Difficulty in list comprehension while a nested for loop could be nasty-CodePudding

I need to get (-2.0EMEXW_1_M_Xx1/F - 2.0ENEXW_1_N_Xx1/F) in my code. Everything is working well using the nested for loop while if I want to get result using the list comprehension then it(aasum) gives duplicate elements.

import sympy as sp
anions = {'X': 1 }
cations = {'M': 1, 'N': 1}
neutrals = {'1' : 0}
aa = list(-(1 / sp.symbols(f'F')) * sp.symbols(f'E{m}') * sp.symbols(f'E{n}') * \
                            ((x   y) / (x * y)) * sp.symbols(f'x{str(list(neutrals.keys())[0])}') *       sp.symbols(f'W_{str(list(neutrals.keys())[0])}_{str(m)}_{str(n)}') \
                            for m in cations.keys() \
                            for x in cations.values() \
                            for n in anions.keys() \
                            for y in anions.values())

aasum = sum(aa)
asum = 0.0
for k in anions:
    for j in cations:
        for n in neutrals:
            asum  = -(1 / sp.symbols(f'F')) * sp.symbols(f'E{str(j)}') * sp.symbols(f'E{str(k)}') * \
            ((cations[j]   anions[k]) / (cations[j] * anions[k])) * sp.symbols(f'x{str(list(neutrals.keys())[0])}') *\
                 sp.symbols(f'W_{str(list(neutrals.keys())[0])}_{str(j)}_{str(k)}')

print(asum,'...', aasum)

result is: asum:-2.0EMEXW_1_M_Xx1/F - 2.0ENEXW_1_N_Xx1/F ... aasum:-4.0EMEXW_1_M_Xx1/F - 4.0ENEXW_1_N_Xx1/F I checked: [https://stackoverflow.com/questions/64091392/nested-list-in-list-comprehension-to-for-loop] But it has nothing to do answering my question. Any suggestion to get correct result using list comprehension would be welcomed.

CodePudding user response：

Use items instead of keys and values:

import sympy as sp
anions = {'X': 1 }
cations = {'M': 1, 'N': 1}
neutrals = {'1' : 0}
aa = [-(1 / sp.symbols(f'F')) * sp.symbols(f'E{m}') * sp.symbols(f'E{n}') *
      ((x   y) / (x * y)) * sp.symbols(f'x{str(list(neutrals.keys())[0])}') * sp.symbols(f'W_{str(list(neutrals.keys())[0])}_{str(m)}_{str(n)}')
          for (m, x) in cations.items()
          for (n, y) in anions.items()]

Output:

>>> aa
[-2.0*EM*EX*W_1_M_X*x1/F, -2.0*EN*EX*W_1_N_X*x1/F]

>>> sum(aa)
-2.0*EM*EX*W_1_M_X*x1/F - 2.0*EN*EX*W_1_N_X*x1/F

Explanation:

In your loop, you iterate over keys only so you get the cartesian product of anions x cations. In your comprehension, you iterate both on keys ans values so you have the product of cations.values x cations.keys, x anions.values x anions.keys. What I did was replicate what you did in your loop except I get the keys and values at the same time. I have the product of (anions.keys, anion.values) x (cations.keys, cations.values). It avoid j and k indexes.