熊猫:在dataframes之间复制一些值

I have two dataframes, called Old and New. Old has 96 rows, and New has 48 rows. I want to take one column of Old, say ['Values'] and split it into two columns in New, say ['First'] and ['Second']. Thus, for a simple example with 6 rows to start; from:

我有两个dataframes，分别叫做Old和New。Old有96行，New有48行。我想取一列旧的，写上['Values']，然后把它分成两个新的列，比如['First']和['Second']。因此，对于一个开始有6行的简单示例;来自:

来

    First  Second
1     10     40
2     20     50
3     30     60

I have a notion that this should be trivially easy, and yet I can't do it because the indices need to be changed. I simply want to copy values, as you see.

我有个想法，这应该很简单，但我做不到，因为指标需要改变。我只是想复制值。

How is this best done?

这是怎么做到的呢?

2 个解决方案

#1

You can use reshape:

您可以使用重塑:

pd.DataFrame(df.values.reshape(-1,2, order='F'), columns=['First','Second'])
Out[12]: 
array([[10, 40],
       [20, 50],
       [30, 60]], dtype=int64)

#2

Using split from numpy, you can split into two or other size and combine them with hstack to form new dataframe:

使用从numpy中分离，您可以分成两个或其他大小，并将它们与hstack合并，形成新的dataframe:

import numpy as np
import pandas as pd

df = pd.DataFrame({'Values': {1: 10, 2: 20, 3: 30, 4: 40, 5: 50, 6: 60}})
print(df)

Input dataframe:

输入dataframe:

Now, using split() then, using hstack():

现在，使用split()，然后使用hstack():

splits = np.split(df, 2)
result_df = pd.DataFrame(np.hstack(splits), columns=['First', 'Second'])
print(result_df)

Result:

结果:

   First  Second
0     10      40
1     20      50
2     30      60

Without using intermediate splits variable, you can try:

不使用中间分割变量，可以尝试:

result_df = pd.DataFrame(np.hstack(np.split(df, 2)), columns=['First', 'Second'])

#1