def test_wtac_one_hot(self):
d = torch.tensor([[1.99, 3.01], [3.0, 2.01]])
labels = torch.tensor([[0, 1], [1, 0]])
actual = competitions.wtac(d, labels)
desired = torch.tensor([[0, 1], [1, 0]])
mismatch = np.testing.assert_array_almost_equal(actual,
desired,
decimal=5)
self.assertIsNone(mismatch)
def test_wtac_unequal_dist(self):
d = torch.tensor([[2.0, 3.0, 4.0], [2.0, 3.0, 1.0]])
labels = torch.tensor([0, 1, 1])
actual = competitions.wtac(d, labels)
desired = torch.tensor([0, 1])
mismatch = np.testing.assert_array_almost_equal(actual,
desired,
decimal=5)
self.assertIsNone(mismatch)
def test_wtac(self):
d = torch.tensor([[2., 3., 1.99, 3.01], [2., 3., 2.01, 3.]])
labels = torch.tensor([0, 1, 2, 3])
actual = competitions.wtac(d, labels)
desired = torch.tensor([2, 0])
mismatch = np.testing.assert_array_almost_equal(actual,
desired,
decimal=5)
self.assertIsNone(mismatch)
# Optimize using SGD optimizer from `torch.optim`
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)
criterion = GLVQLoss(squashing="sigmoid_beta", beta=10)
x_in = torch.Tensor(x_train)
y_in = torch.Tensor(y_train)
# Training loop
title = "Prototype Visualization"
fig = plt.figure(title)
for epoch in range(70):
# Compute loss
dis, plabels = model(x_in)
loss = criterion([dis, plabels], y_in)
with torch.no_grad():
pred = wtac(dis, plabels)
correct = pred.eq(y_in.view_as(pred)).sum().item()
acc = 100. * correct / len(x_train)
print(
f"Epoch: {epoch + 1:03d} Loss: {loss.item():05.02f} Acc: {acc:05.02f}%"
)
# Take a gradient descent step
optimizer.zero_grad()
loss.backward()
optimizer.step()
# Get the prototypes form the model
protos = model.proto_layer.prototypes.data.numpy()
if np.isnan(np.sum(protos)):
print("Stopping training because of `nan` in prototypes.")