How to use pandas tz_convert to convert to multiple different time zones

Question

I have some data as shown below with hour in UTC. I want to create a new column named local_hour based on time_zone. How can I do that? It seems like pandas' tz_convert does not allow a column or pandas series as input to the tz argument.

# Create dataframe
import pandas as pd
df = pd.DataFrame({
    'hour': ['2019-01-01 05:00:00', '2019-01-01 07:00:00', '2019-01-01 08:00:00'],
    'time_zone': ['US/Eastern', 'US/Central', 'US/Mountain']
})

# Convert hour to datetime and localize to UTC
df['hour'] = pd.to_datetime(df['hour']).dt.tz_localize('UTC')

df 
        hour                     time_zone
0   2019-01-01 05:00:00+00:00   US/Eastern
1   2019-01-01 07:00:00+00:00   US/Central
2   2019-01-01 08:00:00+00:00   US/Mountain

# Create local_hour column to convert hour to US/Eastern time (this works)
df['local_hour'] = df['hour'].dt.tz_convert(tz='US/Eastern')
df
    hour                        time_zone   local_hour
0   2019-01-01 05:00:00+00:00   US/Eastern  2019-01-01 00:00:00-05:00
1   2019-01-01 07:00:00+00:00   US/Central  2019-01-01 02:00:00-05:00
2   2019-01-01 08:00:00+00:00   US/Mountain 2019-01-01 03:00:00-05:00

# Try to create local_hour column to convert hour based on time_zone column (this fails) 
df['local_hour'] = df['hour'].dt.tz_convert(tz=df['time_zone'])
ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

Answer 1

dt.tz_convert expects a scalar value for its tz param, not a list of timezone-like values. Use apply, which is essentially a loop:

df['local_hour'] = df.apply(lambda row: row['hour'].tz_convert(row['time_zone']), axis=1)

Answer 2

You can use babel and datetime instead:

import pandas as pd
from datetime import datetime
from babel.dates import format_datetime,get_timezone
# Create dataframe
df = pd.DataFrame({
    'hour': ['2019-01-01 05:00:00', '2019-01-01 07:00:00', '2019-01-01 08:00:00'],
    'time_zone': ['US/Eastern', 'US/Central', 'US/Mountain']
})
# First: convert hour column items to datetime objects
df['hour']=df['hour'].map(lambda hh: datetime.strptime(hh, '%Y-%m-%d %H:%M:%S'))
# Second: Try to create local_hour column to convert hour based on time_zone column
df['local_hour']=df[['hour','time_zone']].apply(lambda x:  format_datetime(x[0], "yyyy-MM-dd HH:mm:ssZZ", 
  tzinfo=get_timezone(x[1]), locale='en'),axis=1)
# Third: Convert hour to datetime and localize to UTC (this was your first step)
df['hour']=df['hour'].map(lambda hh: format_datetime(hh, "yyyy-MM-dd HH:mm:ssZZ",
  tzinfo=get_timezone('UTC'), locale='en'))

df
                        hour     time_zone                 local_hour
0   2019-01-01 05:00:00+0000    US/Eastern   2019-01-01 00:00:00-0500
1   2019-01-01 07:00:00+0000    US/Central   2019-01-01 01:00:00-0600
2   2019-01-01 08:00:00+0000    US/Mountain  2019-01-01 01:00:00-0700