Python 如何使用字符串访问 Pandas DataFrame 日期时间索引
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/36871188/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
How to access pandas DataFrame datetime index using strings
提问by Pedro Braz
This is a very simple and practical question. I have the feeling that it must be a silly detail and that there should be similar questions. I wasn't able to find them tho. If someone does I'll happily delete this one.
这是一个非常简单实用的问题。我有一种感觉，这一定是一个愚蠢的细节，应该有类似的问题。我没能找到他们。如果有人这样做，我会很乐意删除这个。
The closest I found were these: pandas: iterating over DataFrame index with loc
我找到的最接近的是这些： pandas: iterating over DataFrame index with loc
How to select rows within a pandas dataframe based on time only when index is date and time
anyway, the thing is, I have a datetime indexed panda dataframe as follows:
无论如何，问题是，我有一个日期时间索引的熊猫数据框，如下所示：
In[81]: y
Out[81]:
PETR4 CSNA3 VALE5
2008-01-01 0.0 0.0 0.0
2008-01-02 1.0 1.0 1.0
2008-01-03 7.0 7.0 7.0
In[82]: y.index
Out[82]: DatetimeIndex(['2008-01-01', '2008-01-02', '2008-01-03'], dtype='datetime64[ns]', freq=None)
Oddly enough, I can't access its values using none of the following methods:
奇怪的是，我无法使用以下任何方法访问其值：
In[83]: y[datetime.datetime(2008,1,1)]
In[84]: y['2008-1-1']
In[85]: y['1/1/2008']
I get the KeyErrorerror.
我得到了KeyError错误。
Even more weird is that the following methods DO work:
更奇怪的是以下方法确实有效：
In[86]: y['2008']
Out[86]:
PETR4 CSNA3 VALE5
2008-01-01 0.0 0.0 0.0
2008-01-02 1.0 1.0 1.0
2008-01-03 7.0 7.0 7.0
In[87]: y['2008-1']
Out[87]:
PETR4 CSNA3 VALE5
2008-01-01 0.0 0.0 0.0
2008-01-02 1.0 1.0 1.0
2008-01-03 7.0 7.0 7.0
I'm fairly new to pandas so maybe I'm missing something here?
我对大熊猫相当陌生，所以也许我在这里遗漏了一些东西？
回答by piRSquared
pandas is taking what's inside the []and deciding what it should do. If it's a subset of column names, it'll return a DataFrame with those columns. If it's a range of index values, it'll return a subset of those rows. What is does not handle is taking a single index value.
pandas 正在获取里面的内容[]并决定它应该做什么。如果它是列名的子集，它将返回一个包含这些列的 DataFrame。如果它是一系列索引值，它将返回这些行的子集。不处理的是采用单个索引值。
Solution
解决方案
Two work around's
两个解决方法
1.Turn the argument into something pandas interprets as a range.
1. 将参数转换为 pandas 解释为范围的内容。
df['2008-01-01':'2008-01-01']
2.Use the method designed to give you this result. loc[]
2.使用旨在给你这个结果的方法。 loc[]
df.loc['2008-01-01']
回答by Scratch'N'Purr
You can use the to_pydatetimefunction on your index so thus:
您可以在索引上使用to_pydatetime函数，因此：
y[y.index.to_pydatetime() == datetime.datetime(2008,1,1)]
回答by bob_monsen
Reversing your dataframe allows the indexing to work:
反转数据框允许索引工作：
Here is your .csv datafile:
这是您的 .csv 数据文件：
Date,PETR4,CSNA3,VALE5
2008-01-01,0.0,0.0,0.0
2008-01-02,1.0,1.0,1.0
2008-01-03,7.0,7.0,7.0
Use the following incantation to read it into a DataFrame:
使用以下咒语将其读入 DataFrame：
>>> a = pd.read_csv('your.csv', index_col=0, parse_dates=True, infer_datetime_format=True)
Then, try to index a row:
然后，尝试索引一行：
>>> a['2008-01-01']
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/usr/local/lib/python2.7/dist-packages/pandas/core/frame.py", line 1969, in __getitem__
return self._getitem_column(key)
File "/usr/local/lib/python2.7/dist-packages/pandas/core/frame.py", line 1976, in _getitem_column
return self._get_item_cache(key)
File "/usr/local/lib/python2.7/dist-packages/pandas/core/generic.py", line 1091, in _get_item_cache
values = self._data.get(item)
File "/usr/local/lib/python2.7/dist-packages/pandas/core/internals.py", line 3211, in get
loc = self.items.get_loc(item)
File "/usr/local/lib/python2.7/dist-packages/pandas/core/index.py", line 1759, in get_loc
return self._engine.get_loc(key)
File "pandas/index.pyx", line 137, in pandas.index.IndexEngine.get_loc (pandas/index.c:3979)
File "pandas/index.pyx", line 157, in pandas.index.IndexEngine.get_loc (pandas/index.c:3843)
File "pandas/hashtable.pyx", line 668, in pandas.hashtable.PyObjectHashTable.get_item (pandas/hashtable.c:12265)
File "pandas/hashtable.pyx", line 676, in pandas.hashtable.PyObjectHashTable.get_item (pandas/hashtable.c:12216)
KeyError: '2008-01-01'
You end up with a traceback with KeyError
您最终会使用 KeyError 进行回溯
However, if you reverse it, like this:
但是，如果您将其反转，如下所示：
>>> b = a[::-1]
Then try the same index, you get the proper result:
然后尝试相同的索引，你会得到正确的结果：
>>> b['2008-01-01']
PETR4 CSNA3 VALE5
Date
2008-01-01 0 0 0
I do NOTknow why this is the case. Chances are, it has something to do with being a time series one way, but not the other? Someone more knowledgeable should answer that.
我不知道为什么会这样。有可能，它以一种方式与时间序列有关，但与另一种方式无关？应该有更懂行的人来回答。
Update: By RTFM, I discovered this page:
更新：通过 RTFM，我发现了这个页面：
https://pandas.pydata.org/pandas-docs/stable/timeseries.html
https://pandas.pydata.org/pandas-docs/stable/timeseries.html
If you find the section titled "Slice vs. Exact Match", there is a warning that explains this behavior. The problem seems to be that for a TimeSeries, an exact match is interpreted as a column name. For unsorted dataframes, this doesn't happen. See the warning box in the section referenced above. I still find this terribly confusing, but there you go...
如果您找到标题为“切片与精确匹配”的部分，则有一条警告解释了这种行为。问题似乎是对于 TimeSeries，精确匹配被解释为列名。对于未排序的数据帧，这不会发生。请参阅上述部分中的警告框。我仍然觉得这非常令人困惑，但是你去...
Edit: Changed the printout of b, which was wrong in the original.
编辑：修改了b的打印输出，原来是错误的。
Edit1: Update with explanation in python documentation.
编辑 1：在 python 文档中更新解释。
