[FIXED] How to create a new array by combining every element of the 1st with every 2nd?

Issue

Need a quick solution in Numpy to get an array in C. The dimensions and number of elements in the arrays are different.

A = [ [1, 2 , 3],
      [4, 5, 6],
         ...  ]
B = [ [a, b],
      [c, d],
      [e, f],
       ... ]

C = [ [1, 2, 3, a, b],
      [1, 2, 3, c, d],
      [1, 2, 3, e, f],
      [4, 5, 6, a, b],
      [4, 5, 6, c, d],
      [4, 5, 6, e, f],
               ...  ]

Solution

# Example arrays A and B
A = np.array([[1, 2, 3], [4, 5, 6]])
B = np.array([['a', 'b'], ['c', 'd'], ['e', 'f']])

# Repeat each row of A for the number of rows in B
repeated_A = np.repeat(A, B.shape[0], axis=0)

# Tile B for the number of times A has rows
tiled_B = np.tile(B, (A.shape[0], 1))

# Combine repeated_A and tiled_B along the second axis
C = np.hstack((repeated_A, tiled_B))

print(C)

Is it acceptable to use generator? If it is, maybe something like this will work:

rows_A, cols_A = A.shape
rows_B, cols_B = B.shape

def generate_rows():
    for i in range(rows_A):
        for j in range(rows_B):
            yield np.concatenate((A[i], B[j]))

C = np.fromiter(generate_rows(), dtype=object, count=rows_A * rows_B)

Answered By - Milos Stojanovic