PyTorch is one of the most popular Machine Learning frameworks and is commonly used for both research and production. However, in practice, users are often unaware of how to profile and tune their model and miss out on important performance wins. This talk covers the guidelines and best practices for profiling and tuning the models.
Agenda: The talk will describe guidelines and best practices for profiling and tuning PyTorch models.
Target Audience: People using/developing PyTorch models.
Why is this talk important: Users are often unaware of how to profile and tune their model and miss out on important performance wins.
Format: Talk using slides. No hands-on coding. Slides will have code snippets and screenshots.
About 15 min: PyTorch Profiling 101
2-3 min: Code snippets/screenshots for showing how to setup PyTorch profiler.
7-10 min: Walking through output from profiling a PyTorch model on GPU (prepared beforehand).
About 10 min: Tuning PyTorch Models
About 25 min for the talk and 5 min for Q/A in the end