numpy equivalent to list.index

Question

In Python, I can use list.index to find the index of an element, e.g. [1,2,3,4].index(1) returns 0, since the element 1 occurs at position 0. I'd like to do this with multi-dimensional arrays. That is, I'd like to be able to write something like

x = np.array([[0, 1, 2],
              [3, 4, 5]])
y = np.array([0, 1, 2])
print(x.index(y))

The output should be 0, because again the target element is in the first position.

Using the built-in list.index method doesn't work because the == operator for numpy arrays returns an array rather than a boolean. And np.argwhere doesn't do any broadcasting. I know I could always do

idx = 0
while idx < x.shape[0] and (x[idx] != y).any():
  idx += 1
print(idx)

but that feels so clumsy!

Answer 1

There are several ways. Following this example (Find matching rows in 2 dimensional numpy array):

>>> np.where((x == y).all(axis=1))
(array([0]),)