[ Python 常見問題 ] Shuffling Several DataFrames Together

 Source From Here

Question
Is it possible to shuffle several DataFrames together?

For example I have a DataFrame df1 and a DataFrame df2. I want to shuffle the rows randomly, but for both DataFrames in the same way.

Example

view plaincopy to clipboardprint?

1. import pandas as pd  
2.   
3. df1 = pd.DataFrame(data=[[1,2,3],[4,5,6], [7,8,9]], columns=['f1', 'f2', 'f3'])  
4. df2 = pd.DataFrame({'label':[0,1,0]})  

HowTo
I think you can double reindex with applying numpy.random.permutation to index, but it is necessary both DataFrames have same length and same unique index values:

view plaincopy to clipboardprint?

1. import numpy as np  
2.   
3. new_idx = np.random.permutation(df1.index)  
4. df1 = df1.reindex(new_idx)  
5. df2 = df2.reindex(new_idx)  

Alternative with reindex_axis:

view plaincopy to clipboardprint?

1. print (df1.reindex_axis(idx, axis=0))  
2. print (df2.reindex_axis(idx, axis=0))