def __init__(self):
super(spiral_model, self).__init__()
self.fc1 = nn.linear(2, 16)
self.fc2 = nn.linear(16, 16)
self.fc3 = nn.linear(16, 2)
self.tanh1 = nn.tanh()
self.tanh2 = nn.tanh()
self.sig = nn.relu()
def __init__(self):
super(cnn_mnist_model, self).__init__()
self.conv1 = nn.conv2d(1, 32, 3, padding=1)
self.pool = nn.maxpool2d(2, 2)
self.conv2 = nn.conv2d(32, 48, 3)
self.fc1 = nn.linear(48 * 2 * 2, 120) # (599, 192)
self.fc2 = nn.linear(120, 84)
self.fc3 = nn.linear(84, 10)
self.relu1 = nn.relu()
self.relu2 = nn.relu()
self.relu3 = nn.relu()
self.relu4 = nn.relu()
def test_linear():
a = np.random.ranf([3, 5]).astype(np.float32)
t1 = madml.tensor(a)
module = nn.linear(5, 5, use_gpu=True)
t2 = module._forward_cpu(t1)
y = t2.host_data
t3 = module._forward_gpu(t1)
y_hat = t3.download()
t1.gradient.host_data = a
print(y_hat == y)
input()
def test_linear(self):
import madml
import madml.nn as nn
a = np.random.ranf([3, 5]).astype(np.float32)
t1 = madml.tensor(a)
self.assertTrue((t1.host_data == a).all())
module = nn.linear(5, 5)
t2 = module.forward(t1)
y = t2.host_data
module.to(0)
t3 = module.forward(t1)
y_hat = t3.download()
self.assertTrue((y == y_hat).all())
t2.gradient.host_data = a
t2.gradient.upload()
dx = module.backward()
dx_hat = dx.download()
print(dx_hat)
self.assertTrue((dx_hat != 0.0).all())
def __init__(self):
super(identity_model, self).__init__()
self.fc1 = nn.linear(32, 32)
self.fc2 = nn.linear(32, 32)
def __init__(self):
super(dnn_mnist_model, self).__init__()
self.fc1 = nn.linear(8 * 8, 256)
self.fc2 = nn.linear(256, 10)
self.relu1 = nn.relu()
self.relu2 = nn.relu()
