I currently have a dataframe table shown below:
Day_Part Start_Time End_Time
Quarter 1 2014, 1, 1 2015, 3, 1
Quarter 2 2014, 3, 3 2014, 7, 3
The columns "Start_Time" and "End_Time" are pandas Series. I want to convert the datatype of both columns to datetime instead.
I need those two columns to have datetime datatypes because in a later block of code, I say if this column is between this date and this date then I label it Quarter 1.
*Any help is greatly appreciated
You can use to_datetime() with a format string to extract the date:
date = pd.to_datetime(df.Start_Time, format='%Y, %m, %d').dt.date
You can also modify the dates in place:
df[['Start_Time', 'End_Time']] = df[['Start_Time', 'End_Time']].apply(
lambda x: pd.to_datetime(x, format='%Y, %m, %d').dt.date)
Or, you can convert to date when reading the csv:
to_date = lambda x: pd.to_datetime(x, format='%Y, %m, %d').date()
converters = dict(Start_Time=to_date, End_Time=to_date)
df = pd.read_csv(StringIO(data), converters=converters)
A testable example:
import pandas as pd
from io import StringIO
data = u"""
Day_Part,Start_Time,End_Time
"Quarter 1","2014, 1, 1","2015, 3, 1"
"Quarter 2","2014, 3, 3","2014, 7, 3"
"""
df = pd.read_csv(StringIO(data))
# You can use `to_datetime()` with a format string to extract the date:
date = pd.to_datetime(df.Start_Time, format='%Y, %m, %d').dt.date
# The start month in the second row is 3
assert date[1].month == 3
# You can also modify in place
df[['Start_Time', 'End_Time']] = df[['Start_Time', 'End_Time']].apply(
lambda x: pd.to_datetime(x, format='%Y, %m, %d').dt.date)
# The end month in the second row is 7
assert df.End_Time[1].month == 7
# You can convert to date when reading the csv
to_date = lambda x: pd.to_datetime(x, format='%Y, %m, %d').date()
converters = dict(Start_Time=to_date, End_Time=to_date)
df = pd.read_csv(StringIO(data), converters=converters)
# The end month in the first row is 3
assert df.End_Time[0].month == 3
df.head()anddf.dtypes. I suspect your columns haveobjectdtypes.