numpy 学习5

The flat property gives back a numpy.flatiter object.This is the only means to get a flatiter object;we do not have access to a flatiter constructor.The flat iterator enables us to loop through an array as if it were a flat array,as shown in the following code object:
	In [80]: b
	Out[80]: 
	array([[0, 1],
	       [2, 3]])

	In [81]: f = b.flat

	In [82]: f
	Out[82]: <numpy.flatiter at 0x23ab790>

	In [83]: for item in f: print(item)
	0
	1
	2
	3

	In [84]: for item in f: print(item)	#暂时还不了解这个为什么没有输出

	In [85]: f = b.flat

	In [86]: f
	Out[86]: <numpy.flatiter at 0x23afac0>

	In [87]: for item in f:
	    ...:     print(item)
	    ...:     
	0
	1
	2
	3

We can convert a NumPy array to a Python list with the tolist() function:
    In [102]: b
    Out[102]: array([ 1.+1.j,  3.+2.j])

    In [103]: b.astype(int)
    /home/maoxianxin/virtualenv/python3/dataAnalysis/bin/ipython:1: ComplexWarning: Casting complex values to real discards the imaginary part
      #!/home/maoxianxin/virtualenv/python3/dataAnalysis/bin/python3
    Out[103]: array([1, 3])

    In [104]: b.astype(complex)
    Out[104]: array([ 1.+1.j,  3.+2.j])

    In [105]: b.astype('complex')
    Out[105]: array([ 1.+1.j,  3.+2.j])

    In [106]: b.astype('int')
    /home/maoxianxin/virtualenv/python3/dataAnalysis/bin/ipython:1: ComplexWarning: Casting complex values to real discards the imaginary part
      #!/home/maoxianxin/virtualenv/python3/dataAnalysis/bin/python3
    Out[106]: array([1, 3])

We are dropping off the imaginary part when casting from the complex type to int.The astype() function takes the name of a data type as a string too.