Deep Learning With PyTorch

Challenge yourself with CNNs,RNNs,and transformers.

1. What is the fundamental data structure in PyTorch used for numerical computations?
2. PyTorch's autograd package automatically computes gradients for tensor operations.
3. Which of the following are valid activation functions in PyTorch?
4. What does the abbreviation 'CNN' stand for in deep learning?
5. Which PyTorch class is used as the base for defining custom neural networks?
6. In PyTorch, the 'backward()' method is called on a loss tensor to compute gradients.
7. Which of the following are optimizers available in the torch.optim module?
8. What PyTorch function initializes a tensor filled with zeros? (format: torch.zeros)
9. Which method is used to move a PyTorch tensor to a GPU?
10. PyTorch's nn.CrossEntropyLoss combines the LogSoftmax activation and NLLLoss (Negative Log Likelihood Loss).
11. Which of the following are layer classes in PyTorch's nn module?
12. What is the full form of 'ReLU'?
13. By default, what is the data type of a tensor created with torch.tensor([1, 2, 3])?
14. PyTorch requires manual specification of gradients for all tensor operations.
15. Which loss functions are suitable for regression tasks in PyTorch?
16. Name the PyTorch built-in dataset containing handwritten digit images (abbrev.)
17. Which parameter in a PyTorch tensor enables gradient tracking?
18. nn.Sequential is a PyTorch container for stacking modules in sequence, where each module's output is the input to the next.
19. Which of the following are in-place operations in PyTorch (modify the tensor itself)?
20. What PyTorch subpackage provides utilities for loading and preprocessing datasets? (format: torch.utils.data)
Answered 0 of 0 — 0 correct