CODEWITHSUNDEEP
pythonarraysnumpydatetimedata-science
Max Kremlev
Max Kremlev

Reputation: 109

Formatting date data in NumPy array

I would be really grateful for an advice. I had an exercise like it's written bellow:

The first column (index 0) contains year values as four digit numbers in the format YYYY (2016, since all trips in our data set are from 2016). Use assignment to change these values to the YY format (16) in the test_array ndarray.

I used a code to solve it:

test_array[:,0] = test_array[:,0]%100

But I'm sure it has to be more universal and smart way to get the same results with datetime or smth else. But I cant find it. I tried different variations of this code, but I dont get whats wrong:

dt.datetime.strptime(str(test_array[:,0]), "%Y")
test_array[:,0] = dt.datetime.strftime("%y")

Could you help me with this, please?

Thank you

Upvotes: 4

Views: 1497

Answers (1)

bazinga
bazinga

Reputation: 146

In order to carry out the conversion of year from YYYY format to YY format would require intermediate datetime value on which operations such as strftime can be carried out in following manner: 

df.iloc[:, 0] = df.iloc[:, 0].apply(lambda x: pd.datetime(x, 1, 1).strftime('%y'))

Here to obtain the datetime values we needed 3 args: year, month and date, out of which we had year and the values for rest were assumed to be 1 as default.

Upvotes: 1

Related Questions