super(TestPRelu, self).__init__()
self.prelu = nn.PReLU(init=random.random(), num_parameters=inp)
def forward(self, x):
x = self.prelu(x)
return x
if __name__ == '__main__':
max_error = 0
for i in range(100):
inp = np.random.randint(1, 100)
out = np.random.randint(1, 100)
model = TestPRelu(inp, out, inp % 2)
input_np = np.random.uniform(-10, 10, (1, inp))
input_var = Variable(torch.FloatTensor(input_np))
output = model(input_var)
k_model = pytorch_to_keras((inp, ), output)
pytorch_output = output.data.numpy()
keras_output = k_model.predict(input_np)
error = np.max(pytorch_output - keras_output)
print(error)
if max_error < error:
max_error = error
print('Max error: {0}'.format(max_error))
sys.path.append('../pytorch2keras')
from converter import pytorch_to_keras
if __name__ == '__main__':
max_error = 0
for i in range(10):
model = torchvision.models.resnet18()
for m in model.modules():
m.training = False
input_np = np.random.uniform(0, 1, (1, 3, 224, 224))
input_var = Variable(torch.FloatTensor(input_np))
output = model(input_var)
k_model = pytorch_to_keras((
3,
224,
224,
), output)
pytorch_output = output.data.numpy()
keras_output = k_model.predict(input_np)
error = np.max(pytorch_output - keras_output)
print(error)
if max_error < error:
max_error = error
print('Max error: {0}'.format(max_error))
def forward(self, x):
x = self.conv2d(x)
return x
if __name__ == '__main__':
max_error = 0
for i in range(100):
kernel_size = np.random.randint(1, 10)
inp = np.random.randint(kernel_size + 1, 100)
out = np.random.randint(1, 2)
model = TestConv2d(inp + 2, out, kernel_size, inp % 2)
input_np = np.random.uniform(0, 1, (1, inp + 2, inp, inp))
input_var = Variable(torch.FloatTensor(input_np))
output = model(input_var)
k_model = pytorch_to_keras((inp + 2, inp, inp,), output, change_ordering=True)
pytorch_output = output.data.numpy()
keras_output = k_model.predict(input_np.transpose(0, 2, 3, 1))
error = np.max(pytorch_output - keras_output.transpose(0, 3, 1, 2))
print(error)
if max_error < error:
max_error = error
print('Max error: {0}'.format(max_error))
self.sigmoid = nn.Sigmoid()
def forward(self, x):
x = self.linear(x)
x = self.sigmoid(x)
return x
if __name__ == '__main__':
max_error = 0
for i in range(100):
inp = np.random.randint(1, 100)
out = np.random.randint(1, 100)
model = TestSigmoid(inp, out, inp % 2)
input_np = np.random.uniform(-1.0, 1.0, (1, inp))
input_var = Variable(torch.FloatTensor(input_np))
output = model(input_var)
k_model = pytorch_to_keras(model, input_var, (inp, ), verbose=True)
pytorch_output = output.data.numpy()
keras_output = k_model.predict(input_np)
error = np.max(pytorch_output - keras_output)
print(error)
if max_error < error:
max_error = error
print('Max error: {0}'.format(max_error))
class TestEmbedding(nn.Module):
def __init__(self, input_size):
super(TestEmbedding, self).__init__()
self.embedd = nn.Embedding(input_size, 100)
def forward(self, input):
return self.embedd(input).sum(dim=0)
if __name__ == '__main__':
max_error = 0
for i in range(100):
input_np = np.random.randint(0, 10, (1, 1, 4))
input = Variable(torch.LongTensor(input_np))
simple_net = TestEmbedding(1000)
output = simple_net(input)
k_model = pytorch_to_keras(simple_net, input, (1, 4), verbose=True)
pytorch_output = output.data.numpy()
keras_output = k_model.predict(input_np)
error = np.max(pytorch_output - keras_output[0])
print(error)
if max_error < error:
max_error = error
print('Max error: {0}'.format(max_error))
if __name__ == '__main__':
max_error = 0
for i in range(10):
model = resnet18()
for m in model.modules():
m.training = False
input_np = np.random.uniform(0, 1, (1, 3, 224, 224))
input_var = Variable(torch.FloatTensor(input_np))
output = model(input_var)
k_model = pytorch_to_keras((
3,
224,
224,
),
output,
change_ordering=True)
pytorch_output = output.data.numpy()
keras_output = k_model.predict(input_np.transpose(0, 2, 3, 1))
print(pytorch_output.shape, keras_output.shape)
error = np.max(pytorch_output - keras_output)
print(error)
if max_error < error:
max_error = error
print('Max error: {0}'.format(max_error))
max_error = 0
for i in range(100):
kernel_size = np.random.randint(1, 10)
inp = np.random.randint(kernel_size + 1, 100)
out = np.random.randint(1, 2)
model = TestConv2d(inp + 2, out, kernel_size, inp % 2)
input_np = np.random.uniform(0, 1, (1, inp + 2, inp, inp))
input_var = Variable(torch.FloatTensor(input_np))
output = model(input_var)
k_model = pytorch_to_keras(model,
input_var, (
inp + 2,
inp,
inp,
),
change_ordering=True,
verbose=True)
pytorch_output = output.data.numpy()
keras_output = k_model.predict(input_np.transpose(0, 2, 3, 1))
error = np.max(pytorch_output - keras_output.transpose(0, 3, 1, 2))
print(error)
if max_error < error:
max_error = error
print('Max error: {0}'.format(max_error))
self.classif = nn.Linear(1536, num_classes)
def forward(self, x):
x = self.features(x)
x = x.view(x.size(0), -1)
x = self.classif(x)
return x
if __name__ == '__main__':
max_error = 0
for i in range(10):
model = InceptionV4()
model.eval()
input_np = np.random.uniform(0, 1, (4, 3, 299, 299))
input_var = Variable(torch.FloatTensor(input_np))
output = model(input_var)
k_model = pytorch_to_keras((3, 299, 299,), output)
pytorch_output = output.data.numpy()
keras_output = k_model.predict(input_np)
error = np.max(pytorch_output - keras_output)
print(error)
if max_error < error:
max_error = error
print('Max error: {0}'.format(max_error))
model = ResNet(BasicBlock, [2, 2, 2, 2], **kwargs)
if pretrained:
model.load_state_dict(model_zoo.load_url(model_urls['resnet18']))
return model
if __name__ == '__main__':
max_error = 0
for i in range(10):
model = resnet18()
for m in model.modules():
m.training = False
input_np = np.random.uniform(0, 1, (1, 3, 224, 224))
input_var = Variable(torch.FloatTensor(input_np))
output = model(input_var)
k_model = pytorch_to_keras(model, input_var, (3, 224, 224,), verbose=True, change_ordering=True)
pytorch_output = output.data.numpy()
keras_output = k_model.predict(input_np.transpose(0, 2, 3, 1))
print(pytorch_output.shape, keras_output.shape)
error = np.max(pytorch_output - keras_output)
print(error)
if max_error < error:
max_error = error
print('Max error: {0}'.format(max_error))
