Introduction to Neural Networks and PyTorch

In this module, you'll learn how to structure and prepare data for training in PyTorch. You'll create custom dataset classes, implement __len__ and __getitem__, and apply preprocessing steps using transforms and Compose. You'll also work with image datasets and Torchvision patterns. By the end, you'll understand how data flows into a PyTorch model during training.

In this module, you'll learn how to build and train linear regression models in PyTorch. You'll explore how models are defined using nn.Module, how state_dict() stores parameters, and how loss functions measure prediction error. You'll examine cost surfaces, gradient descent, learning rates, and stopping criteria. Through hands-on training loops, you'll see how slope and bias update over time as the model minimizes loss.

In this module, you'll discover how to implement training workflows using PyTorch tools such as DataLoader and optimizers. You'll learn how to compare batch, stochastic, and mini-batch gradient descent, and examine how batch size, epochs, and learning rate affect convergence. You'll learn how to structure full training loops with forward passes, backpropagation, and parameter updates. Finally, you'll explore training, validation, and test splits to evaluate model performance and detect overfitting.

In this module, you'll explore how to extend linear regression to handle multiple input features and multiple outputs. You'll learn how to use nn.Linear and custom modules to build higher-dimensional models and discover how weights and bias expand from scalars to vectors and matrices. You'll practice working with vectorized cost functions, gradient descent, and training workflows using DataLoaders and optimizers. Through hands-on labs, you'll learn how to build, train, and evaluate multi-dimensional and multi-output regression models step by step using real PyTorch code patterns.

In this module, you'll explore how to move from regression to classification. You'll learn how to build logistic regression models using nn.Sequential, apply the sigmoid function to generate probabilities, and convert probabilities into class predictions. You'll examine the Bernoulli distribution and maximum likelihood estimation and discover why cross-entropy loss is preferred over Mean Squared Error (MSE) for classification tasks. You'll also explore optimization and regularization techniques that help improve classification performance.

In this module, you'll apply what you've explored throughout the course in a hands-on classification project. You will build a logistic regression model to predict the outcomes of League of Legends matches. Leveraging various in-game statistics, this project will utilize your knowledge of PyTorch, logistic regression, and data handling to create a robust predictive model. Finally, you can choose between immediate auto-grading using the IBM AI-assisted assessment tool, Mark, or submit your assignment for a human peer review.