mirror of
https://github.com/data61/MP-SPDZ.git
synced 2026-04-20 03:01:31 -04:00
50 lines
1.3 KiB
Python
50 lines
1.3 KiB
Python
# this tests the pretrained ResNet in secure computation
|
|
|
|
program.options_from_args()
|
|
|
|
from Compiler import ml
|
|
|
|
try:
|
|
ml.set_n_threads(int(program.args[2]))
|
|
except:
|
|
pass
|
|
|
|
import torchvision
|
|
import torch
|
|
import numpy
|
|
import requests
|
|
import io
|
|
import PIL
|
|
|
|
from torchvision import transforms
|
|
|
|
model = getattr(torchvision.models.resnet, 'resnet' + program.args[1])(
|
|
weights='DEFAULT')
|
|
|
|
r = requests.get('https://github.com/pytorch/hub/raw/master/images/dog.jpg')
|
|
input_image = PIL.Image.open(io.BytesIO(r.content))
|
|
preprocess = transforms.Compose([
|
|
transforms.Resize(256),
|
|
transforms.CenterCrop(224),
|
|
transforms.ToTensor(),
|
|
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
|
|
])
|
|
input_tensor = preprocess(input_image)
|
|
input_batch = input_tensor.unsqueeze(0) # create a mini-batch as expected by the model
|
|
|
|
with torch.no_grad():
|
|
output = int(model(input_batch).argmax())
|
|
print('Model says %d' % output)
|
|
|
|
secret_input = sfix.input_tensor_via(
|
|
0, numpy.moveaxis(input_batch.numpy(), 1, -1))
|
|
|
|
layers = ml.layers_from_torch(model, secret_input.shape, 1, input_via=0)
|
|
|
|
optimizer = ml.Optimizer(layers, time_layers='time_layers' in program.args)
|
|
|
|
start_timer(1)
|
|
print_ln('Secure computation says %s',
|
|
optimizer.eval(secret_input, top=True)[0].reveal())
|
|
stop_timer(1)
|