Probing the limits of fake-quantised neural networks
From iis-projects
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.
Short Description
Abstract of the project
Status: Available
- Looking for 1 Semester/Master students
- Contact: Matteo Spallanzani, Renzo Andri (Huawei RC Zurich)
Prerequisites
- Experience in C/C++ and Python programming
- Knowledge of numerical analysis
- (Recommended) Experience in the PyTorch deep learning framework
- (Recommended) Knowledge of numerical representation formats and digital arithmetic
Character
- 20% Theory
- 40% C/C++ and Python coding
- 40% Deep learning