mars.tensor.atleast_3d¶

mars.tensor.
atleast_3d
(*tensors)[source]¶ View inputs as tensors with at least three dimensions.
 Parameters
tensors1 (array_like) – One or more tensorlike sequences. Nontensor inputs are converted to tensors. Tensors that already have three or more dimensions are preserved.
tensors2 (array_like) – One or more tensorlike sequences. Nontensor inputs are converted to tensors. Tensors that already have three or more dimensions are preserved.
.. (array_like) – One or more tensorlike sequences. Nontensor inputs are converted to tensors. Tensors that already have three or more dimensions are preserved.
 Returns
res1, res2, … – A tensor, or list of tensors, each with
a.ndim >= 3
. Copies are avoided where possible, and views with three or more dimensions are returned. For example, a 1D tensor of shape(N,)
becomes a view of shape(1, N, 1)
, and a 2D tensor of shape(M, N)
becomes a view of shape(M, N, 1)
. Return type
Tensor
See also
Examples
>>> import mars.tensor as mt
>>> mt.atleast_3d(3.0).execute() array([[[ 3.]]])
>>> x = mt.arange(3.0) >>> mt.atleast_3d(x).shape (1, 3, 1)
>>> x = mt.arange(12.0).reshape(4,3) >>> mt.atleast_3d(x).shape (4, 3, 1)
>>> for arr in mt.atleast_3d([1, 2], [[1, 2]], [[[1, 2]]]).execute(): ... print(arr, arr.shape) ... [[[1] [2]]] (1, 2, 1) [[[1] [2]]] (1, 2, 1) [[[1 2]]] (1, 1, 2)