Paillier Homomorphic Encryption Example

DISCLAIMER: This is a proof-of-concept implementation. It does not represent a remotely product ready implementation or follow proper conventions for security, convenience, or scalability. It is part of a broader proof-of-concept demonstrating the vision of the OpenMined project, its major moving parts, and how they might work together.



In [1]:

    
from syft.he.paillier import KeyPair, PaillierTensor
from syft import TensorBase
import numpy as np

Basic Ops



In [2]:

    
pubkey,prikey = KeyPair().generate()



In [3]:

    
x = PaillierTensor(pubkey, np.array([1, 2, 3, 4, 5.]))



In [4]:

    
x.decrypt(prikey)









    Out[4]:





BaseTensor: array([ 1.,  2.,  3.,  4.,  5.])



In [5]:

    
(x+x[0]).decrypt(prikey)









    Out[5]:





BaseTensor: array([ 2.,  3.,  4.,  5.,  6.])



In [6]:

    
(x*5).decrypt(prikey)









    Out[6]:





BaseTensor: array([  5.,  10.,  15.,  20.,  25.])



In [7]:

    
(x+x/5).decrypt(prikey)









    Out[7]:





BaseTensor: array([ 1.2,  2.4,  3.6,  4.8,  6. ])

Key SerDe



In [8]:

    
pubkey,prikey = KeyPair().generate()



In [9]:

    
x = PaillierTensor(pubkey, np.array([1, 2, 3, 4, 5.]))



In [10]:

    
pubkey_str = pubkey.serialize()
prikey_str = prikey.serialize()



In [11]:

    
pubkey2,prikey2 = KeyPair().deserialize(pubkey_str,prikey_str)



In [12]:

    
prikey2.decrypt(x)









    Out[12]:





BaseTensor: array([ 1.,  2.,  3.,  4.,  5.])



In [13]:

    
y = PaillierTensor(pubkey,(np.ones(5))/2)



In [14]:

    
prikey.decrypt(y)









    Out[14]:





BaseTensor: array([ 0.5,  0.5,  0.5,  0.5,  0.5])

Value SerDe



In [15]:

    
import pickle



In [16]:

    
y_str = pickle.dumps(y)



In [17]:

    
y2 = pickle.loads(y_str)



In [18]:

    
prikey.decrypt(y2)









    Out[18]:





BaseTensor: array([ 0.5,  0.5,  0.5,  0.5,  0.5])



In [ ]: