Daniel Voigt Godoy - Deep Learning with PyTorch Step-by-Step A Beginner’s Guide-leanpub

Output
Linear(in_features=1, out_features=1, bias=True)
Do we still have our b and w parameters? Sure, we do:
linear.state_dict()
Output
OrderedDict([('weight', tensor([[-0.2191]])),
('bias', tensor([0.2018]))])
So, our former parameter b is the bias, and our former parameter w is the weight
(your values will be different since I haven’t set up a random seed for this example).
Now, let’s use PyTorch’s Linear model as an attribute of our own, thus creating a
nested model.
You are not limited to defining parameters, though; models can
contain other models as their attributes as well, so you can easily
nest them. We’ll see an example of this shortly.
Even though this clearly is a contrived example, since we are pretty much wrapping
the underlying model without adding anything useful (or, at all!) to it, it illustrates the
concept well.
Notebook Cell 1.11 - Building a model using PyTorch’s Linear model
class MyLinearRegression(nn.Module):
def __init__(self):
super().__init__()
# Instead of our custom parameters, we use a Linear model
# with a single input and a single output
self.linear = nn.Linear(1, 1)
def forward(self, x):
# Now it only makes a call
self.linear(x)
110 | Chapter 1: A Simple Regression Problem