#vgg16

Transfer learning with VGG-16 for 5 flower classes:

vgg16 = models.vgg16(pretrained=True)
for param in vgg16.features.parameters():
    param.requires_grad = False
vgg16.classifier[6] = nn.Linear(4096, 5)
optimizer = optim.SGD(vgg16.classifier.parameters(), lr=0.001)

vgg16.classifier holds three fully connected layers, the last of which was just replaced. Which parameters change during training?