Turn NaN in dataframe if condition met

Question

I have a column here that looks like this and is part of a dataframe:

df.Days_Since_Earnings
Out[5]: 
0      21.0
2       1.0
4    1000.0
5     500.0
6     119.0
Name: Days_Since_Earnings, Length: 76, dtype: float64

I want to leave it as it is except I want to turn numbers above 120 to 'nan's, so it would look like this:

df.Days_Since_Earnings
Out[5]: 
0      21.0
2       1.0
4       nan
5       nan
6     119.0
Name: Days_Since_Earnings, Length: 76, dtype: float64

thanks to anyone who helps!

Does this answer your question? [How to set a cell to NaN in a pandas dataframe](https://stackoverflow.com/questions/34794067/how-to-set-a-cell-to-nan-in-a-pandas-dataframe) — deadshot, May 21 '20 at 13:30
`idx_days_over_120 = df["Days_Since_Earnings"] > 120` and then `df.loc[idx_days_over_120, "Days_Since_Earnings"] = np.nan` — Dan, May 21 '20 at 13:49

score 2 · Accepted Answer · answered May 21 '20 at 14:07

You can use mask:

df['Days_Since_Earnings'] = df.Days_Since_Earnings.mask(df.Days_Since_Earnings > 120)

or where with reverse condition

df['Days_Since_Earnings'] = df.Days_Since_Earnings.where(df.Days_Since_Earnings <= 120)

or loc assignment:

df.loc[df.Days_Since_Earnings > 120, 'Days_Since_Earnings'] = np.nan

NYC Coder · Answer 2 · 2020-05-21T13:37:12.663

1

df['days'] = df['days'].apply(lambda x: np.nan if x > 120 else x)
print(df)

Or

df[df['days'] > 120] = np.nan

    days
0   21.0
1    1.0
2    NaN
3    NaN
4  119.0

edited May 21 '20 at 13:37

answered May 21 '20 at 13:29

NYC Coder

1

Using `apply` for this is likely to be very inefficient: https://stackoverflow.com/a/55557758/1011724 – Dan May 21 '20 at 13:32

2 Answers2