bool

i.e., first and last rows are NaNed.

user_93321870
Related