python - 在 Pandas 系列中设置日期时间值时的 SettingWithCopyWarning

Question

使用 iloc 在 pandas 系列中设置值时，我收到 SettingWithCopyWarning。我正在使用 Pandas 0.21.1 和 python 3.6.7

import pandas as pd
from datetime import datetime
from pytz import timezone

tz = timezone('CET')
ambiguous_dst = True

expected_dt_series = pd.Series([
    datetime(2018, 10, 28, 1),
    datetime(2018, 10, 28, 2),
    datetime(2018, 10, 28, 3),
    datetime(2018, 10, 28, 4),
])
expected_dt_series = expected_dt_series.dt.tz_localize(
    tz, ambiguous='NaT')

expected_dt_series.iloc[1] = tz.localize(
    datetime(2018, 10, 28, 2), is_dst=ambiguous_dst) # <- line that causes error

expected_dt_series = expected_dt_series.astype('object')

输出:

SettingWithCopyWarning: modifications to a method of a datetimelike object are not supported and are discarded. Change values on the original.
  self._setitem_with_indexer(indexer, value)
  x.py:17: SettingWithCopyWarning: modifications to a method of a datetimelike object are not supported and are discarded. Change values on the original.
    datetime(2018, 10, 28, 2), is_dst=ambiguous_dst)

1. 为什么会这样？我在原始系列中设置一个值
2. 我该怎么做才能避免这种情况？ (关闭警告除外)

Answer 1

确实是这部分导致了问题:

expected_dt_series = expected_dt_series.dt.tz_localize(tz, ambiguous='NaT')

使用copy明确告诉pandas这是它自己的系列

expected_dt_series = expected_dt_series.dt.tz_localize(
    tz, ambiguous='NaT').copy()

expected_dt_series.iloc[1] = tz.localize(
    datetime(2018, 10, 28, 2), is_dst=ambiguous_dst)