Imaging Lunch: Introduction to Pytorch for image analysis
Join us for a hands-on Introductory PyTorch Workshop, led by our in-house expert Florian Aymanns, designed for EPFL PhD students and postdocs interested in applying neural networks to image analysis. (Registration required.)
This session is best suited for participants with basic knowledge of Python who want to learn the foundations of neural networks and their implementation in PyTorch.
We’ll start by introducing tensors, highlighting their differences from NumPy arrays, and demonstrating autograd with simple examples. Moving forward, we’ll cover key neural network components like convolutional and fully connected layers, and use them to construct a neural network from scratch.
Participants will gain an understanding of loss functions, autograd, and stochastic gradient descent, learning how weights are updated during training. Finally, we’ll apply these concepts to train an image classification network using PyTorch.
This workshop offers a code-along experience on EPFL’s RCP cluster and is perfect for those eager to begin their journey into machine learning for image analysis.
Learning Objectives
- Understand the components of a convolutional neural network in code
- Train your first neural network to classify images using Pytorch
Prerequisites
- Basic python knowledge
- Familiarity with NumPy arrays
- Basic understanding of neural networks
- Understanding of object oriented programming (classes and inheritance) is a plus but not required
About the Imaging Lunches: Once per month, the EPFL Center for Imaging organises an event dedicated to all PhD students and postdocs working with/in imaging. Discuss the latest advances in imaging. Connect with imaging peers. Learn about popular imaging tools!
