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

After updating two out of three fundamental parts, our current
state of development is:
• Data Preparation V0
• Model Configuration V1
• Model Training V1
How do we check if our changes introduced any bugs? We can inspect our model’s
state_dict():
# Checks model's parameters
print(model.state_dict())
Output
OrderedDict([('0.weight', tensor([[1.9690]], device='cuda:0')),
('0.bias', tensor([1.0235], device='cuda:0'))])
Let’s give our training loop a rest and focus on our data for a while. So far, we’ve
simply used our Numpy arrays turned into PyTorch tensors. But we can do better;
we can build a…
Dataset
In PyTorch, a dataset is represented by a regular Python class that inherits from
the Dataset class. You can think of it as a list of tuples, each tuple corresponding to
one point (features, label).
The most fundamental methods it needs to implement are:
• __init__(self): This takes whatever arguments are needed to build a list of
tuples—it may be the name of a CSV file that will be loaded and processed; it
may be two tensors, one for features, another one for labels; or anything else,
depending on the task at hand.
134 | Chapter 2: Rethinking the Training Loop