CODEWITHSUNDEEP
datetimematplotlib
Chen Lizi
Chen Lizi

Reputation: 103

convert pandas series index to datetime

I have a pandas series/dataframe that has some built in time index which is of pandas.core.indexes.base.Index type. It goes like this:

Index(['2015-07-13', '2015-07-14',...],
      dtype='object', length=1238)

I want to convert this time index, a list-like thing to a list of strictly datetime.date type thing such that Matplotlib can plot the converted list on the x axis.
To be specific, I want it to be converted and have a format looking like this list:

[datetime.date(2015, 7, 13),datetime.date(2015, 7, 14),datetime.date(2015, 7, 15),
 datetime.date(2015, 7, 16),...]

Btw, the way I the list above is through this simple block of code:

dates = [datetime.date(2015, 7, 13)]

num=len(ts_test.index)
for _ in range(num-1):
    dates.append(dates[-1] + datetime.timedelta(days=1))

The problem is that the time index I try to plot on the x-axis skips weekends and such, so the list "dates" here has all the unwanted weekends and holidays.

so to put it simply, how to convert '2015-07-13', a string to datetime.date(2015, 7, 13)?

Thank you very much!

Upvotes: 0

Views: 432

Answers (2)

Blas Ortellado
Blas Ortellado

Reputation: 76

If you want to convert your index column, of type string, to an index column of type timestamp, a straightforward solution could be:

dates_test.index = pd.to_datetime(dates_test.index)

To extract a list later on:

timeList = dates_test.index.to_list()

Upvotes: 1

Chen Lizi
Chen Lizi

Reputation: 103

I ran into a solution and it works nicely!

dates_test=[]
for i in ts_test.index:
    date_str=i
    format_str='%Y-%m-%d' 
    datetime_obj = datetime.datetime.strptime(date_str, format_str)
    dates_test.append(datetime_obj.date())

So now you have managed to convert a list of str to a list of datetime type of objects which Matplotlib can fully handle and plot nicely along the X-axis!

Upvotes: 0

Related Questions