decrypt_to_pytorch_tensor

Engine.decrypt_to_pytorch_tensor(ciphertext, secret_key)

Decrypts a ciphertext and returns the result as a PyTorch tensor. If the engine's mode is set to GPU, the resulting tensor will also be allocated on the GPU.

• Input:
  • Ciphertext
  • SecretKey
• Output:
  • Plaintext

Examples

CPU

import torch

from desilofhe import Engine

engine = Engine()
secret_key = engine.create_secret_key()

tensor = torch.tensor([1, 2, 3])
encrypted = engine.encrypt_pytorch_tensor(tensor, secret_key)
decrypted = engine.decrypt_to_pytorch_tensor(encrypted, secret_key)

GPU

import torch

from desilofhe import Engine

engine = Engine(mode="gpu")
secret_key = engine.create_secret_key()

tensor = torch.tensor([1, 2, 3], device="cuda")
encrypted = engine.encrypt_pytorch_tensor(tensor, secret_key)
decrypted = engine.decrypt_to_pytorch_tensor(encrypted, secret_key)