A Practical Introduction to the Machine Learning Framework Pytorch
Overview
Pytorch is an open source framework for Deep Learning, known for being flexible and easy to use.
It is based on python and integrated with popular libraries such as NumPy and SciPy. Moreover, PyTorch features
excellent documentation and tutorials, native support for CPU and GPU, and dynamic graph calculation.
This course will cover the basics of PyTorch and include examples of their application in a high energy physics context.
- The course will take place in CIP Pool INF 226 1.305
- Date and time: 7-11 April, 9:00-12:00
Contents
- PyTorch Fundamentals: Tensors, Datasets and Models
- Automatic Differentiation and Optimization
- Geometric Learning: Graph Neural Networks and their Implementation in PyTorch
- Scientific Application: Track Reconstruction in Particle Physics
Refences
PyTorch webpageRequirements
Good Python skills and some Machine Learning basics.
