Questions

1. Why should we convert integer inputs into float values during training?
2. What are the methods used to reshape a tensor object?
3. Why is computation faster with tensor objects than with NumPy arrays?
4. What constitutes the init magic function in a neural network class?
5. Why do we perform zero gradients before performing backpropagation?
6. What magic functions constitute the dataset class?
7. How do we make predictions on new data points?
8. How do we fetch the intermediate layer values of a neural network?
9. How does the Sequential method help simplify the definition of the architecture of a neural network?
10. While updating loss_history, we append loss.item() instead of loss. What does this accomplish, and why is it useful to append loss.item() instead of just loss?
11. What are the advantages of using torch.save(model.state_dict())?

Learn more on Discord

Join our community’s Discord space for discussions...