CODEWITHSUNDEEP
pythonpandas
Héctor Gómez Meneses
Héctor Gómez Meneses

Reputation: 3

how to convert an object type column to datetime in panda

I have one column given as type object and want to convert it to a datetime:

DataFrame: df

Column: date

Value Format: 1880-01-01T00:00:00.000

Dtype: object

When doing: df['date'] = pd.to_datetime(dt['date'])

I got the error: Out of bounds nanosecond timestamp: 1583-01-01 00:00:00

Even specifiyng the format: df['date'] = pd.to_datetime(df['date'], format = '%Y-%m-%dT%H:%M:%S.%f')

I got the same error, any idea why that error occurs and how to solve it?

Upvotes: 0

Views: 181

Answers (1)

Jones
Jones

Reputation: 92

With 1880-01-01T00:00:00.000, it should work. But from your error message, it seems that your value is 1583-01-01T00:00:00.000 and it is out of pandas timestamps representation range.
In nanosecond resolution, the timespan that can be represented using a 64-bit integer is limited to approximately 584 years, which is:

print(pd.Timestamp.min)
Output: 1677-09-21 00:12:43.145225

print(pd.Timestamp.max)
Output: 2262-04-11 23:47:16.854775807

But if your value is 1880-01-01T00:00:00.000, after specifying the format:
df['date'] = pd.to_datetime(df['date']),
when I print df['date'], I get:

1880-01-01 00:00:00
Name: date, dtype: datetime64[ns]

Upvotes: 1

Related Questions