@ospider 十分感谢！！就缺了这一环～

输入：Solve[2^(x - 3) == x, x]
结果：{{x -> -(ProductLog[-(Log[2]/8)]/Log[2])}, {x -> -(
ProductLog[-1, -(Log[2]/8)]/Log[2])}}

@a target="_blank" href="/member/zentorwie" rel="nofollow noopener">zentorwie 我在搜索解决方式的时候，也看到了这篇文章

@kurtrossel 笑死

@unfurl
[Warn信息]:

W: no module named matplotlib.rcParams (top-level import by matplotlib.figure)
W: no module named glib.Idle (top-level import by gobject)
W: no module named nt (conditional import by __main__)
W: no module named numpy.not_equal (top-level import by numpy.numarray.ufuncs)
W: no module named setuptools.Command (top-level import by setuptools.command.setopt)
W: no module named numpy.testing.Tester (top-level import by numpy.polynomial)
W: no module named numpy.compat.formatargspec (top-level import by numpy.ma.core)
W: no module named glib.Source (top-level import by gobject)
W: no module named numpy.testing.Tester (top-level import by numpy.random)
W: no module named msvcrt (conditional import by getpass)
W: no module named java (conditional import by xml.sax._exceptions)
W: no module named numpy.core.ravel (top-level import by numpy.linalg.linalg)
W: no module named numpy.core.atleast_3d (top-level import by numpy.lib.shape_base)
W: no module named matplotlib.MatplotlibDeprecationWarning (top-level import by matplotlib.legend)
W: no module named numpy.core.zeros (delayed import by numpy.testing.utils)
W: no module named numpy.compat.bytes (top-level import by numpy.core.numerictypes)
W: no module named setuptools.Command (top-level import by setuptools.command.easy_install)
W: no module named numpy.core.flatnonzero (top-level import by numpy.linalg.linalg)
W: no module named numpy.tanh (top-level import by numpy.numarray.ufuncs)
W: no module named PyQt4._qt (top-level import by PyQt4.QtCore)
W: no module named numpy.sinh (top-level import by numpy.numarray.ufuncs)
W: no module named glib.SPAWN_LEAVE_DESCRIPTORS_OPEN (top-level import by gobject)
W: no module named numpy.core.isscalar (delayed import by numpy.testing.utils)
W: no module named numpy.left_shift (top-level import by numpy.oldnumeric.ufuncs)
W: no module named Pyrex (delayed, conditional import by numpy.distutils.command.build_src)
W: no module named matplotlib.backends.pylab_setup (top-level import by matplotlib.pyplot)
W: no module named numpy.subtract (top-level import by numpy.numarray.ufuncs)
W: no module named matplotlib.rcParams (top-level import by matplotlib.ticker)
W: no module named numpy.arcsin (top-level import by numpy.numarray.ufuncs)
W: no module named numpy.lexsort (top-level import by numpy.numarray.functions)
W: no module named numpy.core.intc (top-level import by numpy.linalg.linalg)
W: no module named numpy.ndarray (top-level import by numpy.ma.extras)
W: no module named numpy.greater_equal (top-level import by numpy.oldnumeric.ufuncs)
W: no module named numpy.logical_not (top-level import by numpy.oldnumeric.ufuncs)
W: no module named Carbon (delayed import by plistlib)
W: no module named matplotlib.rcParams (top-level import by matplotlib.axis)
W: no module named _winreg (top-level import by distutils.msvccompiler)
W: no module named numpy.swapaxes (top-level import by numpy.oldnumeric.misc)
W: no module named numpy.diagonal (top-level import by numpy.numarray.functions)
W: no module named matplotlib.delaunay.__version__ (delayed, conditional import by matplotlib.mlab)
W: no module named numpy.arccos (top-level import by numpy.numarray.ufuncs)
W: no module named numpy.core.hstack (top-level import by numpy.lib.polynomial)
W: no module named numpy.sinh (top-level import by numpy.oldnumeric.ufuncs)
W: no module named numpy.core.max (conditional import by numpy)
W: no module named numpy.distutils.fcompiler.show_fcompilers (delayed import by numpy.distutils.interactive)
W: no module named numpy.core.concatenate (top-level import by numpy.fft.helper)
W: no module named numpy.arcsinh (top-level import by numpy.numarray.ufuncs)
W: no module named numpy.linalg.lstsq (top-level import by numpy.lib.polynomial)
W: no module named numpy.greater (top-level import by numpy.oldnumeric.ufuncs)
W: no module named numpy.core.atleast_2d (top-level import by numpy.lib.function_base)
W: no module named vms_lib (delayed, conditional import by platform)
W: no module named glib.set_application_name (top-level import by gobject)
W: no module named numpy.floor_divide (top-level import by numpy.numarray.ufuncs)
W: no module named numpy.testing.Tester (conditional import by numpy.distutils)
W: no module named numpy.less (top-level import by numpy.oldnumeric.ufuncs)
W: no module named matplotlib.rcParamsDefault (top-level import by matplotlib.pyplot)
W: no module named numpy_distutils (delayed, conditional import by numpy.f2py.diagnose)
W: no module named glib.MainContext (top-level import by gobject)
W: no module named numpy.ndarray (top-level import by numpy.ma.mrecords)
W: no module named matplotlib.rcParams (top-level import by matplotlib.text)
W: no module named numpy.eye (delayed import by numpy.core.numeric)
W: no module named numpy.searchsorted (top-level import by numpy.numarray.functions)
W: no module named numpy.core.

Mac mini，至于虚拟机，还是玩玩吧，不能干活

第2种，我是业余选手

@binux 你的意思是说，所有定义的self.XXX是函数内的（self）决定的，突然明白了。哈哈，被打醒了，ThankYou，谢谢你的耐心～～

@binux
class AAA():
self.a='5'
def go(self):
pass


b=AAA()
b.a

---------
为什么不可以在class内和def平行关系，定义self.XXX这种实例化变量？

@binux 其实我就想问一句，class下的def XXX里面的定义的变量，加self和不加有什么区别，它是不是和其他的一样？

因为我现在认为__init__()下是一个特殊的区域，在这里定义self.xxx是正常可以引用的实例化变量
但是在class下和def并列的时候，定义 self.XX就不会引用成功
同样，我怀疑在def内，变量加和不加self都不同。

我有点混乱，求指点

@binux 哈哈，文章还是有瑕疵。除了这个地方没说。你知道么，可爱的网友？

为毛不可以修改问题。问题是这样子的：
具体的如下：
class AAA(object):
def Go(self):
self.one = 'hello'

和下面的
class AAA(object):
def Go(self):
One= 'hello'

两者的one，有什么区别？

这个也有帮助http://www.crifan.com/summary_python_variable_effective_scope/

这个有点帮助http://www.crifan.com/summary_the_meaning_of_self_and___init___in_python_and_why_need_them/

@dylanninin 感谢，受用了

感谢他对Python社区做的贡献

第一次听说快手，请问快手可以跨平台么?很感兴趣，想了解

@xavierskip 我在OS X上打包的程序，运行没有反应，可用Py2app却可以，请问为什么？