python - Pandas drop rows where all but one column is NaN

Question

I am reading data from a csv file into a pandas dataframe. Now some of the rows contain data in only one column (can be any column) and rest columns are NaN. The actual data may also contain some NaNs, therefore I want to drop all such rows where exactly one column contains data and rest are NaNs.

Therefore neither df = df.dropna(how='all') nor df = df.dropna(how='any') solves my issue.

Sample dataframe:

1     2    3    ... 19   20
text  NaN  NaN  ... NaN  NaN
NaN   abc  NaN  ... NaN  NaN
NaN   NaN  xyz  ... NaN  NaN
a     b    c    ... NaN    z
.
.

What I want become:

1     2    3    ... 19   20
a     b    c    ... NaN    z
.
.

question from:https://stackoverflow.com/questions/65919332/pandas-drop-rows-where-all-but-one-column-is-nan

Answer 1

Waitting for answers