Convert datatype Pandas Dataframe

Question

Starting from a Pandas Dataframe with it's setup:

B13-111DATA.TIJD    object
dtype: object
                    B13-111DATA.TIJD
StartTime
2020-03-30 00:00:00              292
2020-03-30 00:00:01              292
2020-03-30 00:00:02              292
2020-03-30 00:00:03              292
2020-03-30 00:00:04              292
...                              ...
2020-04-07 23:59:55              333
2020-04-07 23:59:56              333
2020-04-07 23:59:57              333
2020-04-07 23:59:58              333
2020-04-07 23:59:59              333

[777600 rows x 1 columns]

This Pandas Dataframe I would like to transform to a structuce like below:

B13-111DATA.TIJD    int64 
dtype: object

or

B13-111DATA.TIJD    float64
dtype: object

I tried to use following line:

df = df[B13-111DATA.TIJD].astype(float)

But it returns me a simple "float" and errors my code

print(output.columns.values)

with an error "AttributeError: 'Series' object has no attribute 'columns'". It looks my dataFrame turned into a series. Could that be the case?

Pretty sure it is something simple many people already encountered here. Any tip or help would be appreciated.

score 1 · Accepted Answer · answered Apr 22 '20 at 14:46

Problem is there is reaasign DataFrame variable df to Series (column of DataFrame):

df = df['B13-111DATA.TIJD'].astype(float)

For correct converting assign back column, so df stay DataFrame:

df['B13-111DATA.TIJD'] = df['B13-111DATA.TIJD'].astype(float)
print (df)

Convert datatype Pandas Dataframe

1 Answers1