Making a function more efficient

[CatorCanulis]

Hi everyone,

I'm working on a function that avoids transposed numbers. Like if you have 12 you can't have 21 or if you have 126 you can't have 216 oder 621.

What I got so far:

#!/bin/env python3

import itertools
import itertools

from numba import jit, cuda
import numpy as np

#@jit(target_backend='cuda')
def getNonSwappables(iStart, iStop, verbose = False):

  idsClean = []

  ids = list(range(iStart, iStop))

  for i in ids:

    if verbose:
      print("Processing: ", i)
    # Create permutation for current string.
    sPermutations = [''.join(p) for p in itertools.permutations(str(i))]

    # Check if the clean ids list contains any of the permutations.
    if not any(item in str(idsClean) for item in sPermutations):

      # If not so, apped the current integer to the list of clean ids.
      idsClean.append(i)

  return(idsClean)

listIDsClean = getNonSwappables(1, 100000, True)
print(listIDsClean)
print(len(listIDsClean))

For numbers < 10.000 it works somewhat reasonable. But larger numbers take forever. Is there a way I can code this more efficient?

As you can see, I already tried to crunch it with my gpu but compilation failed. That's not really a sustainable option anyway.

Any help is greatly appreciated.

Markus