numpy.true_divide() en Python

(arr1, arr22, /, out=None, *, where=True, casting=’same_kind’, order=’K’, dtype=None, ufunc ‘true_divide’) :
el elemento de array de la primera array se divide por los elementos de segunda array (todo sucede por elementos). Tanto arr1 como arr2 deben tener la misma forma. Devuelve la división verdadera por elementos.

Python sigue tradicionalmente la ‘división de piso’. Independientemente del tipo de entrada, la división real ajusta la respuesta de la mejor manera.
“//” es operador de división de piso.
“/” es el verdadero operador de división.

Parámetros:

arr1     : [array_like]Input array or object which works as numerator.
arr2     : [array_like]Input array or object which works as denominator. 
out      : [ndarray, None, optional]Output array with same dimensions as Input array, 
           placed with result.
**kwargs : allows you to pass keyword variable length of argument to a function. 
           It is used when we want to handle named argument in a function.
where    : [array_like, optional]True value means to calculate the universal 
           functions(ufunc) at that position, False value means to leave the  
           value in the output alone.

Devolver :

If inputs are scalar then scalar; otherwise array with arr1 / arr2(element- wise) 
i.e. true division

 
Código 1: arr1 dividido por arr2

# Python program explaining
# true_divide() function
import numpy as np
  
# input_array
arr1 = [6, 7, 2, 9, 1]
arr2 = [2, 3, 4, 5, 6]
print ("arr1         : ", arr1)
print ("arr1         : ", arr2)
  
# output_array
out = np.true_divide(arr1, arr2)
print ("\nOutput array : \n", out)

Producción :

arr1         :  [6, 7, 2, 9, 1]
arr1         :  [2, 3, 4, 5, 6]

Output array : 
 [ 3.          2.33333333  0.5         1.8         0.16666667]

 
Código 2: elementos de arr1 divididos por divisor

# Python program explaining
# true_divide() function
import numpy as np
  
# input_array
arr1 = [2, 7, 3, 11, 4]
divisor = 3
print ("arr1         : ", arr1)
  
# output_array
out = np.true_divide(arr1, divisor)
print ("\nOutput array : ", out)

Producción :

arr1         :  [2, 7, 3, 11, 4]

Output array :  [ 0.66666667  2.33333333  1.          3.66666667  1.33333333]

 
Código 3: Comparación entre floor_division(//) y true-division(/)

# Python program explaining
# true_divide() function
import numpy as np
  
# input_array
arr1 = np.arange(5)
arr2 = [2, 3, 4, 5, 6]
print ("arr1         : ", arr1)
print ("arr1         : ", arr2)
  
# output_array
out = np.floor_divide(arr1, arr2)
out_arr = np.true_divide(arr1, arr2) 
print ("\nOutput array with floor divide : \n", out)
print ("\nOutput array with true divide  : \n", out_arr)
  
  
print ("\nOutput array with floor divide(//) : \n", arr1//arr2)
print ("\nOutput array with true divide(/)   : \n", arr1/arr2)

Producción :

arr1         :  [0 1 2 3 4]
arr1         :  [2, 3, 4, 5, 6]

Output array with floor divide : 
 [0 0 0 0 0]

Output array with true divide  : 
 [ 0.          0.33333333  0.5         0.6         0.66666667]

Output array with floor divide(//) : 
 [0 0 0 0 0]

Output array with true divide(/)   : 
 [ 0.          0.33333333  0.5         0.6         0.66666667]

Referencias:
https://docs.scipy.org/doc/numpy-1.13.0/reference/generated/numpy.floor_divide.html
.