numpy中，矩阵的三种转置transpose、getT、getH的区别

由于没有时间整理，先贴出官方帮助文档，以便知道对矩阵有这三种转置操作，之间的差别以后遇到问题了再整理。

help(np.matrix.H)

Help on property:

    Returns the (complex) conjugate transpose of `self`.

    

    Equivalent to ``np.transpose(self)`` if `self` is real-valued.

    

    Parameters

    ----------

    None

    

    Returns

    -------

    ret : matrix object

        complex conjugate transpose of `self`

    

    Examples

    --------

x = np.matrix(np.arange(12).reshape((3,4)))

z = x - 1j*x; z

    matrix([[  0. +0.j,   1. -1.j,   2. -2.j,   3. -3.j],

            [  4. -4.j,   5. -5.j,   6. -6.j,   7. -7.j],

            [  8. -8.j,   9. -9.j,  10.-10.j,  11.-11.j]])

z.getH()

    matrix([[  0. +0.j,   4. +4.j,   8. +8.j],

            [  1. +1.j,   5. +5.j,   9. +9.j],

            [  2. +2.j,   6. +6.j,  10.+10.j],

            [  3. +3.j,   7. +7.j,  11.+11.j]])

help(np.matrix.T)

Help on property:

    Returns the transpose of the matrix.

    

    Does *not* conjugate!  For the complex conjugate transpose, use ``.H``.

    

    Parameters

    ----------

    None

    

    Returns

    -------

    ret : matrix object

        The (non-conjugated) transpose of the matrix.

    

    See Also

    --------

    transpose, getH

    

    Examples

    --------

m = np.matrix('[1, 2; 3, 4]')

m

    matrix([[1, 2],

            [3, 4]])

m.getT()

    matrix([[1, 3],

            [2, 4]])

help(np.matrix.transpose)

Help on method_descriptor:

transpose(...)

    a.transpose(*axes)

    

    Returns a view of the array with axes transposed.

    

    For a 1-D array, this has no effect. (To change between column and

    row vectors, first cast the 1-D array into a matrix object.)

    For a 2-D array, this is the usual matrix transpose.

    For an n-D array, if axes are given, their order indicates how the

    axes are permuted (see Examples). If axes are not provided and

    ``a.shape = (i[0], i[1], ... i[n-2], i[n-1])``, then

    ``a.transpose().shape = (i[n-1], i[n-2], ... i[1], i
4000
[0])``.

    

    Parameters

    ----------

    axes : None, tuple of ints, or `n` ints

    

     * None or no argument: reverses the order of the axes.

    

     * tuple of ints: `i` in the `j`-th place in the tuple means `a`'s

       `i`-th axis becomes `a.transpose()`'s `j`-th axis.

    

     * `n` ints: same as an n-tuple of the same ints (this form is

       intended simply as a "convenience" alternative to the tuple form)

    

    Returns

    -------

    out : ndarray

        View of `a`, with axes suitably permuted.

    

    See Also

    --------

    ndarray.T : Array property returning the array transposed.

    

    Examples

    --------

a = np.array([[1, 2], [3, 4]])

a

    array([[1, 2],

           [3, 4]])

a.transpose()

    array([[1, 3],

           [2, 4]])

a.transpose((1, 0))

    array([[1, 3],

           [2, 4]])

a.transpose(1, 0)

    array([[1, 3],

           [2, 4]])