def features(self, x: torch.Tensor) -> torch.Tensor:
"""
Returns the non-activated output of the second-last layer.
:param x: input tensor (batch_size, input_size)
:return: output tensor (100)
"""
x = x.view(-1, num_flat_features(x))
return self._features(x)
def forward(self, x: torch.Tensor) -> torch.Tensor:
"""
Compute a forward pass.
:param x: input tensor (batch_size, input_size)
:return: output tensor (output_size)
"""
x = x.view(-1, num_flat_features(x))
return self.net(x)
def forward(self, x: torch.Tensor) -> torch.Tensor:
"""
Compute a forward pass.
:param x: input tensor (batch_size, input_size)
:return: output tensor (output_size)
"""
x = x.view(-1, num_flat_features(x))
if len(self.old_cols) > 0:
with torch.no_grad():
fc1_kb = [old(x) for old in self.old_fc1s]
fc2_kb = [old(fc1_kb[i]) for i, old in enumerate(self.old_fc2s)]
x = F.relu(self.fc1(x))
y = self.adaptor1(torch.cat(fc1_kb, 1))
x = F.relu(self.fc2(x) + y)
y = self.adaptor2(torch.cat(fc2_kb, 1))
x = self.classifier(x) + y
else:
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.classifier(x)
return x