Dealing with 'OutOfBoundsDatetime: Out of bounds nanosecond timestamp:'

Question

I am working on a DataFrame looking at baseball games Date and their Attendance so I can create a Calendar Heatmap.

    Date    Attendance
1   Apr 7   44723.0
2   Apr 8   42719.0
3   Apr 9   36139.0
4   Apr 10  41253.0
5   Apr 11  20480.0

I’ve tried different solutions that I’ve come across…

- df['Date'] = df['Date'].astype('datetime64[ns]')
- df['Date'] = pd.to_datetime(df['Date'])

but I’ll get the error of

‘Out of bounds nanosecond timestamp: 1-04-07 00:00:00’.

From looking at my data, I don’t even have a date that goes with that timestamp. I also looked at other posts on this site, and 1 potential problem is that my Dates are NOT zero padded? Could that be the cause?

Asked By: Addison Choi

||

Source

Answer 1

you can convert to datetime if you supply a format; Ex:

df
Out[33]: 
     Date   Attendance
1   Apr 7      44723.0
2   Apr 8      42719.0
3   Apr 9      36139.0
4  Apr 10      41253.0
5  Apr 11      20480.0

pd.to_datetime(df["Date"], format="%b %d")
Out[35]: 
1   1900-04-07
2   1900-04-08
3   1900-04-09
4   1900-04-10
5   1900-04-11
Name: Date, dtype: datetime64[ns]

If you’re unhappy with the base year 1900, you can add a date offset, for example

df["datetime"] = pd.to_datetime(df["Date"], format="%b %d")

df["datetime"] += pd.tseries.offsets.DateOffset(years=100)

df["datetime"]
1   2000-04-07
2   2000-04-08
3   2000-04-09
4   2000-04-10
5   2000-04-11
Name: datetime, dtype: datetime64[ns]

Answered By: FObersteiner

Dealing with 'OutOfBoundsDatetime: Out of bounds nanosecond timestamp:'

Question:

Answers: