create matrix structure using pandas

Question

I have loaded the below CSV file containing code and coefficient data into the below dataframe df:

CODE|COEFFICIENT  
A|0.5  
B|0.4  
C|0.3

import pandas as pd
import numpy as np
df= pd.read_csv('cod_coeff.csv', delimiter='|', encoding="utf-8-sig")

giving

  ITEM   COEFFICIENT  
0    A       0.5  
1    B       0.4  
2    C       0.3  

From the above dataframe, I need to create a final dataframe as below which has a matrix structure with the product of the coefficients:

     A         B         C        
A   0.25      0.2        0.15  
B   0.2       0.16       0.12  
C   0.15      0.12       0.09

I am using np.multiply but I am not successful in producing the result.

Solution 1

numpy as a faster alternative

pd.DataFrame(np.outer(df, df), df.index, df.index)

---

Timing

Given sample

30,000 rows

df = pd.concat([df for _ in range(10000)], ignore_index=True)

Solution 2

You want to do the math between a vector and its tranposition. Transpose with .T and apply the matrix dot function between the two dataframes.

df = df.set_index('CODE')

df.T
Out[10]: 
CODE             A    B    C
COEFFICIENT    0.5  0.4  0.3

df.dot(df.T)
Out[11]: 
CODE     A     B     C
CODE                  
A     0.25  0.20  0.15
B     0.20  0.16  0.12
C     0.15  0.12  0.09