-1

I am trying to create and use a dynamic variable in Python for the first time.

for i in range(0,len(test_data)):
     globals()["test_list_{}".format(test_data[i])]=[]
     globals()["test_calculation_{}".format(test_data[i])]=0

First, I created test_list_number and test_calculation_number as global variables.

Then I want to use this in for and use it for calculations. The code I wrote here was made by simplifying the code I'm going to use.

How do I change the numbers in the two for statements below?

enter image description here

1

--------------Below is a code example. -------------

import pandas as pd
import numpy as np

X=list(np.random.random(100)*100)
Y=list(np.random.random(100)*100)

test_data= [2,5,7,8]

test_dict={(i,j):np.hypot(X[i]-X[j],Y[i]-Y[j]) for i in range(0,100) for j in range(0,100)}



test_df_data2={
    'index' : [1,2,3],
    'data1' : [3,5,6],
    'data2' : [2,5,6]
}



test_df_data5={
    'index' : [1,2,3],
    'data1' : [8,3,1],
    'data2' : [3,2,7]
}

test_df_2  =pd.DataFrame(test_df_data2)
test_df_5  =pd.DataFrame(test_df_data5)


for i in range(0,len(test_data)):
    globals()["test_list_{}".format(test_data[i])]=[]
    globals()["test_calculation_{}".format(test_data[i])]=0

    
    
for i in range(0, len(test_df_2     )  ):
    test_list_2       .append((test_df_2   .data1[i],test_df_2     .data2      [i]))

for i in range(len(test_list_2    )):
    test_calculation_2    = test_calculation_2     + test_dict[test_list_2       [i] ]

    
print( test_calculation_2)
Soo
  • 35
  • 4
  • Use a dictionary with the "dynamic variable" names as your keys. – Craig Sep 02 '21 at 04:07
  • Does this answer your question? [How do I create variable variables?](https://stackoverflow.com/questions/1373164/how-do-i-create-variable-variables) – Craig Sep 02 '21 at 04:08
  • Have found anything in Python, numpy or pandas docs that talks about "dynamic variable". I don't recall anything with that name, though I can guess what you are trying to do. You are better off using documented, accepted practices, as opposed to importing some concept from another language. – hpaulj Sep 02 '21 at 07:01

1 Answers1

2

Short answer, do not do this!

It is widely accepted to be a bad practice (see for example). You have a high risk of doing something difficult to debug, to overwrite existing variables, or to use the wrong namespace. In addition there are eventual pitfalls and lack of robust methods to do this.

Use a container instead. You will have a shorter, cleaner, and easier to debug code.

Dictionaries are ideal for this:

# initialize
my_dfs = {}

# computation
for i in …:
    my_dfs[f'df_computation_{i}'] = …

# access
my_dfs['df_computation_42']
mozway
  • 194,879
  • 13
  • 39
  • 75
  • I used the wrong method because I didn't study enough. thank you for telling me. I'll do ask before asking another question. – Soo Sep 02 '21 at 08:37
  • @Soo don't feel bad, many want to do this. It feels like a attractive/natural way to handle multiple datasets. I think it's mostly with experience that your acquire enough abstraction to realize this is not ideal/needed. – mozway Sep 02 '21 at 08:40