Intro to PyTorch Flashcards
(20 cards)
What is Deep Learning?
A type of machine learning that uses neural networks to learn patterns from large datasets.
How does Deep Learning differ from traditional programming?
In DL, the model learns rules from data and outputs, instead of being given rules.
When should you use Deep Learning?
When rules are complex, environment is dynamic, and lots of data is available.
Name some applications of Deep Learning.
Image recognition, speech-to-text, translation, sentiment analysis, recommendation systems.
What are the three main learning paradigms?
Supervised learning, unsupervised learning, and transfer learning.
What is a Neural Network?
A layered structure of nodes that learns to map input data to outputs using learned weights.
What is PyTorch?
An open-source deep learning framework developed by Facebook used for building and training neural networks.
Why is PyTorch popular?
Pythonic, GPU-compatible, fast, research-friendly, and widely used in industry.
What is a GPU?
Graphics Processing Unit – it performs fast numerical computations, especially matrix operations.
What is a Tensor?
A multi-dimensional array used as the basic data structure in PyTorch.
What are the dimensions of a tensor used for images?
Typically 3D or 4D: [channels, height, width] or [batch, channels, height, width].
How is a tensor different from a NumPy array?
Tensors can run on GPUs and are used for auto-differentiation in PyTorch.
What are the key steps in a PyTorch workflow?
Create tensors, preprocess data, build model, train, predict, evaluate, save/load.
What Python library do you import to use PyTorch?
import torch
Which week covers Supervised Learning in PyTorch?
Week 10
Which week includes your Test 2?
Week 12
Where can you run PyTorch online?
On Google Colab: colab.google.com
What function is used to create a tensor in PyTorch?
torch.tensor([…])
What makes PyTorch suitable for research?
It’s dynamic, easy to debug, and supports fast prototyping.
