How to check and replace NaN from a dataframe column-CodePudding

I have a DataFrame with 2 columns.

col_1	col_2
apple	NaN
pear	NaN
mango	NaN
NaN	Strawberry

I want to check col_2 for NaN, If the value is not NULL or NaN copy value of col_2 to col_1.

df = {'col_1': ['apple', 'pear', 'mango',''],
      'col_2': [Nan,Nan,Nan,'Strawberry' ]}
print(df)
for idx, row in df.iterrows():
       if df.loc[idx, col_2].notna:
            print("inside if")
       else:
            print("inside else")

every time it goes in if and never goes in else.

CodePudding user response：

I think this will solve your problem, which says to copy non-NaN values from col_2 to col_1. Since that's a rather unusual operation, I'm uncertain whether you meant to do that or maybe meant copy values from col1 to col2.

Nan = np.nan
df = pd.DataFrame({'col_1': ['apple', 'pear', 'mango',''],
      'col_2': [Nan,Nan,Nan,'Strawberry' ]})
df.col_1.mask(df.col_2.notna(), df.col_2)

CodePudding user response：

Backfill values from col_2 to col_1.

Given:

   col_1       col_2
0  apple         NaN
1   pear         NaN
2  mango         NaN
3    NaN  Strawberry

Doing:

df = df.bfill(axis=1)
print(df)

Output:

        col_1       col_2
0       apple         NaN
1        pear         NaN
2       mango         NaN
3  Strawberry  Strawberry

CodePudding user response：

This solved the issue.

df['col_2'].fillna("", inplace = True)