如何将所有dataframe列合并为一列? [英] How to concatenate all dataframe column into one column?

问题描述

我有一个大致像这样的数据框:

I have a dataframe that looks roughly like:

           2020-01-01   2020-01-02   2020-01-03   2020-01-05
00:00:00     11             47          54           10 
01:00:00     12             49          46           22
...
23:00:00     15             34          22           40

预期输出...

2020-01-01 00:00:00   11
2020-01-01 01:00:00   12
...
2020-01-01 23:00:00   12
2020-01-02 00:00:00   47
2020-01-02 01:00:00   49
...
2020-01-01 23:00:00   34
...

推荐答案

使用

Use DataFrame.melt with convert index to column, then convert columns to datetimes and timedeltas and join together with DataFrame.pop for extract columns:

df = df.reset_index().melt('index')
df.index = pd.to_datetime(df.pop('variable')) + pd.to_timedelta(df.pop('index'))

#alternative with join strings
#df.index = pd.to_datetime(df.pop('variable') + ' ' + df.pop('index'))
print (df)
                     value
2020-01-01 00:00:00     11
2020-01-01 01:00:00     12
2020-01-01 23:00:00     15
2020-01-02 00:00:00     47
2020-01-02 01:00:00     49
2020-01-02 23:00:00     34
2020-01-03 00:00:00     54
2020-01-03 01:00:00     46
2020-01-03 23:00:00     22
2020-01-05 00:00:00     10
2020-01-05 01:00:00     22
2020-01-05 23:00:00     40

如果未使用pop是必要的，请通过 DataFrame.drop :

If pop is not used is necessary remove columns by DataFrame.drop:

df = df.reset_index().melt('index')
df.index = pd.to_datetime(df['variable']) + pd.to_timedelta(df['index'])
df = df.drop(['index','variable'], axis=1)

使用 DataFrame.unstack 并将MultiIndex与map结合在一起并转换为DatetimeIndex，输出为Series:

Another idea with DataFrame.unstack and join MultiIndex with map and convert to DatetimeIndex, output is Series:

s = df.unstack()
s.index = pd.to_datetime(s.index.map('{0[0]} {0[1]}'.format))
print (s)
2020-01-01 00:00:00    11
2020-01-01 01:00:00    12
2020-01-01 23:00:00    15
2020-01-02 00:00:00    47
2020-01-02 01:00:00    49
2020-01-02 23:00:00    34
2020-01-03 00:00:00    54
2020-01-03 01:00:00    46
2020-01-03 23:00:00    22
2020-01-05 00:00:00    10
2020-01-05 01:00:00    22
2020-01-05 23:00:00    40
dtype: int64

这篇关于如何将所有dataframe列合并为一列?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！