Automatic activation checkpointing and tensor swapping for DNN training using torch.fx graph profiling.
-
Updated
Aug 30, 2026 - Jupyter Notebook
Automatic activation checkpointing and tensor swapping for DNN training using torch.fx graph profiling.
From-scratch graph optimizer for torch.compile: constant folding, CSE and DCE on PyTorch FX graphs.
To associate your repository with the torch-fx topic, visit your repo's landing page and select "manage topics."