Methods to convert matrices into lists and other functions in Python's numpy library_python

>>> x = np.matrix(np.arange(12).reshape((3,4))); x
matrix([[ 0, 1, 2, 3],
  [ 4, 5, 6, 7],
  [ 8, 9, 10, 11]])
>>> x.tolist()
[[0, 1, 2, 3], [4, 5, 6, 7], [8, 9, 10, 11]]

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

>>> a = np.array([[1, 2], [3, 4]]) #对所有元素求均值
>>> np.mean(a)
2.5
>>> np.mean(a, axis=0) #对每一列求均值
array([ 2., 3.])
>>> np.mean(a, axis=1) #对每一行求均值
array([ 1.5, 3.5])

>>> a = np.array([[1, 2], [3, 4]]) #对所有元素求标准差 
>>> np.std(a)
1.1180339887498949
>>> np.std(a, axis=0) #对每一列求标准差
array([ 1., 1.])
>>> np.std(a, axis=1) #对每一行求标准差
array([ 0.5, 0.5])

>>> a=np.array([[1,2,3],[4,5,6],[7,8,9]]) #先输入3行2列的数组a
>>> b=a[:,:2] 
>>> b.shape #当数组的行与列都大于1时，不需增加维度
(3, 2)
>>> c=a[:,2] 
>>> c.shape #可以看到，当数组只有一列时，缺少列的维度
(3,)
>>> c
array([3, 6, 9])

>>> d=a[:,2,np.newaxis] #np.newaxis实现增加列的维度
>>> d
array([[3],
  [6],
  [9]])
>>> d.shape  #d的维度成了3行1列（3,1）
(3, 1)
>>> e=a[:,2,None] #None与np.newaxis实现相同的功能
>>> e
array([[3],
  [6],
  [9]])
>>> e.shape
(3, 1)

>>> index = [i for i in range(10)] 
>>> index 
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9] 
>>> np.random.shuffle(index) 
>>> index 
[7, 9, 3, 0, 4, 1, 5, 2, 8, 6]

>>> import numpy as np 
>>> a = np.arange(15).reshape(5,3) #构造一个5行3列的二维数组 
>>> a 
array([[ 0, 1, 2], 
  [ 3, 4, 5], 
  [ 6, 7, 8], 
  [ 9, 10, 11], 
  [12, 13, 14]]) 
>>> b = a[:,0].min() ##取第0列的最小值，其他列同理 
>>> b 
0 
>>> c = a[0,:].max() ##取第0行的最大值，其他行同理 
>>> c 
2

n = np.array(np.random.randn(4,2)) 
 
n 
Out[153]: 
array([[ 0.17234 , -0.01480043], 
  [-0.33356669, -1.33565616], 
  [-1.11680009, 0.64230761], 
  [-0.51233174, -0.10359941]]) 
 
l = np.array([1,2,3,4]) 
 
l 
Out[155]: array([1, 2, 3, 4]) 
 
l.shape 
Out[156]: (4,)

n = np.hstack((n,l)) 
ValueError: all the input arrays must have same number of dimensions

n = np.hstack((n,l[:,np.newaxis])) ##注意：在使用np.hstack()时必须用()把变量括起来，因为它只接受一个变量 
 
n 
Out[161]: 
array([[ 0.17234 , -0.01480043, 1.  ], 
  [-0.33356669, -1.33565616, 2.  ], 
  [-1.11680009, 0.64230761, 3.  ], 
  [-0.51233174, -0.10359941, 4.  ]])

n = np.array([[1,2,3,4,5,6],[11,22,33,44,55,66],[111,222,333,444,555,666]]) ##产生一个三行六列容易区分的数组 
 
n 
Out[166]: 
array([[ 1, 2, 3, 4, 5, 6], 
  [ 11, 22, 33, 44, 55, 66], 
  [111, 222, 333, 444, 555, 666]]) 
 
sample = [[]for i in range(3)] ##产生三行一列的空列表 
Out[172]: [[], [], []] 
for i in range(0,6,2): ##每间隔一列便添加到sample中 
 sample = np.hstack((sample,n[:,i,np.newaxis]))  
  
 
sample 
Out[170]: 
array([[ 1., 3., 5.], 
  [ 11., 33., 55.], 
  [ 111., 333., 555.]])