tile的用法

tile函数在numpy库中

python文档如下：

tile(A, reps)
    Construct an array by repeating A the number of times given by reps.

通过由reps给出的参数，重复A相应的次数，构造一个新的数组
    
    If `reps` has length ``d``, the result will have dimension of
    ``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.
    
    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).
    
    Note : Although tile may be used for broadcasting, it is strongly
    recommended to use numpy's broadcasting operations and functions.
    
    Parameters
    ----------
    A : array_like
        The input array.
    reps : array_like
        The number of repetitions of `A` along each axis.
    
    Returns
    -------
    c : ndarray
        The tiled output array.
    
    See Also
    --------
    repeat : Repeat elements of an array.
    broadcast_to : Broadcast an array to a new shape
    
    Examples
    --------
    >>> a = np.array([0, 1, 2])
    >>> np.tile(a, 2)
    array([0, 1, 2, 0, 1, 2])
    >>> np.tile(a, (2, 2))
    array([[0, 1, 2, 0, 1, 2],
           [0, 1, 2, 0, 1, 2]])
    >>> np.tile(a, (2, 1, 2))
    array([[[0, 1, 2, 0, 1, 2]],
           [[0, 1, 2, 0, 1, 2]]])
    
    >>> b = np.array([[1, 2], [3, 4]])
    >>> np.tile(b, 2)
    array([[1, 2, 1, 2],
           [3, 4, 3, 4]])
    >>> np.tile(b, (2, 1))
    array([[1, 2],
           [3, 4],
           [1, 2],
           [3, 4]])
    
    >>> c = np.array([1,2,3,4])
    >>> np.tile(c,(4,1))
    array([[1, 2, 3, 4],
           [1, 2, 3, 4],
           [1, 2, 3, 4],
           [1, 2, 3, 4]])

自己添加几个示例：

>>> tile([1,2],(4,1))
array([[1, 2],
       [1, 2],
       [1, 2],
       [1, 2]])
>>> tile([1,2],(4,2))
array([[1, 2, 1, 2],
       [1, 2, 1, 2],
       [1, 2, 1, 2],
       [1, 2, 1, 2]])

>>> tile([[1,2],[3,4]],(4,2))
array([[1, 2, 1, 2],
       [3, 4, 3, 4],
       [1, 2, 1, 2],
       [3, 4, 3, 4],
       [1, 2, 1, 2],
       [3, 4, 3, 4],
       [1, 2, 1, 2],
       [3, 4, 3, 4]])

>>> tile([[1,2],[3,4]],(4,1))
array([[1, 2],
       [3, 4],
       [1, 2],
       [3, 4],
       [1, 2],
       [3, 4],
       [1, 2],
       [3, 4]])