CODEWITHSUNDEEP
numpy
ZEWEI CHU
ZEWEI CHU

Reputation: 447

how to gather elements of specific indices in numpy?

I want to gather elements of specified indices in specified axis like following. 

x = [[1,2,3], [4,5,6]]
index = [[2,1], [0, 1]]
x[:, index] = [[3, 2], [4, 5]]

This is essentially gather operation in pytorch, but as you know, this is not achievable in numpy this way. I am wondering if there is such a "gather" operation in numpy?

Upvotes: 7

Views: 13745

Answers (4)

banma
banma

Reputation: 141

numpy.take_along_axis is what I need, take elements according to the index. It can be used like gather method in PyTorch.

This is an example from the manual:

>>> a = np.array([[10, 30, 20], [60, 40, 50]])
>>> ai = np.expand_dims(np.argmax(a, axis=1), axis=1)
>>> ai
array([[1],
       [0]])
>>> np.take_along_axis(a, ai, axis=1)
array([[30],
       [60]])

Upvotes: 11

Leonardo Barazza
Leonardo Barazza

Reputation: 395

Use numpy.take() function which has most of PyTorch's gather function functionalities.

Upvotes: 3

Sia Rezaei
Sia Rezaei

Reputation: 437

I wrote this awhile ago to replicate PyTorch's gather in Numpy. In this case self is your x

def gather(self, dim, index):
    """
    Gathers values along an axis specified by ``dim``.

    For a 3-D tensor the output is specified by:
        out[i][j][k] = input[index[i][j][k]][j][k]  # if dim == 0
        out[i][j][k] = input[i][index[i][j][k]][k]  # if dim == 1
        out[i][j][k] = input[i][j][index[i][j][k]]  # if dim == 2

    Parameters
    ----------
    dim:
        The axis along which to index
    index:
        A tensor of indices of elements to gather

    Returns
    -------
    Output Tensor
    """
    idx_xsection_shape = index.shape[:dim] + \
        index.shape[dim + 1:]
    self_xsection_shape = self.shape[:dim] + self.shape[dim + 1:]
    if idx_xsection_shape != self_xsection_shape:
        raise ValueError("Except for dimension " + str(dim) +
                         ", all dimensions of index and self should be the same size")
    if index.dtype != np.dtype('int_'):
        raise TypeError("The values of index must be integers")
    data_swaped = np.swapaxes(self, 0, dim)
    index_swaped = np.swapaxes(index, 0, dim)
    gathered = np.choose(index_swaped, data_swaped)
    return np.swapaxes(gathered, 0, dim)

These are the test cases:

# Test 1
    t = np.array([[65, 17], [14, 25], [76, 22]])
    idx = np.array([[0], [1], [0]])
    dim = 1
    result = gather(t, dim=dim, index=idx)
    expected = np.array([[65], [25], [76]])
    print(np.array_equal(result, expected))

# Test 2
    t = np.array([[47, 74, 44], [56, 9, 37]])
    idx = np.array([[0, 0, 1], [1, 1, 0], [0, 1, 0]])
    dim = 0
    result = gather(t, dim=dim, index=idx)
    expected = np.array([[47, 74, 37], [56, 9, 44.], [47, 9, 44]])
    print(np.array_equal(result, expected))

Upvotes: 6

Sam17
Sam17

Reputation: 91

>>> x = np.array([[1,2,3], [4,5,6]])
>>> index = np.array([[2,1], [0, 1]])
>>> x_axis_index=np.tile(np.arange(len(x)), (index.shape[1],1)).transpose() 
>>> print x_axis_index
[[0 0]
 [1 1]]
>>> print x[x_axis_index,index]
[[3 2]
 [4 5]]

Upvotes: 2

Related Questions