CODEWITHSUNDEEP
pythonpandasdatetime
Jonas Palačionis
Jonas Palačionis

Reputation: 4842

Converting timestamp to date and keeping only date in a list inside pandas column

I have a df which looks like:

df

date

[2021-08-31 00:00:00]
[2021-07-02 00:00:00, 2021-07-02 00:00:00]
[2021-08-31 00:00:00, 2021-09-15 00:00:00]

When I export it to .csv I get a column looking like :

date

[Timestamp('2021-08-31 00:00:00')]
[Timestamp('2021-07-02 00:00:00'), Timestamp('2021-07-02 00:00:00')]
[Timestamp('2021-08-31 00:00:00'), Timestamp('2021-09-15 00:00:00')]

I want the .csv file to have the same look as in df without the Timestamp and preferably without the hours, minutes, and seconds.

What I tried:

Checked if I can convert it like so:

for nr, item in enumerate(df['date']):
    print(pd.to_datetime(item[nr],format='%Y%m%d'))
    print(type(item[nr]))
    break

But it still returns seconds which I don't understand why if I added the format parameter.

2021-08-31 00:00:00
<class 'pandas._libs.tslibs.timestamps.Timestamp'>

I tried:

def to_date(df):
    res = []
    for nr, item in enumerate(df['date']):
        res.append(item[nr].date())
    return res

df['test'] = to_date(df)

But I get:

IndexError: list index out of range

And I am not sure how I should convert it so that in my .csv file I would get dates that would look like this 2021-08-31.

Upvotes: 0

Views: 556

Answers (1)

Anurag Dabas
Anurag Dabas

Reputation: 24304

try via map() and list comprehension:

df['date']=df['date'].map(lambda x:[str(y) for y in x])
#you can also use apply() or agg() but they are slower then map()

Finally use to_csv() method:

df.to_csv('filename.csv')

Update:

If you want only date part then:

df['date']=df['date'].map(lambda x:[str(y.date()) for y in x])
#as suggested by @MrFuppes

Upvotes: 1

Related Questions