mars.tensor.
tile
Construct a tensor by repeating A the number of times given by reps.
If reps has length d, the result will have dimension of max(d, A.ndim).
d
max(d, A.ndim)
If A.ndim < d, A is promoted to be d-dimensional by prepending new axes. So a shape (3,) array is promoted to (1, 3) for 2-D replication, or shape (1, 1, 3) for 3-D replication. If this is not the desired behavior, promote A to d-dimensions manually before calling this function.
A.ndim < d
If A.ndim > d, reps is promoted to A.ndim by pre-pending 1’s to it. Thus for an A of shape (2, 3, 4, 5), a reps of (2, 2) is treated as (1, 1, 2, 2).
A.ndim > d
Note : Although tile may be used for broadcasting, it is strongly recommended to use Mars’ broadcasting operations and functions.
A (array_like) – The input tensor.
reps (array_like) – The number of repetitions of A along each axis.
c – The tiled output tensor.
Tensor
See also
repeat
Repeat elements of a tensor.
broadcast_to
Broadcast a tensor to a new shape
Examples
>>> import mars.tensor as mt
>>> a = mt.array([0, 1, 2]) >>> mt.tile(a, 2).execute() array([0, 1, 2, 0, 1, 2]) >>> mt.tile(a, (2, 2)).execute() array([[0, 1, 2, 0, 1, 2], [0, 1, 2, 0, 1, 2]]) >>> mt.tile(a, (2, 1, 2)).execute() array([[[0, 1, 2, 0, 1, 2]], [[0, 1, 2, 0, 1, 2]]])
>>> b = mt.array([[1, 2], [3, 4]]) >>> mt.tile(b, 2).execute() array([[1, 2, 1, 2], [3, 4, 3, 4]]) >>> mt.tile(b, (2, 1)).execute() array([[1, 2], [3, 4], [1, 2], [3, 4]])
>>> c = mt.array([1,2,3,4]) >>> mt.tile(c,(4,1)).execute() array([[1, 2, 3, 4], [1, 2, 3, 4], [1, 2, 3, 4], [1, 2, 3, 4]])