mars.tensor.take(a, indices, axis=None, out=None)[source]

Take elements from a tensor along an axis.

When axis is not None, this function does the same thing as “fancy” indexing (indexing arrays using tensors); however, it can be easier to use if you need elements along a given axis. A call such as mt.take(arr, indices, axis=3) is equivalent to arr[:,:,:,indices,...].

Explained without fancy indexing, this is equivalent to the following use of ndindex, which sets each of ii, jj, and kk to a tuple of indices:

Ni, Nk = a.shape[:axis], a.shape[axis+1:]
Nj = indices.shape
for ii in ndindex(Ni):
    for jj in ndindex(Nj):
        for kk in ndindex(Nk):
            out[ii + jj + kk] = a[ii + (indices[jj],) + kk]
  • a (array_like (Ni..., M, Nk...)) – The source tensor.

  • indices (array_like (Nj...)) –

    The indices of the values to extract.

    Also allow scalars for indices.

  • axis (int, optional) – The axis over which to select values. By default, the flattened input tensor is used.

  • out (Tensor, optional (Ni..., Nj..., Nk...)) – If provided, the result will be placed in this tensor. It should be of the appropriate shape and dtype.

  • mode ({'raise', 'wrap', 'clip'}, optional) –

    Specifies how out-of-bounds indices will behave.

    • ’raise’ – raise an error (default)

    • ’wrap’ – wrap around

    • ’clip’ – clip to the range

    ’clip’ mode means that all indices that are too large are replaced by the index that addresses the last element along that axis. Note that this disables indexing with negative numbers.


out – The returned tensor has the same type as a.

Return type

Tensor (Ni…, Nj…, Nk…)

See also


Take elements using a boolean mask


equivalent method


By eliminating the inner loop in the description above, and using s_ to build simple slice objects, take can be expressed in terms of applying fancy indexing to each 1-d slice:

Ni, Nk = a.shape[:axis], a.shape[axis+1:]
for ii in ndindex(Ni):
    for kk in ndindex(Nj):
        out[ii + s_[...,] + kk] = a[ii + s_[:,] + kk][indices]

For this reason, it is equivalent to (but faster than) the following use of apply_along_axis:

out = mt.apply_along_axis(lambda a_1d: a_1d[indices], axis, a)


>>> import mars.tensor as mt
>>> a = [4, 3, 5, 7, 6, 8]
>>> indices = [0, 1, 4]
>>> mt.take(a, indices).execute()
array([4, 3, 6])

In this example if a is a tensor, “fancy” indexing can be used.

>>> a = mt.array(a)
>>> a[indices].execute()
array([4, 3, 6])

If indices is not one dimensional, the output also has these dimensions.

>>> mt.take(a, [[0, 1], [2, 3]]).execute()
array([[4, 3],
       [5, 7]])