norm归一化处理——按照行进行？？？

hi,又又遇到了这个问题，源于搞这个博文，作者按照行进行归一化，这是我的疑问

每列是个特征，为啥按照行进行归一化呢？想不通啊。如下举例说明。

For Video Recommendation in Deep learning QQ Group 277356808
 
For Visual in deep learning QQ Group 629530787
 
I'm here waiting for you
 
wanna have a date with someone in Beijing please join in the QQ Group 737813700
 
 

 如下：我之前整的东西不知道放哪了，woc

>>> xx=np.random.randn(3,4)
>>> xx
array([[ 0.18874834,  0.37971162,  0.8287003 , -0.95896989],
       [-0.07977954,  0.04206023, -0.23647192, -0.36731412],
       [ 1.77722951,  0.68746666, -1.77812892,  0.54136854]])
>>> np.linalg.norm(xx, axis=1, keepdims=True)
array([[1.33647832],
       [0.4460633 ],
       [2.66194994]])
>>> xx/np.linalg.norm(xx, axis=1, keepdims=Tru