PyTorch Module - Deep Learning Course
PyTorch module as part of the Deep Learning course in the Master's degree program in Data Science (LM-DATA) at the University of Catania. This module covers practical deep learning implementation using PyTorch, including neural network construction, training, and optimization.
Instructor: Salvatore Alfio Sambataro
Term: Winter
Location: University of Catania, Department of Mathematics and Computer Science
Overview
This module is an integral part of the Deep Learning course in the Master’s degree program in Data Science (Laurea Magistrale in Data Science) at the University of Catania. It provides hands-on experience with PyTorch, one of the most popular deep learning frameworks.
Topics Covered
- Introduction to PyTorch: Framework overview, tensor operations, and computational graphs
- Building Neural Networks: Layer types, activation functions, and model architecture design
- Training Deep Models: Loss functions, optimizers, and gradient descent strategies
- Advanced Architectures: Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Transformers
- Practical Implementation: Image classification, sequence modeling, and transfer learning
- Debugging and Optimization: Performance profiling, mixed precision training, and distributed computing
- Real-world Applications: Computer vision, natural language processing, and generative models