Intro to Deep Learning with PyTorch

Working in Python with the PyTorch library, participants will learn how to prepare real neuroscience datasets for deep learning, build and train neural network models, and enhance model performance through methods such as data augmentation and regularization. The course will explore the defining characteristics of feed-forward, recurrent, and convolutional neural networks, and teach you how to choose the right architecture for different kinds of data. Finally, participants will see how object-oriented programming can be used to make scalable and efficient training pipelines. By the end of the workshop, attendees will have the knowledge and experience to design, implement, and apply neural network models to real-world neuroscience problems.