CUDA graph trees are the internal implementation of CUDA graphs used in PT2 when you say mode="reduce-overhead". Their primary innovation is that they allow the reuse of memory across multiple CUDA graphs, as long as they form a tree structure of potential paths you can go down with the CUDA graph. This greatly reduced the memory usage of CUDA graphs in PT2. There are some operational implications to using CUDA graphs which are described in the podcast.

Rss Apple Podcaster
Podden och tillhörande omslagsbild på den här sidan tillhör Edward Yang, Team PyTorch. Innehållet i podden är skapat av Edward Yang, Team PyTorch och inte av, eller tillsammans med, Poddtoppen.