更改pandas中日期时间列的时区并添加为分层索引

I have data with a time-stamp in UTC. I'd like to convert the timezone of this timestamp to 'US/Pacific' and add it as a hierarchical index to a pandas DataFrame. I've been able to convert the timestamp as an Index, but it loses the timezone formatting when I try to add it back into the DataFrame, either as a column or as an index.

我的数据带有UTC时间戳。我想将此时间戳的时区转换为“US / Pacific”,并将其作为分层索引添加到pandas DataFrame中。我已经能够将时间戳转换为索引,但是当我尝试将其添加回DataFrame时,它会丢失时区格式,无论是作为列还是作为索引。

>>> import pandas as pd
>>> dat = pd.DataFrame({'label':['a', 'a', 'a', 'b', 'b', 'b'], 'datetime':['2011-07-19 07:00:00', '2011-07-19 08:00:00', '2011-07-19 09:00:00', '2011-07-19 07:00:00', '2011-07-19 08:00:00', '2011-07-19 09:00:00'], 'value':range(6)})
>>> dat.dtypes
#datetime    object
#label       object
#value        int64
#dtype: object

Now if I try to convert the Series directly I run into an error.

现在,如果我尝试直接转换系列,我会遇到错误。

>>> times = pd.to_datetime(dat['datetime'])
>>> times.tz_localize('UTC')
#Traceback (most recent call last):
#  File "<stdin>", line 1, in <module>
#  File "/Users/erikshilts/workspace/schedule-detection/python/pysched/env/lib/python2.7/site-packages/pandas/core/series.py", line 3170, in tz_localize
#    raise Exception('Cannot tz-localize non-time series')
#Exception: Cannot tz-localize non-time series

If I convert it to an Index then I can manipulate it as a timeseries. Notice that the index now has the Pacific timezone.

如果我将其转换为索引,那么我可以将其作为时间序列进行操作。请注意,索引现在具有太平洋时区。

>>> times_index = pd.Index(times)
>>> times_index_pacific = times_index.tz_localize('UTC').tz_convert('US/Pacific')
>>> times_index_pacific
#<class 'pandas.tseries.index.DatetimeIndex'>
#[2011-07-19 00:00:00, ..., 2011-07-19 02:00:00]
#Length: 6, Freq: None, Timezone: US/Pacific

However, now I run into problems adding the index back to the dataframe as it loses its timezone formatting:

但是,现在我遇到了将索引添加回数据帧的问题,因为它丢失了时区格式:

>>> dat_index = dat.set_index([dat['label'], times_index_pacific])
>>> dat_index
#                                      datetime label  value
#label                                                      
#a     2011-07-19 07:00:00  2011-07-19 07:00:00     a      0
#      2011-07-19 08:00:00  2011-07-19 08:00:00     a      1
#      2011-07-19 09:00:00  2011-07-19 09:00:00     a      2
#b     2011-07-19 07:00:00  2011-07-19 07:00:00     b      3
#      2011-07-19 08:00:00  2011-07-19 08:00:00     b      4
#      2011-07-19 09:00:00  2011-07-19 09:00:00     b      5

You'll notice the index is back on the UTC timezone instead of the converted Pacific timezone.

您会注意到索引返回UTC时区而不是转换后的太平洋时区。

How can I change the timezone and add it as an index to a DataFrame?

如何更改时区并将其添加为DataFrame的索引?

4 个解决方案

#1

By now this has been fixed. For example, you can now call:

到目前为止,这已得到修复。例如,您现在可以调用:

dataframe.tz_localize('UTC', level=0)

You'll have to call it twice for the given example, though. (I.e., once for each level.)

但是,对于给定的示例,您必须为它调用两次。 (即,每个级别一次。)

#2

If you set it as the index, it's automatically converted to an Index:

如果将其设置为索引,它会自动转换为索引:

In [11]: dat.index = pd.to_datetime(dat.pop('datetime'), utc=True)

In [12]: dat
Out[12]:
                    label  value
datetime
2011-07-19 07:00:00     a      0
2011-07-19 08:00:00     a      1
2011-07-19 09:00:00     a      2
2011-07-19 07:00:00     b      3
2011-07-19 08:00:00     b      4
2011-07-19 09:00:00     b      5

Then do the tz_localize:

然后执行tz_localize:

In [12]: dat.index = dat.index.tz_localize('UTC').tz_convert('US/Pacific')

In [13]: dat
Out[13]:
                          label  value
datetime
2011-07-19 00:00:00-07:00     a      0
2011-07-19 01:00:00-07:00     a      1
2011-07-19 02:00:00-07:00     a      2
2011-07-19 00:00:00-07:00     b      3
2011-07-19 01:00:00-07:00     b      4
2011-07-19 02:00:00-07:00     b      5

~~And then you can append the label column to the index:~~

然后,您可以将标签列附加到索引:

Hmmm this is definitely a bug!

嗯,这绝对是一个错误!

In [14]: dat.set_index('label', append=True).swaplevel(0, 1)
Out[14]:
                           value
label datetime
a     2011-07-19 07:00:00      0
      2011-07-19 08:00:00      1
      2011-07-19 09:00:00      2
b     2011-07-19 07:00:00      3
      2011-07-19 08:00:00      4
      2011-07-19 09:00:00      5

A hacky workaround is to convert the (datetime) level directly (when it's already a MultiIndex):

一个hacky解决方法是直接转换(datetime)级别(当它已经是MultiIndex时):

In [15]: dat.index.levels[1] = dat.index.get_level_values(1).tz_localize('UTC').tz_convert('US/Pacific')

In [16]: dat1
Out[16]:
                                 value
label datetime
a     2011-07-19 00:00:00-07:00      0
      2011-07-19 01:00:00-07:00      1
      2011-07-19 02:00:00-07:00      2
b     2011-07-19 00:00:00-07:00      3
      2011-07-19 01:00:00-07:00      4
      2011-07-19 02:00:00-07:00      5

#3

An other workaround which works in pandas 0.13.1, and solves the FrozenList can not be assigned problem:

在pandas 0.13.1中工作的另一种解决方法,并解决了FrozenList无法分配的问题:

index.levels = pandas.core.base.FrozenList([
    index.levels[0].tz_localize('UTC').tz_convert(tz),
    index.levels[1].tz_localize('UTC').tz_convert(tz)
])

Struggling a lot with this issue, MultiIndex loses tz in many other conditions too.

在这个问题上苦苦挣扎,MultiIndex在许多其他条件下也失去了tz。

#4

The workaround does not seem to work because the index levels of a hierarchical index seem to be immutable (FrozenList is immutable).

解决方法似乎不起作用,因为层次索引的索引级别似乎是不可变的(FrozenList是不可变的)。

Starting with a singular index and appending also does not work.

从单数索引开始并附加也不起作用。

Creating a lambda function that casts as Timestamp and converts each member of the Series returned by to_datetime() also does not work.

创建一个转换为Timestamp并转换由to_datetime()返回的Series的每个成员的lambda函数也不起作用。

Is there a way to create timezone aware Series and then insert them into a dataframe/make them an index?

有没有办法创建时区感知系列,然后将它们插入数据帧/使它们成为索引?

joined_event_df = joined_event_df.set_index(['pandasTime'])
joined_event_df.index = joined_event_df.index.get_level_values(1).tz_localize('UTC').tz_convert('US/Central')
# we have tz-awareness above this line
joined_event_df = joined_event_df.set_index('sequence', append = True)
# we lose tz-awareness in the index as soon as we add another index
joined_event_df = joined_event_df.swaplevel(0,1)

#1