How to build an array of all combinations of two NumPy arrays?

Sometimes we need to find the combination of elements of two or more arrays. Numpy has a function to compute the combination of 2 or more Numpy arrays named as “numpy.meshgrid()“. This function is used to create a rectangular grid out of two given one-dimensional arrays representing the Cartesian indexing or Matrix indexing.

Syntax:

numpy.meshgrid(*xi, copy=True, sparse=False, indexing='xy')

Example 1: Computing combinations of elements of Two NumPy arrays

Python3

# importing Numpy package
import numpy as np

# creating 2 numpy arrays
array_1 = np.array([1, 2])
array_2 = np.array([4, 6])

print("Array-1")
print(array_1)

print("\nArray-2")
print(array_2)

# combination of elements of array_1 and array_2
# using numpy.meshgrid().T.reshape()
comb_array = np.array(np.meshgrid(array_1, array_2)).T.reshape(-1, 2)

print("\nCombine array:")
print(comb_array)

Output:

In the above example, we combine elements of ‘array_1‘ and ‘array_2‘ using numpy.meshgrid().T.reshape() 

Example 2: Computing combinations of elements of Three NumPy arrays

Python3

# importing Numpy package
import numpy as np

# creating 3 numpy arrays
array_1 = np.array([1, 2, 3])
array_2 = np.array([4, 6, 4])
array_3 = np.array([3, 6])

print("Array-1")
print(array_1)

print("Array-2")
print(array_2)

print("Array-3")
print(array_3)


# combination of elements of array_1,
# array_2 and array_3 using
# numpy.meshgrid().T.reshape()
comb_array = np.array(
  np.meshgrid(array_1, array_2, array_3)).T.reshape(-1, 3)

print("\nCombine array:")
print(comb_array)

Output:

In the above example, we combine elements of ‘array_1‘, ‘array_2‘ and ‘array_3‘ using numpy.meshgrid().T.reshape()

Example 3: Computing combinations of elements of Four NumPy arrays

Python3

# importing Numpy package
import numpy as np

# creating 4 numpy arrays
array_1 = np.array([50, 21])
array_2 = np.array([4, 4])
array_3 = np.array([1, 10])
array_4 = np.array([7, 14])


print("Array-1")
print(array_1)

print("Array-2")
print(array_2)

print("Array-3")
print(array_3)

print("Array-4")
print(array_4)


# combination of elements of array_1,
# array_2, array_3 and array_4
# using numpy.meshgrid().T.reshape()
comb_array = np.array(np.meshgrid(
    array_1, array_2, array_3, array_4)).T.reshape(-1, 4)

print("\nCombine array:")
print(comb_array)

Output:

In the above example, we combine elements of ‘array_1‘, ‘array_2‘, ‘array_3‘ and ‘array_4‘ using numpy.meshgrid().T.reshape()