dpnp.tensor.top_k

dpnp.tensor.top_k(x, k, /, *, axis=None, mode='largest')

Returns the k largest or smallest values and their indices in the input array x along the specified axis axis.

Parameters:

• x (usm_ndarray) -- Input array.

• k (int) -- Number of elements to find. Must be a positive integer value.

• axis ({None, int}, optional) --

  Axis along which to search. If None, the search will be performed over the flattened array.

  Default: None.

• mode ({"largest", "smallest"}, optional) --

  Search mode. Must be one of the following modes:

  • "largest": return the k largest elements.

  • "smallest": return the k smallest elements.

  Default: "largest".

Returns:

out -- A namedtuple (values, indices) whose

• first element values will be an array containing the k largest or smallest elements of x. The array has the same data type as x. If axis was None, values will be a one-dimensional array with shape (k,) and otherwise, values will have shape x.shape[:axis] + (k,) + x.shape[axis+1:]

• second element indices will be an array containing indices of x that result in values. The array will have the same shape as values and will have the default array index data type.

Return type:

tuple of usm_ndarrays