I am new to Python and trying to store results from a for-loop into dataframe columns (Python 3). Let's say I have the following data:

time=[1,2,3,4]
i=[1,2,3,4,5]
j=[10,20,30,40,50]
k=[100,200,300,400,500]
data_ijk=list(zip(i,j,k))

[(1, 10, 100), (2, 20, 200), (3, 30, 300), (4, 40, 400), (5, 50, 500)]

I want to loop through the data_ijk to solve the following equation:

eq=(i j k)*t

and put the results of equation into dataframe with column labels col_labels=["A","B","C","D","E"] for each set of ijk and rows considering time (t).

The expected output should look something like this:

I know i need to define dataframe

df = pd.DataFrame(columns=[col_labels])

and I assume I should create nested for loop with data_abc and time, but I have no idea how to define to put results in different columns

for i,j,k in data_abc:
  for t in time:
  ...???
   print((i j k)*t)

CodePudding user response：

Try:

df = pd.DataFrame(data=[[sum(x)*t for x in zip(i,j,k)] for t in time], 
                  columns=list("ABCDE"), 
                  index=time)

>>> df
     A    B     C     D     E
1  111  222   333   444   555
2  222  444   666   888  1110
3  333  666   999  1332  1665
4  444  888  1332  1776  2220

CodePudding user response：

Try:

import pandas as pd

# In case, the column names are actually words than single letters
col_labels=["A","B","C","D","E"]

time=[1,2,3,4]
data = dict(i=[1,2,3,4,5],
            j=[10,20,30,40,50],
            k=[100,200,300,400,500])

df = pd.DataFrame(data)
cons_sum_ijk = df.apply(sum, axis=1).to_list()


df = pd.DataFrame(columns=col_labels, data=[cons_sum_ijk]* len(time) )
df['t'] = time

df[col_labels] = df.apply(lambda x: x[col_labels]*x['t'], axis=1)
df = df[['t']   col_labels]

>>> df
    t     A    B     C     D     E
0   1    111  222   333   444   555
1   2    222  444   666   888  1110
2   3    333  666   999  1332  1665
3   4    444  888  1332  1776  2220

CodePudding user response：

Using Numpy's einsum

import numpy as np
import pandas as pd

time = [1, 2, 3, 4]
i = [1, 2, 3, 4, 5]
j = [10, 20, 30, 40, 50]
k = [100, 200, 300, 400, 500]

df = pd.DataFrame(np.einsum('a,bc->ac', time, np.array([i, j, k])), columns=[*'ABCDE'])

df

     A    B     C     D     E
0  111  222   333   444   555
1  222  444   666   888  1110
2  333  666   999  1332  1665
3  444  888  1332  1776  2220